review article broadband over power lines systems...

Hindawi Publishing CorporationISRN Power EngineeringVolume 2013 Article ID 517940 30 pageshttpdxdoiorg1011552013517940

Review ArticleBroadband over Power Lines Systems ConvergenceMultiple-Input Multiple-Output Communications Analysis ofOverhead and Underground Low-Voltage andMedium-Voltage BPL Networks

Athanasios G Lazaropoulos

School of Electrical and Computer Engineering National Technical University of Athens9 Iroon Polytechniou Street Zografou 15780 Athens Greece

Correspondence should be addressed to Athanasios G Lazaropoulos aglazaropoulosgmailcom

Received 26 April 2013 Accepted 6 June 2013

Academic Editors V Fernao Pires and W Jaikla

Copyright copy 2013 Athanasios G LazaropoulosThis is an open access article distributed under the Creative Commons AttributionLicense which permits unrestricted use distribution and reproduction in any medium provided the original work is properlycited

This review paper reveals the broadband potential of overhead and underground low-voltage (LV) and medium-voltage (MV)broadband over power lines (BPL) networks associated withmultiple-inputmultiple-output (MIMO) technologyThe contributionof this review paper is fourfold First the unified value decomposition (UVD)modal analysis is introducedUVDmodal analysis is anew technique that unifies eigenvalue decomposition (EVD) and singular value decomposition (SVD)modal analyses achieving thecommon handling of traditional SISOBPL and upcoming MIMOBPL systems The validity of UVD modal analysis is examinedby comparing its simulation results with those of other exact analytical models Second based on the proposed UVD modalanalysis theMIMOchannels of overhead and underground LV andMVBPL networks (distribution BPL networks) are investigatedwith regard to their inherent characteristics Towards that direction an extended collection of well-validated metrics from thecommunications literature such as channel attenuation average channel gain (ACG) root-mean-square delay spread (RMS-DS)coherence bandwidth (CB) cumulative capacity capacity complementary cumulative distribution function (CCDF) and capacitygain (GC) is first applied in overhead and underground MIMOLV and MIMOMV BPL channels and systems It is found thatthe results of the aforementioned metrics portfolio depend drastically on the frequency the power grid type (either overhead orunderground either LV orMV) theMIMO scheme configuration properties theMTL configuration the physical properties of thecables used the end-to-end distance and the number the electrical length and the terminations of the branches encountered alongthe end-to-end BPL signal propagationThird three interesting findings concerning the statistical properties of MIMO channels ofdistribution BPL networks are demonstrated namely (i) the ACG RMS-DS and cumulative capacity lognormal distributions(ii) the correlation between RMS-DS and ACG and (iii) the correlation between RMS-DS and CB By fitting the numericalresults unified regression distributions appropriate for MIMOBPL channels and systems are proposed These three fundamentalproperties can play significant role in the evaluation of recently proposed statistical channelmodels for various BPL systems Fourththe potential of transformation of overhead and underground LVBPL and MVBPL distribution grids to an alternative solutionto fiber-to-the-building (FTTB) technology is first revealed By examining the capacity characteristics of various MIMO schemeconfigurations and by comparing these capacity results against SISO ones a new promising urban backbone network seems to beborn in a smart grid (SG) environment

1 Introduction

The distribution power gridsmdashthat is overhead and un-derground low-voltage (LV) and medium-voltage (MV)

networksmdashcan become the key to delivering broadband last-mile access in remote andor underdeveloped areas andsimultaneously to the development of an advanced IP-basedpower system [1ndash6]This power grid upgrade can be achieved

2 ISRN Power Engineering

through the deployment of broadbandover power lines (BPL)networks [7ndash14]

Despite their great broadband promises distributionpower grids constitute a rather hostile medium for commu-nications signal transmission Attenuation multipath due tovarious reflections noise and electromagnetic interferenceare the main problematic issues [15ndash22] Since overhead andunderground LVBPL and MVBPL channels mix the nastybehavior of a power line with that of a communication chan-nel each of the aforementioned adverse factors significantlydegrades the overall performance of BPL networks [23ndash25]

Recently new interest arises due to developments regard-ing multiple-input multiple-output (MIMO) transmissionscheme configurations for BPLnetworks andBPL coexistencewith other broadband technologies [13 14 26ndash30] Actuallythe need of coexistence among overhead and undergroundLVBPL and MVBPL networks (intraoperability) will be anecessary prerequisite before investigating the cooperativecommunications between distribution BPL networks andother broadband technologies (interoperability) [28 31ndash38] Despite the urgent demand for systems intraoperabil-ityinteroperability todayrsquos distribution BPL network capac-ity performance seems significantly worse than that of othercompetitive broadband technologiesmdashwired such as fiberand DSL and wireless such as Wi-Fi and WiMAX Hencethere is a need of not only utilizing existing technical devel-opments but also inventing and exploiting novel ones suchas MIMO technology

Being prominent from the wireless world [38ndash46] andrecently applied in various BPL networks [13 14 33 47ndash56]the potential of integrating MIMO technology with distri-bution BPL networks is investigated Towards that directionMIMO capabilities of distribution BPL networks are com-pared with todayrsquos single-input single-output (SISO) oneseither in transmission or in spectral efficiency fields

For efficient MIMOBPL (either MIMOLVBPL orMIMOMVBPL) communications a thorough understand-ing of the MIMO signal transmission at high frequenciesalong the overhead and underground LV andMVpower linesis required To facilitate MIMO analysis the hybrid modelthat is usually employed to examine the behavior of BPLtransmission channels installed onmulticonductor transmis-sion line (MTL) structures is also used in this review paper[2 3 5 6 9 10 12ndash14] This hybrid method based on acombination of bottom-up [57ndash65] and top-down [8 15 1824 58 62 64 66ndash69] approaches receives as inputs accu-rately determined parameters such as the MIMO schemeimplementation the power grid type (either overhead orunderground either LV or MV) the MTL configuration thephysical properties of cables and the grid topology and deliv-ers as outputs accurate results in terms of channel attenuationActually exploiting the acquired MIMOBPL knowledge of[13 14] MIMO expansion of the hybrid method is firstpresented in order to (i) achieve a smooth expansion ofSISO to MIMO analysis and (ii) unify eigenvalue decom-position (EVD) and singular value decomposition (SVD)modal analyses under the aegis of the proposed unified valuedecomposition (UVD) modal analysis

Based on the outputs of UVDmodal analysis broadbandtransmission characteristics statistical performance metricsand capacity measures are investigated through numericalresults concerning simulated overhead and undergroundMIMOLVBPL and MIMOMVBPL networks [57 62 6467 70] The numerical results of UVD modal analysis arevalidated against already verified results of the accurate TM2method which are analytically presented in [13 14]

More specifically a set of important statistical perfor-mance metrics such as the average channel attenuation theaverage channel gain (ACG) the root-mean-square delayspread (RMS-DS) and the coherence bandwidth (CB) triesto elicit the commonand the different aspects of overhead andunderground MIMOLVBPL and MIMOMVBPL chan-nels In addition two fundamental properties of severalwireline channels (eg DSL coaxial and phone links) arefirst validated in overhead and undergroundMIMOLVBPLandMIMOMVBPL systems namely (i) ACG and RMS-DSare negatively correlated lognormal random variables beingin agreement with recently proposed statistical BPL channelmodels [71ndash79] A new approximationmdashUN1 approachmdashsuitable for the design of overhead and undergroundMIMOBPL channels is proposed and compared to existingdistributions and (ii) CB and RMS-DS correlation behaviorcan be described by hyperbolic functions By fitting thesimulation results a new approximationmdashUN2 approachmdashappropriate for MIMOBPL channels is proposed [78ndash82]Moreover lognormal distributions for the cumulative dis-tribution functions (CDFs) of ACG RMS-DS and cumu-lative capacitymdashsuitable for overhead and undergroundMIMOLVBPL andMIMOMVBPL channelsmdashare used soas to (a) be imposed by design in the upcoming BPL statisticalchannel modeling and (b) operate as evaluation measures ofthe statistical channel models proposed [71ndash73 76ndash79]

Apart from statistical performance metrics suitablecapacitymetrics such as the cumulative capacity the capacitycomplementary cumulative distribution functions (capacityCCDF) and the capacity gain (GC) of MIMOBPL systemsare applied It is unveiled thatMIMOBPL systems can definean alternative technological solution to the existing fiber-optic technology and especially to the fiber-to-the-building(FTTB) systems Depending on BPL system investmentbudget technoeconomic and socioeconomic characteristicsrequired system complexity power grid type power gridtopology IPSDM limits applied operation frequency bandand BPL intraoperabilityinteroperability requirements dif-ferent single- and multiport implementations may be appliedin improving todayrsquos system capacities By fully exploitingthe MIMO technology the increase of the capacity of bothoverhead and underground MIMOLVBPL and MIMOMVBPL systems is notable exceeding 8Gbps Using onlythe existing infrastructure and electromagnetic compatibilitylimits the imminent MIMOBPL technology implementa-tions promise better capacity results by a factor of up to365 in comparison with the best todayrsquos capacity results forgiven power grid type and power grid topology Hence theforthcoming MIMOBPL capacity boost indicates that theBPL technology should be examined as (i) an emerging

ISRN Power Engineering 3

urban backbone network and (ii) a promise towards BPLintraoperabilityinteroperability

The rest of this review paper is organized as followsIn Section 2 the overhead and underground LV and MVconfigurations adopted in this paper are presented Section 3introduces theMTL theory andmodal analysis alongwith thenecessary assumptions concerning BPL transmission Apartfrom the presentation of EVD and SVD modal analyses theUVD modal analysis is highlighted in order to realize thenecessary upgrade from SISOBPL to MIMOBPL consid-eration Section 4 details the statistical performance metricsand the capacity measures applied in this review paper forthe upcoming MIMOBPL analysis In Section 5 a seriesof numerical results and conclusions is provided aimingat marking out how the various features of the overheadand underground MIMOLV and MIMOMV channels andsystems influence BPL transmission statistical performancemetrics and capacitymeasures On the basis of the confirmedfundamental properties new regression approximations areproposed Moreover after providing the SISO capacity ofoverhead and underground LVBPL and MVBPL channelsand systems different MIMO scheme configurations aredemonstrated and compared regarding their suitability andperformance for various overhead and underground LVBPLand MVBPL systems A thorough MIMO scheme config-uration investigation reveals the great capacity potential ofoverhead and underground MIMOLVBPL and MIMOMVBPL distribution power networks Section 6 concludesthis review paper

2 The Physical BPL Layer

The overhead and underground LV and MV MTL config-urations as well as the ground properties used across thefollowing MIMOBPL system analysis are outlined in thissection [8 15 17 23 57 59ndash61 66 67 83ndash86]

21 Overhead LV and MV Power Distribution Networks Atypical case of overhead LV distribution line is depicted inFigure 1(a) Four parallel noninsulated conductors are sus-pended one above the other spaced by Δ LV and located atheights ℎLV above ground for the lowest conductorThe upperconductor is the neutral while the lower three conductorsare the three phases This three-phase four-conductor (119899 =4) overhead LV distribution line configuration is consideredin the present work consisting of ASTER conductors Theexact dimensions for the overhead LVMTL configuration aredetailed in [60 61 83ndash85 87]

Overhead MV distribution lines hang at typical heightsℎMV Typically three parallel noninsulated phase conductorsspaced by ΔMV are used above lossy ground This three-phase three-conductor (119899 = 3) overheadMVdistribution lineconfiguration is considered in the present work consisting ofACSR conductorsmdashsee Figure 1(b) The exact properties ofthe overhead MV MTL configuration are detailed in [15 1657 67 83 86 87]

22 Underground LV and MV Power Distribution NetworkThe underground LV distribution line that will be examined

in the simulations is the three-phase four-conductor (119899 =

4) core-type YJV XLPE underground LV distribution cableburied 1m inside the ground The layout of this cable isdepicted in Figure 1(c) The cable arrangement consists ofthe three-phase three-core-type conductors one core-typeneutral conductor and one shield conductor The shield isgrounded at both ends The exact dimensions of this under-ground LVMTL configuration are reported in [58 59 62 6383 84 87 88]

The underground MV distribution line that will beexamined is the three-phase three-conductor (119899 = 3) sector-type PILC 810 kV distribution-class cable buried 1m insidethe ground [15ndash18 89 90] The cable arrangement consistsof the three-phase three-sector-type conductors one shieldconductor and one armor conductormdashsee Figure 1(d) Theshield and the armor are grounded at both ends The under-ground MVMTL configuration specifications are detailed in[17 91ndash94]

As it concerns BPL signal propagation in undergroundMTL configurations due to the common practice of ground-ing at both ends the shield acts as a ground return pathand as a reference conductor Hence the inner conductorsare separated electrostatically andmagnetostatically from theremaining set On the basis of this separation the analysismay be focused only on the remaining five- or four-conductortransmission line (TL) of interest for underground LVmdashthatis the three phases the neutral conductor and the shieldconductormdashor MV configurationsmdashthat is the three phasesand the shield conductor respectively Repeatedly applied in[17 58 59 62 63 66 83 87 91ndash98] this is the commonprocedure either in theoretical analyses or in measurements

23 Ground Properties The properties of the ground areassumed common either in overhead or in underground BPLconfigurations In detail the conductivity of the ground isassumed 120590119892 = 5 mSm and its relative permittivity 120576rg = 13In accordancewith [15ndash17 57ndash59 62 63 67 95 99ndash102] theseground properties are suitable for the BPL signal propagationanalysis at high frequencies (from 1MHz to approximately100MHz) either in overhead or in underground BPL systems

3 Modal Analysis of Overhead andUnderground LVBPL and MVBPL Systems

In this section through a matrix approach the standardTL analysis is extended to the MTL case which involvesmore than two conductors First a brief introduction toMTLtheory is provided which is accompanied with the necessaryassumptions concerning BPL transmission via the entiredistribution power grid Second a detailed modal analysis ofoverhead and underground LVBPL and MVBPL systems ishighlighted in particular UVDmodal analysis is proposed inorder to (i) unify EVD and SVDmodal analyses (ii) describethe analysis convergence of SISOBPL and MIMOBPL con-figurations and (iii) suggest appropriatemethodology for theperformance and capacity study of MIMOBPL systems

31 MTL Theory of Overhead and Underground LVBPL andMVBPL Systems Signal transmission via overhead power


Δ LV

Δ LVΔ LV

ΔMV ΔMVrMVp

120583o

120576g 120583o

hMV

120590g

hLV

(a) (b)

rMVs

rMVc

rMVΔMV

120579MV

x

y

z

Armor

Shield

Conductor 1Conductor 2

Conductor 3

Armor insulation

Armor-shield insulationInner insulation

(d)

Neutral conductorΔ LV

rLVc

rLVp

rLVn

rLVs

rLV

(c)

120576o

Figure 1 Typical multiconductor structures [15 17 58 95] (a) Overhead LV (b) Overhead MV (c) Underground LV (d) Underground MV

lines has been analyzed in [17 58 59 62 63 95 99ndash102]whereas signal transmission via three-phase undergroundpower lines has been detailed in [17 58 59 62 63 95 99ndash102]These formulations have the advantage that contrary to otheravailable models for overhead and underground power lines[85 103ndash105] they are appropriate for broadband applicationsof overhead LVBPL and MVBPL systems

Already described in [15 17 31 56 57 60 61 67]compared to a two-conductor line supporting one forward-and one backward-traveling wave an MTL structure with119899 + 1 conductors parallel to the 119911 axis as depicted inFigures 1(a)ndash1(d) may support 119899 pairs of forward- andbackward-traveling waves with corresponding propagationconstants These waves may be described by a coupled set of

2n first-order partial differential equations relating the linevoltages 119881119894(119911 119905) 119894 = 1 119899 to the line currents 119868119894(119911 119905) 119894 =1 119899 Each pair of forward- and backward-traveling wavesis referred to as a mode and presents its own propagationconstant

According to TM1 methodmdashdetailed in [15 16 57 5865 106] and briefly presented in Section 32mdashthese modesmay be examined separately across the overall overhead orunderground LV or MV power distribution network underthe following two assumptions

(A1) The branching cables are assumed to be identical tothe distribution cables Hence cables with identicalmodes are used throughout the network ensuring


that the mode propagation constants of all the cablesegments are the same

(A2) The termination points are assumed to be eitherideal matchesmdashachieved using adaptive modal im-pedance matching [107 108]mdashor open circuit ter-minations Hence these termination points behaveindependently of frequency Due to this frequency-independent nature of terminations the branches andtermination points are perfectly balanced ensuringthat no mode mixing occurs anywhere in the net-work

These two assumptions are general and consider BPLtransmission regardless of the power grid type (either over-head or underground either LV or MV) and MIMO schemeconfiguration Because of the above assumptions the 119899modessupported by the BPL configurations are completely separategiving rise to 119899 independent transmission channels whichsimultaneously carry BPL signals [15ndash18 57 58 65 91 106]

32 UVD Modal Analysis of Overhead and UndergroundLVBPL and MVBPL Systems Since the number of transmitports 119899119879 and receive ports 119899119877 in MIMO systems is not alwaysequal existing EVD modal analysis for SISOBPL networksneeds to be integrated with SVD modal analysis for MIMOcommunications networks so as to permit the general studyof MIMOBPL systems [39 40 48 49 52]

Taking advantage of the simplicity of TM1 method UVDmodal analysis is the technique that perceives the modularconsideration of EVD and SVDmodal analyses concatenatesthem and achieves the common handling of traditionalSISOBPL and upcoming MIMOBPL systems Based onEVD and SVD modal transfer functions UVD modal anal-ysis computes the extended channel transfer functions andvarious performance and capacity metrics imperative for thefollowing MIMOBPL analysismdashfurther details concerningthe metrics are provided in Section 4

More specifically the EVD modal voltages V119898(119911) =

[119881119898

1(119911) sdot sdot sdot 119881

119898

119899(119911)]119879 and the EVD modal currents I119898(119911) =

[119868119898

1(119911) sdot sdot sdot 119868

119898

119899(119911)]119879 may be related to the respective EVD

line quantities V(119911) = [1198811(119911) sdot sdot sdot 119881119899(119911)]119879 and I(119911) =

[1198681(119911) sdot sdot sdot 119868119899(119911)]119879 via the EVD transformations [15ndash18 58

60 61]

V (119911) = T119881 sdot V119898(119911) (1)

I (119911) = T119868 sdot I119898(119911) (2)

where [sdot]119879 denotes the transpose of a matrix T119881 and T119868 are119899 times 119899 matrices depending on the frequency the power gridtype the physical properties of the cables and the geometryof the MTL configuration [15 17 58 60 61]

The TM1 methodmdashbased on the scattering matrix theory[15 18 19 68 69] and presented analytically in [18]mdashmodelsthe spectral relationship between 119881119898

119894(119911) 119894 = 1 119899 and

119881119898

119894(0) 119894 = 1 119899 introducing operators119867119898

119894sdot 119894 = 1 119899

so that

V119898 (119911) = H119898 V119898 (0) (3)

where

H119898 sdot = diag 1198671198981sdot sdot sdot sdot 119867

119898

119899sdot (4)

is a diagonal matrix operator whose elements 119867119898119894sdot 119894 =

1 119899 are the EVD modal transfer functions [15ndash18]Combining (1) and (4) the 119899times119899 channel transfer function

matrixHsdot relating V(119911) with V(0) through

V (119911) = H V (0) (5)

is determined from

H sdot = T119881 sdotH119898sdot sdot Tminus1119881 (6)

Based on (4) and (6) the 119899 times 119899 channel transfer functionmatrixHsdot of the overhead and underground SISOLVBPLand SISOMVBPL distribution networks is determined

Based on (5) at the transmitting end side of the over-head and underground LVBPL and MVBPL distributionnetworks n independent transmit portsmdashthat is 119881119894(0) 119894 =1 119899mdashwhich share the same return reference conductorcan be used At the receiving end side due to couplingeffects that provide interactions between different ports nindependent receive ports are availablemdashthat is 119881119894(119911) 119894 =1 119899 Consequently the 119899 times 119899 channel transfer functionmatrix Hsdot which is presented in (5) and (6) can describemultiport communications systems with 119899119879 (119899119879 le 119899)transmit and 119899119877 (119899119877 le 119899) receive ports This multiportcommunications system is referred to as 119899119879 times 119899119877MIMOBPLsystem [13 14 31 56 76 79] During the following analysis itis assumed that 119873119879 and 119873119877 are the active transmit port andthe active receive port set respectively hereafter

As it concerns the characterization of channels of theupcomingMIMOBPL systems from (6) the elements119867119894119895sdot119894 119895 = 1 119899 of channel transfer function matrix Hsdot with119894 = 119895 are the cochannel (CC) transfer functions while thosewith 119894 = 119895 are the cross-channel (XC) transfer functions where119867119894119895 119894 119895 = 1 119899 denotes the element of matrixHsdot in row 119894of column 119895 Altogether119867119894119895sdot 119894 119895 = 1 119899 are the transferfunctions of MIMO channels (either CCs or XCs)

On the basis of the previous channel characterization it isclear that the analysis concerning 119899119879times119899119877MIMOBPL systemsis sustained on the generalization of the analysis concerningMIMO channels of a full 119899 times 119899MIMOBPL systemmdashthat isall the available transmit and receive ports are active either4 times 4 MIMOLVBPL or 3 times 3 MIMOMVBPLmdashvia theproposed UVD modal analysis As shown in the followinganalysis theUVD modal analysis permits the study of full119899 times 119899 MIMOBPL and SISOBPL systems as case studies ofa general 119899119879 times 119899119877MIMOBPL system [39 40 49 52]

More specifically based on the above definition of theEVD line voltages the associated UVD transmit vectorV+(0) = [119881+

1(0) sdot sdot sdot 119881

+

119899(0)]119879 is related to the EVD line volt-

ages vector V(0) through a simple matrix relationship of theform

V (0) = D119873119879119881sdot V+ (0) (7)

where D119873119879119881

is the 119899 times 119899 UVD transmit voltage matrix withzero elements except in row 119895 of column 119895 where the value


is equal to 1 when transmit port 119895 belongs to the set 119873119879 Forexample in the case of full 119899 times 119899MIMOBPL and SISOBPLsystemsD119873119879

119881is equal to 119899times119899 identity matrix and 119899times119899matrix

with only one 1 on the diagonal at the position of the activetransmit port respectively Likewise the UVD receive vectorV+(119911) = [119881+

1(119911) sdot sdot sdot 119881

+

119899(119911)]119879 can be related to the EVD line

voltages vector V(119911) through a similar matrix relationship

V+ (119911) = D119873119877119881sdot V (119911) (8)

where D119873119877119881

is the 119899 times 119899 UVD receive voltage matrix withzero elements except in row 119894 of column 119894 where the valueis equal to 1 when receive port 119894 belongs to the set 119873119877 Forexample in the case of full 119899 times 119899MIMOBPL and SISOBPLsystems D119873119877

119881is equal to the 119899 times 119899 identity matrix and the

119899times119899matrix with only one 1 on the diagonal at the position ofthe active receive port respectively Combining (5) (7) and(8) the 119899times119899 extended channel transfer functionmatrixH+sdotis determined from

H+ sdot = D119873119877119881sdotH sdot sdotD119873119879

119881 (9)

Apart from the insertion of D119873119879119881

and D119873119877119881 UVD modal

analysis achieves the MIMO channel analysis generalizationby exploiting both already existing EVD and SVD modalanalyses [13 14]

More specifically adopting SVD modal analysis pre-sented in [13 14 39 40 49 52] the 119899 times 119899 extended channeltransfer function matrix H+sdot can be decomposed intomin119899119879 119899119877 parallel and independent SISO channels wheremin119909 119910 returns the smallest value between 119909 and 119910 Thenthe SVD modal transfer function matrix operator is given by

H119898

sdot =

T119867

119881sdotH+ sdot sdot

T119868

= (

T119867

119881sdotD119873119877119881sdot T119881) sdotH

119898sdot sdot (Tminus1

119881sdotD119873119879119881sdot

T119868)

(10)

where

H119898

sdot

= [

[

diag

119867

119898

119894sdot 0(min119899119877119899119879)times(119899minusmin119899119877119899119879)

0(119899minusmin119899119877119899119879)times(min119899119877119899119879) 0(119899minusmin119899119877119899119879)times(119899minusmin119899119877119899119879)

]

]

119894 = 1 min 119899119877 119899119879 (11)

is an 119899 times 119899 matrix operator whose nonzero elements

119867

119898

119894sdot

119894 = 1 min119899119877 119899119879 are (i) the singular values of H+sdotand (ii) the end-to-end SVD modal transfer functions [sdot]119867denotes the Hermitian conjugate of a matrix 0119898times119899 is an119898 times 119899 matrix with zero elements and

T119881 and

T119868 are 119899 times 119899unitary matrices [39 40 49 52] From (10) and (11) the SVDmodal transfer function matrix

H119898

sdot suitable for the MIMOcapacity computations of general 119899119879times119899119877MIMOBPL systemsis determined

4 Performance and Capacity Metrics ofOverhead and Underground MIMOLVBPLand MIMOMVBPL Systems

In this section several useful transmission and capacity met-rics which are well proven and used in the communicationsliterature [3 5 12ndash14] are synopsized in order to investigatethe behavior of MIMOBPL channels and systems whenUVDmodal analysis is applied

More particularly the ACG the RMS-DS the CB andvarious capacity measuresmdashsuch as cumulative capacitycapacity CCDF and GCmdashare reported and applied withthe purpose of investigating the transmission and capacityproperties of overhead and underground MIMOLVBPLand MIMOMVBPL channels and systems All these per-formances and capacity metrics are based on (i) extendedchannel transfer functions elements 119867+

119894119895sdot 119894 119895 = 1 119899

of MIMOBPL transmission channelsmdashas given by (9) and(ii) end-to-end SVD modal transfer function elements

119867119894sdot119894 = 1 min119899119877 119899119879mdashas given by (10)mdashof MIMOBPLtransmission channels

41 The Discrete Impulse Response Once the extended chan-nel transfer functions 119867+

119894119895sdot 119894 119895 = 1 119899 of MIMO

channels are known from (9) discrete extended impulseresponses ℎ+

119894119895119901= ℎ+

119894119895(119905 = 119901119879119904) 119901 = 0 119869 minus 1 119894 119895 = 1 119899

are obtained as the power of two J-point inverse discreteFourier transform (IDFT) of the discrete extended channeltransfer functions of MIMO channels

119867+

119894119895119902=

10038161003816100381610038161003816119867+

119894119895119902

10038161003816100381610038161003816119890119895120593+

119902 119902 = 0 119870 minus 1

0 119902 = 119870 119869 minus 1

= 119867+

119894119895(119891 = 119902119891119904) 119902 = 0 119870 minus 1

0 119902 = 119870 119869 minus 1 119894 119895 = 1 119899

(12)

where 119865119904 = 1119879119904 is the Nyquist sampling rate 119870 le 1198692

is the number of subchannels in the BPL signal frequencyrange of interest 119891119904 = 119865119904119869 is the flat-fading subchannelfrequency spacing and |119867+

119894119895119902| 119894 119895 = 1 119899 and 120601+

119894119895119902 119894 119895 =

1 119899 are the amplitude responses and the phase responsesof the discrete extended channel transfer functions of MIMOchannels respectively [71ndash73]

42 The ACG Since the BPL channels are frequency selec-tive the ACGs of MIMO channels |119867+

119894119895|2 119894 119895 = 1 119899 can

be calculated by averaging over frequency [71ndash73 76ndash79]

10038161003816100381610038161003816119867+

119894119895

10038161003816100381610038161003816

2

=

119869minus1

sum

119901=0

10038161003816100381610038161003816ℎ+

119894119895119901

10038161003816100381610038161003816

2

=1

119869

119869minus1

sum

119902=0

10038161003816100381610038161003816119867+

119894119895119902

10038161003816100381610038161003816

2

119894 119895 = 1 119899 (13)

43The RMS-DS It is a measure of the multipath richness ofa BPL channel The RMS-DS of MIMO channels 120590+

119894119895120591 119894 119895 =

1 119899 is determined from the following [71ndash73 76ndash79]

120590+

119894119895120591= 119879119904

radic120583(2)

1198941198950minus (1205831198941198950)

2

119894 119895 = 1 119899 (14)


where

1205831198941198950 =

sum119869minus1

119901=011990110038161003816100381610038161003816ℎ+

119894119895119901

10038161003816100381610038161003816

2

sum119869minus1

119901=0

10038161003816100381610038161003816ℎ+

119894119895119901

10038161003816100381610038161003816

2 119894 119895 = 1 119899 (15a)

120583(2)

1198941198950=

sum119869minus1

119901=0119901210038161003816100381610038161003816ℎ+

119894119895119901

10038161003816100381610038161003816

2

sum119869minus1

119901=0

10038161003816100381610038161003816ℎ+

119894119895119901

10038161003816100381610038161003816

2 119894 119895 = 1 119899 (15b)

44 The CB It is the range of frequencies over which thenormalized autocorrelation function of the channel transferfunction is over a certainCB correlation level119883 (usually set to09 07 or 05) that is the maximum bandwidth in which thesubchannels can be approximately considered flat-fading Asthe phase response of the extended channel transfer functionmay be assumed as uniformly distributed over [0 2120587] theCB ofMIMO channels can be determined from the following[47 80ndash82 109]

119861119894119895119888 (Δ119891) =

119864 119867+

119894119895119902sdot [119867+

119894119895(119891 = 119902119891119904 + Δ119891)]

lowast

119864 10038161003816100381610038161003816119867+

119894119895119902

10038161003816100381610038161003816

2

119902 = 0 1 lfloor(119869 minus 2) +Δ119891

119891119904

rfloor 119894 119895 = 1 119899

(16)

where Δ119891 is the frequency shift [sdot]lowast denotes the complexconjugate of an element and lfloor119909rfloor is the largest integer notgreater than 119909 From (16) CB119894119895119883 119894 119895 = 1 119899 is that valueof Δf such that 119861119894119895119888(Δ119891) = 119883 119894 119895 = 1 119899

45 The Capacity Capacity is the maximum achievabletransmission rate over a BPL channel Note that in the restof this review paper the common case is examined wherethe transmitting end does not have channel state information(CSI) [13 14] In the general case of 119899119879 times 119899119877 MIMOBPLsystems the MIMO capacity is calculated by using thefollowing [13 14 33 41 42 48 50 53ndash55]

119862MIMO

= 119891119904

119869minus1

sum

119902=0

min119899119879119899119877

sum

119894=1

log21 +

SNR (119902119891119904)119899119879

sdot

1003816100381610038161003816100381610038161003816

119867

119898

119894(119902119891119904)

1003816100381610038161003816100381610038161003816

2

(17)

where

SNR (119891) =119901 (119891)

119873 (119891)(18)

is the BPL signal-to-noise ratio (SNR) 119901(119891) is the selectedinjected power spectral density mask (IPSDM)mdashthat is rec-ommended by the regulation authoritiesmdashand 119873(119891) is thechannel noise PSD [16 37 110]

As less transmit and receive ports are deployed the capac-ity results tend to present the capacity results of SISOBPLsystems More specifically depending on the number oftransmit and receive ports appropriate well-known capacityexpressions can be usedmdashsee also [13 14 41 42]

(i) Single-Input Multiple-Output (SIMO) Capacity Ex-pression When one transmit and 119899119877 1 lt 119899119877 le 119899receive ports occur

(ii) Multiple-Input Single-Output (MISO) Capacity Ex-pression When 119899119879 1 lt 119899119879 le 119899 transmit and onereceive ports occur

(iii) SISO Capacity Expression When one transmit andone receive ports occur

According to the aforementioned differentMIMO SIMOMISO and SISO scheme configurations the resulting single-and multiport diversities may be classified into two majorclasses [13 14]

(i) Pure Scheme Configurations This class contains theelementary single- and multiport implementationsnamely 119899 times 119899 MIMO 1 times 119899 SIMO 119899 times 1 MISO andall SISO systems

(ii) Mixed Scheme Configurations This class contains allthe other multiport implementations that may bedeployed

On the basis of the previous capacity expressions threeadditional capacity measures are reported and are applied inthe following analysis

(i) Cumulative Capacity The cumulative capacity is thecumulative upper limit of information which can bereliably transmitted over the end-to-endBPL channelIn fact cumulative capacity describes the aggregatecapacity effect of all subchannels of the examinedfrequency band [13 14]

(ii) Capacity CCDFThe capacityCCDFdefines the statis-tics of cumulative capacity of BPL channels specify-ing the cumulative capacity probability of exceeding aspecific cumulative capacity threshold [13 14]

(iii) GC The GC is introduced as the ratio of the MIMOcumulative capacity of a BPL system over its respec-tive SISO one (usually set to SISOCC which is thebest todayrsquos BPL system cumulative capacity)

5 Numerical Results and Discussion

The simulations of various types of overhead and under-ground MIMOLVBPL and MIMOMVBPL transmissionchannels and systems aim at investigating (a) their broad-band MIMO transmission characteristics (b) how theirspectral behavior is affected by several factors (c) statisticalremarks obtained on the basis of simulated overhead andundergroundLVBPL andMVBPLMTL configurations and(d) how the capacity measures are influenced by the imple-mentation of various MIMOBPL scheme configurations

For the numerical computations the overhead andunderground LV and MV distribution line configurationsdepicted in Figures 1(a)ndash1(d) have been considered Asmentioned in Section 31 since the modes supported by theoverhead and underground LVBPL andMVBPL configura-tions may be examined separately it is assumed for simplicity


that the BPL signal is injected directly into the EVD modes[15ndash18 57ndash65 67] thus the complicated EVD and SVDmodal analyses of [13 14 41 42 60 61 81] briefly outlined inSection 32 are bypassed and the practical implementationof UVD modal analysis is further facilitated

As it concerns the simulation model properties theNyquist sampling rate 119865119904 and flat-fading subchannel fre-quency spacing 119891119904 are assumed equal to 200MHz and01MHz respectively Moreover the number of subchannels119870 in the BPL signal frequency range of interest and theJ-point IDFT are assumed equal to 991 and 2048 respectively[71ndash73 111 112]

As it regards the overhead and underground LVBPL andMVBPL MTL configurations the simple BPL topology ofFigure 2 having 119873 branches is considered Based on [15ndash1857 58 67 87] and with reference to Figure 2 the transmittingand the receiving ends are assumedmatched to the character-istic impedance of the mode considered whereas the branchterminations 119885119887119896 119896 = 1 119873 connected at positions1198601015840

119896 119896 = 1 119873 respectively are assumed open circuits

Path lengths sum119873+1119896=1

119871119896 of the order of 200m are encounteredin underground BPL transmission [83 87 92] These pathlengths are quite shorter compared to the overhead BPLcase where path lengths up to 1000m may be encountered[16 18 20 21 57 58 62 67 83 85 87 92 99 113 114]

With reference to Figure 2 five indicative overhead topol-ogies which are common for both overhead LVBPL andMVBPL systems concerning end-to-end connections ofaverage lengths equal to 1000m have been examined [13 1416]

(1) A typical overhead urban topology (overhead urbancase A) with119873 = 3 branches (L1 = 500m L2 = 200mL3 = 100m L4 = 200m L1198871 = 8m L1198872 = 13m andL1198873 = 10m)

(2) An aggravated overhead urban topology (overheadurban case B) with 119873 = 5 branches (L1 = 200mL2 = 50m L3 = 100m L4 = 200m L5 = 300m L6 =150m L1198871 = 12m L1198872 = 5m L1198873 = 28m L1198874 = 41mand L1198875 = 17m)

(3) A typical overhead suburban topology (overheadsuburban case) with 119873 = 2 branches (L1 = 500mL2 = 400m L3 = 100m L1198871 = 50m and L1198872 = 10m)

(4) A typical overhead rural topology (overhead ruralcase) with only 119873 = 1 branch (L1 = 600m L2 =400m and L1198871 = 300m)

(5) The overhead ldquoLOSrdquo transmission along the sameend-to-end distance L= L1 + sdot sdot sdot + L119873+1 = 1000mwhen no branches are encountered It corresponds tothe line-of-sight (LOS) path in wireless transmission

The respective indicative underground BPL topologieswhich are common for both underground LVBPL and MVBPL systems concern 200m long end-to-end connectionsThese topologies are [13 14 16 18] as follows

(1) A typical underground urban topology (undergroundurban case A) with 119873 = 3 branches (L1 = 70m

L2 = 55m L3 = 45m L4 = 30m L1198871 = 12m L1198872 = 7mand L1198873 = 21m)

(2) An aggravated underground urban topology (under-ground urban case B) with 119873 = 5 branches(L1 = 40m L2 = 10m L3 = 20m L4 = 40m L5 = 60mL6 = 30m L1198871 = 22m L1198872 = 12m L1198873 = 8m L1198874 = 2mand L1198875 = 17m)

(3) A typical underground suburban topology (under-ground suburban case) with 119873 = 2 branches(L1 = 50m L2 = 100m L3 = 50m L1198871 = 60m andL1198872 = 30m)

(4) A typical underground rural topology (rural case)with only 119873 = 1 branch (L1 = 50m L2 = 150m andL1198871 = 100m)

(5) The underground ldquoLOSrdquo transmission along the sameend-to-end distance L= L1 + sdot sdot sdot + L119873+1 = 200mwhen no branches are encountered

Finally as it concerns the properties of MIMOBPLscheme configurations the proposed study of general 119899119879times119899119877MIMOBPL systemsmdashoutlined in Sections 31 and 32mdashviaUVD modal analysis is considered To evaluate the capacityperformance of overhead and underground MIMOLVBPLand MIMOMVBPL transmission channels and systems inthe case of overhead BPL transmission a uniform additivewhite Gaussian noise (AWGN) PSD level is assumed equal tominus105 dBmHz [16 18 57 67] that is higher compared to thesignificantly lowerAWGNPSD encountered in undergroundBPL transmission assumed equal to minus135 dBmHz [18 97115ndash117] As it is usually done [13 14] the noise AWGNPSDof each power grid type is assumed common (i) betweencorresponding LV andMV systems and (ii) among all MIMOchannels of the same MTL configuration With regard topower constraints in the 3ndash88MHz frequency range the sameIPSDM limits between LVBPL and MVBPL are assumed[13 14 16 18 118 119] In detail at frequencies below 30MHzaccording to Ofcom in the 1705ndash30MHz frequency rangemaximum levels of minus60 dBmHz and minus40 dBmHz consti-tute appropriate IPSDMs for overhead and undergroundMIMOBPL systems respectively providing presumption ofcompliance with the current FCC Part 15 limits [16 18 118119] In the 30ndash88MHz frequency range maximum IPSDMlevels equal to minus77 dBmHz and minus57 dBmHz for overheadand underground MIMOBPL systems respectively areassumed to provide a presumption of compliance in thisfrequency range [16 18 118 119]

51 Channel Attenuation of Overhead and UndergroundMIMOLVBPL and MIMOMVBPL Channels Comparisonof UVD Modal AnalysismdashResults with Other Already Vali-dated Ones In Figures 3(a) and 3(b) the median values ofchannel attenuation in the frequency band 1ndash100MHz forCCs (solid lines) and for XCs (dashed lines) are plottedwith respect to frequency for the aforementioned indicativeoverhead LVBPL and MVBPL topologies respectively inthe frequency range 1ndash100MHz In Figures 3(c) and 3(d) thesame curves are plotted in the case of underground LVBPLand MVBPL systems respectively


V+in

L2L1 Lk LN+1

A998400N

A998400kA998400

2A9984001

A BTransmittingend end

Vminusout

ReceivingLb1 Lb2 Lbk LbN

A1 A2 Ak AN

L3 middot middot middot LK+1 middot middot middot LN

middot middot middot middot middot middot

Figure 2 End-to-end BPL connection with119873 branches [16 18]

0

0

50

50

100

100

150

Frequency (MHz)

Chan

nel a

ttenu

atio

n (d

B)

(a)

0 50 100

Frequency (MHz)

0

50

100

150

Chan

nel a

ttenu

atio

n (d

B)

(b)

0 50 100Frequency (MHz)

0

50

100

150

Urban case AUrban case B

Rural caseldquoLOSrdquo case

Suburban case

Chan

nel a

ttenu

atio

n (d

B)

(c)


0

50

100

150



Suburban case

Chan

nel a

ttenu

atio

n (d

B)

(d)

Figure 3 Frequency spectra of attenuation coefficients of MIMOBPL channels CCs (solid lines) and XCs (dashed lines)mdashthe subchannelfrequency spacing is equal to 1MHz (a) Overhead LV (b) Overhead MV (c) Underground LV (d) Underground MV


Table 1 Statistical values of ACG for various overhead and underground MIMOLVBPL and MIMOMVBPL channels

ACG (dB)Min Max Mean Standard deviation

OV UN OV UN OV UN OV UNCC

LV minus1270 minus1918 minus463 minus946 minus836 minus1413 303 363MV minus1301 minus2764 minus492 minus1844 minus867 minus2340 304 362

XCLV minus2954 minus3130 minus2317 minus2323 minus2604 minus2694 267 305MV minus2464 minus6119 minus1799 minus5414 minus2099 minus5771 274 283

AllLV minus2954 minus3130 minus463 minus946 minus1720 minus2054 970 745MV minus2464 minus6119 minus492 minus1844 minus1483 minus4055 704 1834

OV overhead UN underground

Observations based on Figures 3(a)ndash3(d) were made asfollows

(i) Comparing Figures 3(a) and 3(c) with the respectiveFigures 3(a) and 3(b) of [14] it is evident that UVDmodal analysis combined with TM1 method givesalmost identical channel attenuation results with thecorresponding ones of SVD modal analysis whenit is combined with the accurate TM2 method foroverhead and underground MIMOLVBPL chan-nels respectively The same high channel attenuationprecision is also achieved when Figures 3(b) and3(d) are compared with the respective Figures 3(a)and 3(c) of [13] Again UVD modal analysis com-bined with TM1 method gives the same attenuationresults with those derived from the combination ofSVD modal analysis with TM1 method for over-head and underground MIMOMVBPL channelsrespectively Therefore UVD modal analysis offersexcellent accuracy for all overhead and undergroundMIMOLV and MIMOMV BPL channels that areexamined across this review paper

(ii) Due to the common bus-bar system topology [1314 54 78] in overhead and underground LVBPLandMVBPL systems the channel attenuation curvesbetween CCs andXCs present similarities concerningtheir curve slopes and the position and extent ofspectral notches This indicates the strong couplingbetween the individual MIMO channels Actuallythe attenuation of MIMO channels depends dras-tically on the frequency the power grid type theMIMO channel type the MTL configuration thephysical properties of the cables used the end-to-endmdashldquoLOSrdquomdashdistance and the number the elec-trical length and the terminations of the branchesencountered along the end-to-end BPL signal trans-mission path [15 16 18 24 66 92 99 113 120]

(iii) Regardless of the power grid type and power gridtopology XCs present higher channel attenuations incomparison with CCs ones [48 53 55] The channel

attenuation difference between CCs and XCs for thesame topology exceeds 15 dB in most of the cases[13 14]

(iv) In accordance with [15ndash18 37] and based on thepicture of the spectral behavior observed fromFigures 3(a)ndash3(d) the overhead and undergroundMIMOLVBPL and MIMOMVBPL channels maybe classified into three channel classes ldquoLOSrdquo goodand bad channels

(v) Due to the favorable low-loss transmission char-acteristics of overhead MIMOLVBPL overheadMIMOMVBPL and underground MIMOLVBPLchannels these power grid types can define anexcellent cooperative communications platform thatis suitable for last mile broadband communicationsservices [15ndash18 83 87 91 92 94 121]

52 ACG Hereafter it is assumed for simplicity that onlythe median values of ACG over CCs and XCs for each ofthe aforementioned indicative overhead and undergroundLVBPL and MVBPL topologies will be studied in thefrequency band 1ndash100MHz

In Table 1 the minimum value the maximum value themean and the standard deviation of the ACG for the over-head and underground MIMOLVBPL and MIMOMVBPL channels (CC XC and all together) are reported

In Figure 4(a) the quantile-quantile plots of the ACGof MIMO channels for the indicative overhead LVBPL andMVBPL topologies are depicted in order to examine the nor-mality of the ACG distribution functions [78] In these plotsthe sample quantiles of ACGs are displayed versus theoreticalquantiles from normal distributions In Figure 4(b) the ACGCDF curves of MIMO channels (dashed lines) for the sameoverhead LVBPL and MVBPL topologies are also plottedFour CDF curves of four random variables normally dis-tributedwithmean and standard deviation reported inTable 1(overhead LVCC overhead LVXC overhead MVCC andoverhead MVXC) are also drawn (bold lines) Finally aunified ACGCDF curve which is normally distributed and is


Standard normal quantiles

minus1 0

0

1minus70

minus60

minus50

minus40

minus30

minus20

minus10

ACG

(dB)

Overhead XC

Overhead CC

(a)

1

08

06

04

02

0minus40minus60 minus20 0

ACG (dB)

Overhead XC

Overhead CC

ACG

CD

F (A

CGlt

absc

issa)

(b)

0

minus70

minus60

minus50

minus40

minus30

minus20

minus10

ACG

(dB)

Standard normal quantilesminus1 0 1

Underground XC

Underground CC

LVBPL casesMVBPL cases

(c)

1

08

06

04

02


ACG (dB)

Underground XC

Underground CC

ACG

CD

F (A

CGlt

absc

issa)

LVBPL casesMVBPL casesMIMOLVBPL channelsMIMOMVBPL channelsUN1 ACG CDF cases

(d)

Figure 4 (a) Quantile-quantile ACG plots of overhead MIMOLVBPL (+) and MIMOMVBPL (+) channels compared to theoreticalnormal distribution lines of overhead LVBPL and MVBPL cases (b) ACG CDF curves of overhead MIMOLVBPL and MIMOMVBPLchannels compared to theoretical normal distribution lines of overhead LVBPL overheadMVBPL andUN1ACGCDF cases (c) (d) Similarplots in underground BPL cases

common for all overhead and underground MIMOLVBPLand MIMOMVBPL channelsmdashdetailed in the followinganalysismdashis also given In Figures 4(c) and 4(d) similarplots are displayed in the case of underground LVBPL andMVBPL MIMO channels

From Figures 4(a)ndash4(d) the following are clearly demon-strated

(i) Despite the power grid type ACG curves of MIMOchannels follow normal distributions This is


explained observing the simulation plots that areclose to linear ones (bold lines) The lognormality(normality) of the ACG may be justified by themultipath nature of BPL signal propagation and theTL modeling based on cascaded two-port networkmodules [4 71ndash73]

(ii) The ACG curves can be fitted by appropriate lognor-mal distributions for each overhead and undergroundLVBPL andMVBPL topology Small deviationsmayoccur for BPL topologies with small electrical lengthRecently suitable lognormal distributions have beenproposed in the analysis of various SISOBPL chan-nels [4 31 56 71ndash73 78 79]

(iii) As expected the spectral behavior of overhead andunderground MIMOLVBPL and MIMOMVBPLchannels depicted in Figures 3(a)ndash3(d) is reflectedon respective ACG distributions Indeed the ACGdistributions of underground MVBPL transmissionare significantly worse than those of other power gridtypes

(iv) Examining ACG metric it is verified that CCsdemonstrate better transmission characteristics incomparison with XCs regardless of the power gridtype and power grid topology Due to these favorablecharacteristics of CCs their exploitation is mainlypromoted during the implementation of SISOBPLsystems Actually this is the typical SISOBPL imple-mentation scheme in todayrsquos BPL networks

(v) From Table 1 and Figures 4(b) and 4(d) in orderto promote the common ACG CDF curve analysisof overhead and underground MIMOLVBPL andMIMOMVBPL channels a unified ACGCDF curveis consideredThis ACGCDF curve which is denotedas UN1 ACG CDF is normally distributed withmean (mean of the eight abovementioned ACG CDFmeans) and standard deviation (mean of the eightabovementioned ACG CDF standard deviations)equal to 2328120583s and 308 120583s respectively The UN1ACG CDF curve allows the introduction of commonACG CDF curve analysis regardless of the overheadand underground LVBPL and MVBPL topologiesdelivering a first statistical analysis intuition towardsthe existing deterministic analysis [71]

53 RMS-DS In this subsection it is assumed that only themedian values of RMS-DS over CCs and XCs for each ofthe aforementioned indicative overhead and undergroundLVBPL topologies will be studied in the frequency band1ndash100MHz

In Table 2 the minimum value the maximum value themean and the standard deviation of theRMS-DS for the over-head and underground MIMOLVBPL and MIMOMVBPL channels (CC XC and all together) are reported

In Figure 5(a) the quantile-quantile plots of the RMS-DSof MIMO channels for the overhead LVBPL and MVBPLtopologies are depicted in order to test the normality ofthe RMS-DS distribution functions [78] In these plots the

sample quantiles of RMS-DS are displayed versus theoreticalquantiles from normal distributions In Figure 5(b) theRMS-DS CDF curves of MIMO channels (dashed lines)for the overhead LVBPL and MVBPL topologies are plot-ted Four CDF curves of four random variables that arenormally distributed with mean and standard deviationreported in Table 2 (overhead LVCC overhead LVXCoverhead MVCC and overhead MVXC) are also plotted(bold lines) Finally a unified RMS-DSCDF curve that is nor-mally distributed which is common for both overhead andundergroundMIMOLVBPL andMIMOMVBPL channelsand is detailed in the following analysis is also given InFigures 5(c) and 5(d) similar plots are given in the caseof undergroundMIMOLVBPL andMIMOMVBPL chan-nels

From Figures 5(a)ndash5(d) the following are observed

(i) RMS-DS is lognormally distributed with goodapproximation in all overhead and undergroundMIMOLVBPL and MIMOMVBPL channels forall overhead and underground LVBPL and MVBPLtopologies [4 31 56 71ndash73 78 79] However RMS-DS is highly variable depending on power grid typeand power grid topology while RMS-DS increases asthe topology complexity rises In fact bad channelclass presents higher RMS-DS values than those ofldquoLOSrdquo and good channel classes due to the existingaggravated multipath environment [47 122]

(ii) In general overheadMIMOLVBPL overheadMIMOMVBPL and underground MIMOMVBPL chan-nels present comparable RMS-DS results HoweverundergroundMIMOLVBPL channels present lowerRMS-DS values than those of the aforementionedthreeMIMOBPL systemsThis renders undergroundMIMOLVBPL transmission as more multipathresistant and more appropriate for sensitive SG appli-cations

(iii) From Table 2 and Figures 5(b) and 5(d) in orderto bypass the RMS-DS CDF dependency on powergrid type and power grid topologies a unified RMS-DS CDFmdashdenoted as UN1 RMS-DS CDFmdashnormallydistributed with mean (mean of the eight aforemen-tioned RMS-DS CDF means) and standard deviation(mean of the eight aforementioned RMS-DS CDFstandard deviations) equal to 083 120583s and 040 120583srespectively is considered The UN1 RMS-DS CDFcurve allows the replication of the variability ofoverhead and underground LVBPL and MVBPLtopologies conferring a stochastic aspect to the exist-ing deterministic analysis [71]

Analyzing the statistical performance metrics of Tables 1and 2 a fundamental property of several wireline channels(eg DSL coaxial and phone links) may be also pointedout in the overhead and underground MIMOLVBPL andMIMOMVBPL channels the correlation betweenACG andRMS-DS This correlation can be easily confirmed simply byobserving the scatter plot presented in Figures 6(a) and 6(b)where the above set of numerical results of MIMO channels


Overhead XC

Overhead CC


0

05

15

2

1

minus2 0 2

RMS-

DS

(120583s)

(a)

RMS-DS (120583s)0 1 2

1

08

06

04

02

0

Overhead XC

Overhead CC

RMS-

DS

CDF

(RM

S-D

Slt

absc

issa)

(b)

0

05

15

2

1

RMS-

DS

(120583s)


Underground XC

Underground XC

Underground CC


(c)

1

08

06

04

02

0

RMS-

DS

CDF

(RM

S-D

Slt

absc

issa)

RMS-DS (120583s)0 1 2

Underground XC

Underground CC

LVBPL casesMVBPL casesMIMOLVBPL channelsMIMOMVBPL channelsUN1 RMS-DS CDF cases

(d)

Figure 5 (a) Quantile-quantile plots of RMS-DS of overheadMIMOLVBPL (+) andMIMOMVBPL (+) channels compared to theoreticalnormal distribution lines of overhead LVBPL and MVBPL cases (b) RMS-DS CDFs of overhead MIMOLVBPL and MIMOMVBPLchannels compared to theoretical normal distribution lines of overhead LVBPL overhead MVBPL and UN1 RMS-DS CDF cases(c) (d) Similar plots in underground BPL cases


Table 2 Statistical values of RMS-DS for various overhead and underground MIMOLVBPL and MIMOMVBPL channels

RMS-DS (120583s)Min Max Mean Standard deviation


LV 009 026 112 083 068 053 042 020MV 038 061 108 146 078 100 030 034

XCLV 011 011 192 078 106 050 066 024MV 011 056 184 159 104 106 063 037

AllLV 011 011 192 078 106 050 066 025MV 028 056 181 159 111 106 055 037


for the aforementioned indicative overhead andundergroundBPL topologies respectively are plotted

In Figures 6(a) and 6(b) except for the numerical resultsa set of regression trend lines is also presented namely

(i) regression line set of the form (120590120591)120583s = minusV sdot (|119867|2)dB +

119908 where (|119867|2)dB is the ACG in dB and (120590120591)120583s is theRMS-DS in 120583s of the MIMOLVBPL and MIMOMVBPL channels which are examined Analyticallythe regression trend lines are as follows

(1) ANT approach as given by [78] with robustregression parameters V and119908 assumed equal to00197 120583sdB and 0 120583s respectively

(2) GAL approach as given by [71ndash73] withrobust regression parameters assumed V =

00075 120583sdB and 119908 = 0183 120583s(3) the proposed UN1OV UN1UN and UN1

approaches with regression parameters[V 119908] equal to [minus00303 120583sdB 04066 120583s][minus001265 120583sdB 03829 120583s] and[minus001029 120583sdB 0591 120583s] respectively Theleast squares fitting method is applied to theaforementioned overhead underground andall together numerical results for the evaluationof parameters [V 119908] of UN1OV UN1UN andUN1 approaches respectively

From Figures 6(a) and 6(b) the following should bementioned

(i) In spite of the power grid type power grid topol-ogy and MIMOBPL channel type the negativeslopes of all regression lines clearly confirm that theACG and RMS-DS of overhead and undergroundMIMOLVBPL and MIMOMVBPL channels arenegatively correlated lognormal random variables

(ii) The existing regression trend lines (ANT and GALapproaches) in the literature [71ndash73 78] betterdescribe the correlation between RMS-DS and ACG

in underground BPL systems rather than in overheadones This is due to the fact that the 1000m longoverhead connections present high-multipath behav-ior compared to the 200m long connections of theunderground case where ldquoLOSrdquo attenuation is theprimary attenuation mechanism

(iii) On the basis of the correlation between ACG andRMS-DS three new approximations are proposedwhich are suitable for the BPL statistical channelmodeling where this correlation is imposed by design[71ndash73] The proposed UN1OV UN1UN and UN1regression lines suggest a unified regression approachthat describes the ACG and RMS-DS correlationbetween overhead underground and all togetherMIMO channels respectively regardless of the powergrid type level (either LV or MV) power grid topol-ogy and MIMOBPL channel type

54 CB In the rest of this subsection it is assumed thatonly the median values of CB over CCs and XCs for eachof the aforementioned indicative overhead and undergroundLVBPL and MVBPL topologies will be studied in thefrequency band 1ndash100MHz

In Table 3 the minimum value the maximum value themean and the standard deviation of the CB05 for the over-head and underground MIMOLVBPL and MIMOMVBPL channels (CC XC and all together) are presented

Observing statistical performance metrics of Tables 2and 3 a second fundamental property of several wirelinesystems is also validated in the overhead and undergroundMIMOLVBPL and MIMOMVBPL channel cases thecorrelation between CB and RMS-DS In Figures 7(a) and7(b) the scatter plots of the RMS-DS versus the CB05 forthe numerical results of overhead and underground channelsrespectively are presented Apart from the numerical resultsa set of proposed regression trend lines is also given

(i) hyperbolic trend lines set of the form (120590120591)120583s =

119910 sdot (CB05)MHz where (CB05)MHz is the CB05 inMHz (i) the proposed UN2OV UN2UN and UN2


0

05

15

2

1

RMS-

DS

(120583s)

0minus60 minus50 minus40 minus30 minus20 minus10

ACG (dB)

Overhead LVCCOverhead LVXCOverhead MVCCOverhead MVXC

UN1 approachUN1OV approach

ANT approachGAL approach

(a)

0

05

15

2

1

RMS-

DS

(120583s)

Underground LVCCUnderground LVXCUnderground MVCCUnderground MVXC

UN1 approachUN1UN approach



ACG (dB)

(b)

Figure 6 Scatter plot of RMS-DS versus ACG for simulated MIMOLVBPL and MIMOMVBPL channels and various regressionapproaches (a) Overhead (b) Underground

0 10 20 30 40 50CB (MHz)



0

05

15

2

1

RMS-

DS

(120583s)

(a)

0 10 20 30 40 50CB (MHz)



0

05

15

2

1RM

S-D

S (120583

s)

(b)

Figure 7 Scatter plot of RMS-DS versus CB05

for simulated MIMOLVBPL and MIMOMVBPL channels and various regressionapproaches (a) Overhead (b) Underground

approaches with parameter 119910 assumed equal to03921120583ssdotMHz 04555 120583ssdotMHz and 04155120583ssdotMHzrespectively obtained using robust linear least squareerror fitting between the trend line and the observedoverhead underground and all together numericalresults respectively

On the basis of the second fundamental property whichis confirmed for overhead and underground MIMOLVBPLand MIMOMVBPL channels additional observations areadded [100 123ndash125]

(i) CB and RMS-DS are inversely related and their rela-tion can be approximated by appropriate hyperbolicfunctions UN2OV UN2UN and UN2 approaches arein very good agreement with the overhead under-ground and all together simulation results respec-tively proposing a unified consideration of the CBRMS-DS correlation

(ii) The variability of the pair values of CB05 andRMS-DSprimarily depends on the power grid type the over-head and underground MIMOLVBPL and MIMOMVBPL channel types and the power grid topologycharacteristics

(iii) UN2OV UN2UN and UN2 approaches quantifythe effects of frequency selective behavior Despitethe CB and RMS-DS differences between overhead

and underground LVBPL and MVBPL channelsdue to multipath environment and ldquoLOSrdquo attenu-ation UN2OV and UN2UN approaches accuratelyfit the numerical results of overhead and under-ground MIMOBPL channels respectively Takingunder consideration the almost identical UN2OV

and UN2UN behaviors UN2 approach promotes anelementary step towards a common transmissionhandling between overhead and underground LVBPL and MVBPL systems More specifically UN2approach fits very well either numerical results orUN2OV and UN2UN approaches regardless of thepower grid type power grid topology and MIMOchannel type

(iv) Except for UN1OV UN1UN and UN1 regression linesthat have been proposed for the wireline statisticalchannel modeling by design UN2OV UN2UN andUN2 approachesmay operate as evaluationmetrics ofthe applied statistical channel models [71ndash79]

55 Capacity The potential transmission performance interms of capacity in the frequency range 3ndash88MHz isevaluated under the assumption of OfcomIPSDMs andAWGNPSDs that have been presented in the introductionof Section 4 Moreover in this subsection it is assumed thatonly the median values of SISO capacities over CCs and


Table 3 Statistical values of CB05 for various overhead and underground MIMOLVBPL and MIMOMVBPL channels

CB05 (MHz)Min Max Mean Standard deviation


LV 070 080 4950 37 1182 1232 2114 1631MV 080 050 4760 560 1178 232 2016 228

XCLV 020 040 3150 48 700 1020 1372 2113MV 020 030 3220 880 716 206 1402 377

AllLV 020 040 3250 48 720 1020 1416 2113MV 020 030 27 880 576 206 1188 377


XCs for each of the aforementioned indicative overheadand underground LVBPL and MVBPL topologies will beinvestigated (see details in [13 14])

In Figures 8(a) and 8(b) the respective SISO cumulativecapacities of the LVBPL and MVBPL systems for theaforementioned indicative overhead topologies are plottedwith respect to frequency In Figure 8(c) the SISO capacityCCDF curves for the overhead LVBPL andMVBPL systemsare drawn In Figures 8(d)ndash8(f) similar plots are given in thecase of underground LVBPL and MVBPL systems

From Figures 8(a)ndash8(f) the following are observed

(i) As it has already been mentioned the analysis con-cerning general 119899119879 times 119899119877MIMOBPL systems is basedonUVDmodal analysis concerningMIMO channelsOnly in the case of SISOBPL systems their behavioris defined directly by the corresponding SISOBPLmdasheither SISOCC or SISOXCmdashchannel characteriza-tion

(ii) The extremely strong attenuation exhibited by under-ground MVBPL systems is counterbalanced by thesignificantly higher IPSDM limits imposed on under-ground MVBPL transmission lower AWGNPSDand shorter average end-to-end transmission dis-tances Thus the transmission efficiency of under-ground MIMOMVBPL systems becomes compara-ble to that of overhead LVBPL and MVBPL systems[13 14 16 18]

(iii) The significantly higher IPSDM limits combined withshorter average end-to-end transmission distanceslow channel attenuation and low noise characteristicsmake the transmission efficiency of undergroundSISOLVBPL systems incomparable to the rest ofBPL systems [16 18] In detail the capacities ofunderground SISOLVBPL systems are multiple ofan approximate factor of 3 in comparison with thecapacities of the respective overhead SISOLV BPLoverhead SISOMVBPL and underground SISOMVBPL systems This feature renders underground

LVBPL systems as a BPL backbone network enhanc-ing the necessary intraoperability between overheadand underground LVBPL and MVBPL networks

(iv) In both overhead and underground LVBPL andMVBPL topologies CCs are those that statisticallyconvey higher SISO capacities than XCs For theoverhead LVBPL and MVBPL topologies the 80thpercentile of the SISO capacity CCDF curves forCCs is equal to 608Mbps and 596Mbps respectivelywhile the respective CCDF curves for XCs are equalto 206Mbps and 293Mbps respectively For theunderground LVBPL and MVBPL topologies the80th percentile of the SISO capacity CCDF curves forCCs is equal to 1849Mbps and 790Mbps respectivelywhile the respective CCDF curves for XCs are equalto 1514Mbps and 334Mbps respectively Even if CCsandXCs present similarities as they concern the chan-nel attenuation curves due to the common bus-barsystem topologies their significant difference affectscritically the achievable SISO capacities This yieldsa disadvantage in practical applications where thecoupling scheme applied at transmitting end needs tobe the same at the receiving end [78 126]

Continuing MIMOBPL capacity analysis differentMIMOBPL scheme configurations which are diversified bytheir number of transmit and receive ports are investigatedfor both overhead and underground MIMOLVBPL andMIMOMVBPL systems [41 42 48 53 55 127 128] Asalready described in [13 14] it is assumed that only themedian values of all possible SIMO MISO and MIMOsystem capacities for each of the aforementioned indicativeoverhead and underground LVBPL andMVBPL topologieswill be studied

To investigate the capacity potential of various single- andmultiport BPL implementations pure scheme configurationsare first studied and compared by means of their cumulativecapacity and capacity CCDF curves In Figures 9(a) 9(c) and9(e) the 1 times 4 SIMO cumulative capacity the 4 times 1MISOcapacity and the 4 times 4MIMO capacity respectively for the


20 40 60 80Frequency (MHz)

00

500

1000

2000

1500

Cum

ulat

ive c

apac

ity (M

bps)

Overhead LVXC

Overhead LVCC

(a)

0

500

1000

2000

1500

Cum

ulat

ive c

apac

ity (M

bps)

0 20 40 60 80Frequency (MHz)

Overhead MVCC

Overhead MVXC

(b)

0

Capa

city

CCD

F (c

apac

itygt

absc

issa)

01

02

03

04

05

06

07

08

09

1

0 500 1000 1500 2000

Overhead LVCC

Overhead MVCC

Overhead MVXC

Overhead LVXC

Capacity (Mbps)

(c)

0

500

1000

2000

1500

Cum

ulat

ive c

apac

ity (M

bps)

UndergroundLVCC

UndergroundLVXC



Suburban case


0

(d)

0

500

1000

2000

1500

Cum

ulat

ive c

apac

ity (M

bps)

Underground MVXC

UndergroundMVCC



Suburban case


0

(e)

0 500 1000 1500 20000

01

02

03

04

05

06

07

08

09

1

UndergroundLVCC

UndergroundLVXC

UndergroundMVCC

Underground

Capacity (Mbps)

Capa

city

CCD

F (c

apac

itygt

absc

issa)

MVXC

(f)

Figure 8 SISO capacity characteristics of the MIMO systems for the aforementioned indicative overhead and underground LVBPL andMVBPL topologies (a) (b) SISO cumulative capacity of overhead LVBPL and MVBPL systems respectively (c) SISO capacity CCDFof overhead LVBPL and MVBPL systems (d) (e) SISO cumulative capacity of underground LVBPL and MVBPL systems respectively(f) SISO capacity CCDF of underground LVBPL and MVBPL systems

aforementioned indicative overhead LVBPL topologies areplotted with respect to frequency In Figures 9(b) 9(d) and9(f) the 1 times 3 SIMO cumulative capacity the 3 times 1MISOcapacity and the 3 times 3 MIMO capacity respectively for

the aforementioned indicative overhead MVBPL topologiesare displayed with respect to frequency In Figure 9(g) theSISOCC capacity CCDF the SISOXC capacity CCDF the1 times 4 SIMO capacity CCDF the 4 times 1MISO capacity CCDF








0

500

1000

2000

3000

1500

2500

0

500

1000

2000

3000

1500

2500

Cum

ulat

ive c

apac

ity (M

bps)

0

500

1000

2000

3000

1500

2500

Cum

ulat

ive c

apac

ity (M

bps)

0

500

1000

2000

3000

1500

2500

Cum

ulat

ive c

apac

ity (M

bps)

0

500

1000

2000

3000

1500

2500

Cum

ulat

ive c

apac

ity (M

bps)

0

500

1000

2000

3000

1500

2500

Cum

ulat

ive c

apac

ity (M

bps)

Cum

ulat

ive c

apac

ity (M

bps)

00

1000 2000 3000

Capa

city

CCD

F (c

apac

itygt

absc

issa)

01

02

03

04

05

06

07

08

09

1

(g)Capacity (Mbps)

0 1000 2000 3000C (Mb )

3times3 MIMOMV3times1 MISOMV1times3 SIMOMV

SISOCCLVSISOXCLV1 times 4 SIMOLV

4 times 4 MIMOLVSISOCCMVSISOXCMV

4 times 1 MISOLV



Suburban case



Suburban case

(a) (b)

(c)

(e) (f)

(d)

Figure 9 Capacity characteristics of various pure scheme configurations for the aforementioned indicative overhead LVBPL and MVBPLtopologies (a) 1 times 4 SIMOLV cumulative capacity (b) 1 times 3 SIMOMV cumulative capacity (c) 4 times 1 MISOLV cumulative capacity(d) 3 times 1 MISOMV cumulative capacity (e) 4 times 4MIMOLV cumulative capacity (f) 3 times 3MIMOMV cumulative capacity (g) SISOCCLVSISOXCLV 1 times 4 SIMOLV 4 times 1 MISOLV 4 times 4 MIMOLV SISOCCMV SISOXCMV 1 times 3 SIMOMV 3 times 1 MISOMV and 3 times 3MIMOMV capacity CCDF of overhead BPL systems [13 14]


and the 4 times 4MIMO capacity CCDF in the case of overheadLVBPL systems for the aforementioned indicative topologiesare also given Finally in Figure 9(g) the SISOCC capacityCCDF the SISOXC capacity CCDF the 1 times 3 SIMOcapacityCCDF the 3 times 1 MISO capacity CCDF and the 3 times 3

MIMO capacity CCDF in the case of overhead MVBPLsystems for the aforementioned indicative topologies are alsoplotted In Figures 10(a)ndash10(g) the respective plots in the caseof underground LVBPL and MVBPL systems are drawn

From the previous figures several interesting conclusionsmay be drawn concerning the pure scheme configurations

(i) As it concerns SIMO and MISO systemsmdash1 times 4SIMOLV 4 times 1 MISOLV 1 times 3 SIMOMV and3 times 1MISOMV BPL systemsmdashincreasing the num-ber of receive and transmit ports only results in alogarithmic increase in average capacity validatingthe above capacity analysis In the case of full 119899 times 119899MIMOBPL systemsmdashthat is 4 times 4MIMOLV and3 times 3MIMOMVBPL systemsmdashthere is the capacitymultiplication effect due to the log

2sdot presence in

the sum of SVD modal capacities confirming theresults of MIMO capacity analysismdashsee (17) [41 4256] Thus the MIMOBPL systems offer a linear (inmin119899119877 119899119879) increase in channel capacity for no addi-tional power or bandwidth expenditure [39 44ndash46]

(ii) The full 119899 times 119899 MIMOMVBPL systems notablyimprove the average BPL system capacity In allthe cases there is a significant increase of 119899 times 119899

MIMO system capacities compared to the respec-tive SISOCC ones Analytically in overhead BPLsystems examined the 4 times 4 MIMOLV and 3 times

3 MIMOMV cumulative capacities are equal to2943Mbps and 2288Mbps respectively when therespective SISOCC BPL system capacities are equalto 905Mbps and 892Mbps respectively In under-ground BPL systems the maximum 4 times 4 MIMOLV and 3 times 3 MIMOMV capacities are equal to7926Mbps and 2736Mbps respectively when therespective SISOCC BPL system capacities are equalto 2152Mbps and 939Mbps respectively Further-more observing Figures 9(g) and 10(g) the MIMOGC remains almost the same over the frequencyband 3ndash88MHzmdashdenoted as GC inertia [48] Conse-quently the capacities of the 4 times 4MIMOLV and 3 times3 MIMOMV BPL systems regardless of the powergrid type range by approximate GC of 35 and 27respectively in comparison with respective SISOCCsystem capacitiesTheMIMOGC ismost effective forlow IPSDM levels aswell as for channels which exhibithigh ldquoLOSrdquo attenuation and frequency selectivity [53]

(iii) Analyzing the overhead and underground MIMOLVBPL and MIMOMVBPL capacity results of Fig-ures 9(g) and 10(g) in the case of the overhead andunderground full 4 times 4 MIMOLVBPL systemsthe use of three extra degrees of full phasial freedom(simultaneous insertion of three active transmit andthree active receive ports) achieves a GC increasein the approximate range from 212 to 265mdashin

comparison with the respective SISOCC systemsSimilarly in overhead and underground full 3 times 3MIMOMVBPL capacity results despite the differ-ences in channel attenuations IPSDM limits andAWGNPSDs the two extra degrees of full phasialfreedom offer a GC increase ranging approximatelyfrom 148 to 167mdashin comparison with the respectiveSISOCC systems Consequently taking also intoaccount the GC inertia each extra degree of fullphasial freedom offers an approximate frequency-stable capacity increase factor of 08 regardless of thepower grid type and power grid topology

(iv) If CSI is available at transmitting end optimal powerallocation to the subchannels and transmit ports isachieved by applying the water-filling (W-F) algo-rithm [39 129] W-F algorithm permits the furtherexploitation of MIMOBPL capacity

Apart from the pure scheme configurations dependingon BPL system investment budget technoeconomic andsocioeconomic characteristics required system complex-ity power grid type power grid topology IPSDM limitsapplied frequency band assigned and BPL intraoperabil-ityinteroperability differentmultiport implementationsmaybe applied exploiting unequal number of transmit and receiveportsThis case defines the mixed scheme configuration classthat is compared against pure scheme configuration class inthe following analysis [48 50 51]

In Table 4 the minimum value the maximum value themean and the standard deviation of the cumulative capacitiesfor the aforementioned indicative overhead andundergroundLVBPL and MVBPL topologies for various single- andmultiport implementationsmdasheither pure or mixed schemeconfigurationsmdashare presented Moreover the possible num-ber of different circuits (NuI) per each single- or multiportimplementation that has been thoroughly examinedmdashthatis all possible combinations among transmit and receiveports for a given scheme configurationmdashfor each overheadand underground LVBPL and MVBPL topology are alsoreported in Table 4

In Figure 11(a) the quantile-quantile plots of the cumu-lative capacities of the aforementioned overhead LVBPL andMVBPL topologies for a representative set of pure andmixedscheme configurations (SISOCCLV 1 times 4 SIMOLV 4 times 3MIMOLV 4 times 4 MIMOLV SISOCCMV 1 times 3 SIMOMV3times 2MIMOMV and 3times 3MIMOMV) are depicted in orderto test the normality of the cumulative capacity distributionfunctions In these plots the sample quantiles of cumulativecapacities are displayed versus theoretical quantiles fromnormal distribution In Figure 11(b) the cumulative capacityCDF of the considered pure and mixed scheme configura-tions (dashed lines) for the overhead LVBPL and MVBPLtopologies is plotted Eight CDF curves of eight randomvariables normally distributed with mean and standard devi-ation reported in Table 4 (SISOCCLV 1 times 4 SIMOLV 4 times 3MIMOLV 4 times 4 MIMOLV SISOCCMV 1 times 3 SIMOMV3 times 2 MIMOMV and 3 times 3 MIMOMV resp) are alsodisplayed (bold lines) In Figures 11(c) and 11(d) similar plots


0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)

00

1000

2000

2000

3000

4000

4000

5000

6000

6000

7000

8000

8000

Cum

ulat

ive c

apac

ity (M

bps)

0

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)




0 20 40 60 800 20 40 60 80Frequency (MHz)Frequency (MHz)


Capa

city

CCD

F (c

apac

itygt

absc

issa)

01

02

03

04

05

06

07

08

09

1SISOCCLVSISOXCLV1 times 4 SIMOLV4 times 1 MISOLV

3 times 1 MISOMV

4 times 4 MIMOLV

3 times 3 MIMOMV

SISOCCMVSISOXCMV

(a) (b)

(c) (d)

(e) (f)

(g)Capacity (Mbps)



Suburban case



Suburban case

1 times 3 SIMOMV

Figure 10 Capacity characteristics of various pure scheme configurations for the aforementioned indicative underground LVBPL andMVBPL topologies (a) 1 times 4 SIMOLV cumulative capacity (b) 1 times 3 SIMOMV cumulative capacity (c) 4 times 1 MISOLV cumulative capacity(d) 3 times 1 MISOMV cumulative capacity (e) 4 times 4MIMOLV cumulative capacity (f) 3 times 3MIMOMV cumulative capacity (g) SISOCCLVSISOXCLV 1 times 4 SIMOLV 4 times 1 MISOLV 4 times 4 MIMOLV SISOCCMV SISOXCMV 1 times 3 SIMOMV 3 times 1 MISOMV and 3 times 3MIMOMV capacity CCDF of underground BPL systems [13 14]


Table 4 Cumulative capacity and NuI for various overhead and undergroundMIMOLVBPL andMIMOMVBPL systems and single- andmultiport scheme configurations

Cumulative capacity (Mbps)NuIMin Max Mean Standard deviation

OV UN OV UN OV UN OV UNSISOCC

LV 471 1634 904 2152 700 1928 168 200 4MV 459 664 892 1019 688 855 168 137 3

1 times 2 SIMOLV 345 1497 693 1999 530 1785 140 194 24MV 375 471 744 769 572 632 149 114 9

1 times 3 SIMOLV 476 1646 907 2161 704 1938 168 198 16MV 477 664 906 1019 705 855 168 137 3

1 times 4 SIMOLV 479 1655 910 2168 707 1947 168 198 4MV mdash mdash mdash mdash mdash mdash mdash mdash mdash

2 times 1 MISOLV 286 1397 604 1896 453 1684 128 193 24MV 399 613 813 951 616 794 161 1303 9

3 times 1 MISOLV 370 1511 772 2026 579 1804 156 198 16MV 370 583 770 912 579 760 156 127 3

4 times 1 MISOLV 348 1485 739 1998 552 1777 152 198 4MV mdash mdash mdash mdash mdash mdash mdash mdash mdash

2 times 2 MIMOLV 537 2770 1114 3779 854 3349 241 390 36MV 602 828 1261 1381 949 1129 264 212 9

2 times 3 MIMOLV 536 2770 1137 3779 851 3349 240 390 24MV 736 1166 1538 1825 1155 1520 312 253 3

2 times 4 MIMOLV 694 2957 1479 3987 1102 3542 305 397 6MV mdash mdash mdash mdash mdash mdash mdash mdash mdash

3 times 2 MIMOLV 530 2770 1135 3779 847 3349 243 390 24MV 732 1166 1535 1824 1151 1520 312 253 3

3 times 3 MIMOLV 774 3058 1591 4092 1201 3645 317 398 16MV 1082 1750 2288 2736 1710 2280 468 380 1



Table 4 Continued


OV UN OV UN OV UN OV UN4 times 2 MIMO

LV 540 2770 1147 3779 859 3349 243 390 6MV mdash mdash mdash mdash mdash mdash mdash mdash mdash




are drawn in the case of underground MIMOLVBPL andMIMOMVBPL systems

56 GC Analysis To investigate further the GC potential ofvarious single- and multiport implementations all possiblepure and mixed scheme configurations are studied andcompared by means of GC assuming the median value ofthe aforementioned indicative topologies for each overheadand undergroundMIMOLVBPL andMIMOMVBPL sys-tem In Figures 12(a) and 12(b) the GC is contour plottedwith respect to the active transmit and receive ports forthe overhead MIMOLVBPL and MIMOMVBPL systemsrespectively This contour plot displays GC isolines of theoccurring pure and mixed scheme configurations In Figures12(c) and 12(d) similar plots are displayed in the case ofunderground LVBPL and MVBPL systems respectively

From Table 4 Figures 11(a)ndash11(d) and Figures 12(a)ndash12(d) several interesting conclusions may be deduced

(i) The cumulative capacities of pure and mixed schemeconfigurations of overhead and underground LVBPLand MVBPL systems follow normal distributionssince the simulation plots are close to linear onesmdashseebold lines in Figures 11(a) and 11(c)mdashregardless of thescheme configuration and the power grid topologyconsidered [4 71ndash73 78] Meanwhile similar tothe cases of ACG and RMS-DS suitable lognormaldistributions for each overhead and undergroundLVBPL and MVBPL pure and mixed scheme con-figurations have been derived by appropriately fittingthe simulation results In all the cases the proposedlognormal distributions are in excellent agreementwith simulation results providing an evaluation toolof the imminent statistical channel models [71ndash79]

(ii) Compared to the respective SISOCC capacities GCof pure and mixed scheme configurations is consid-ered to be dependent on two components the arraygain at the receiving end that corresponds to the gainin the average power of the signal combination onreceive ports and the diversity gain that corresponds

to the gain from increasing the system dimensionality(rank ofH+sdot) It should be noted that diversity gainmainly depends on phasial correlation between sig-nals which are expressed by the correlation between119867+

119894119895 119894 isin 119873119879 119895 isin 119873119877 [130ndash133]

(iii) In the majority of pure and mixed scheme config-urations examined there is a significant increase ofthe capacity compared to SISOCC configurationsHowever apart from the factors that have alreadybeen mentioned GC varies for different transmitand receive port combinations when a scheme con-figuration is given depending on the path specificattenuationmdashthat is the number of CCs and XCsincluded in the givenMIMO scheme configurationmdashand the correlation between the different paths Con-sequently when NuI of a scheme configuration isdominated by XCsmdashfor example 1 times 2 SIMO 1 times 3SIMO 1 times 4 SIMO and 2 times 1 MISO scheme con-figurations for overhead and underground LVBPLsystems and 1 times 2 SIMO 1times3 SIMO and 2times1MISOscheme configurations for overhead and under-ground MVBPL systemsmdashthe cumulative capacityand GC of these multiport implementations are lowercompared to the respective values of SISOCC onesContrary to the above the remaining scheme config-urations are less affected primarily due to the extradegrees ofmdasheither full or partialmdashphasial freedomand secondarily due to the vast presence of CCsduring their implementations

(iv) The effect of transmit and receive port increase ofSIMO MISO and MIMO systems on the capacityis concretely highlighted in Figures 11(a)ndash11(d) andFigures 12(a)ndash12(d) As expected in the cases of equaldiversity order due to the array gain the increase ofreceive ports with constant transmit ports is moreefficient regarding the resulting increase in capacitythan the increase in transmit ports with constantreceive ports Anyway for a given equal diversityorder the full exploitation of MIMO technology


0

500

1000

2000

3000

1500

2500Cu

mul

ativ

e cap

acity

(Mbp

s)


1 times 4 SIMO

4 times 4 MIMO

4 times 3 MIMO3 times 3 MIMO

3 times 2 MIMO

1 times 3 SIMO

SISOCCLV SISOCCMV

(a)

1

08

09

06

07

04

05

03

01

02

00 1000 2000 3000

Cumulative capacity (Mbps)

SISOCCLV

SISOCCMV

Cum

ulat

ive c

apac

ity C

DF

(cum

ulat

ive c

apac

itylt

absc

issa)

4 times 4 MIMO

4 times 3 MIMO

1 times 3 SIMO

1 times 4 SIMO

3 times 2 MIMO

3 times 3 MIMO

(b)

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)

Standard normal quantiles0 1

SISOCCLV

SISOCCMVminus1

4 times 4 MIMO

4 times 3 MIMO

3 times 3 MIMO

1 times 3 SIMO

1 times 4 SIMO

3 times 2 MIMO


(c)

0 2000 4000 6000 8000Cumulative capacity (Mbps)

08

09

1

06

07

04

05

03

01

02

0

SISOCCLV

SISOCCMV

Cum

ulat

ive c

apac

ity C

DF

(cum

ulat

ive c

apac

itylt

absc

issa)

3 times 3 MIMO

3 times 2 MIMO

1 times 4 SIMO

1 times 3 SIMO

4 times 3 MIMO

4 times 4 MIMO

LVBPL casesMVBPL casesLVBPL topologiesMVBPL topologies

(d)

Figure 11 (a) Quantile-quantile plots of cumulative capacities of overhead LVBPL (+) and MVBPL (+) systems compared to theoreticalnormal distribution lines of the respective overhead LVBPL and MVBPL cases for representative pure and mixed scheme configurations(b) Cumulative capacity CDF curves of overhead LVBPL and MVBPL topologies compared to theoretical normal distribution lines of therespective overhead LVBPL and MVBPL pure and mixed scheme configurations (c) (d) Similar plots in underground BPL cases


11

1

1

11

2

2

2

3

3

4

4

079083064

12

12

119

312

171

15

25

171

121

101 165157

Rece

ive p

orts

(nR

)

Transmit ports (nT)

075

(a)

111 1 1

1

2

2

2

3

3

Transmit ports (nT)

12

12

14

14

16

22

18

167

102 248

082

089 084

136

167Re

ceiv

e por

ts (n

R)

(b)

11

1

2

3

4 365101

101

187184

174

174 174 174

189

093

088 092094

189

1

1

15

15

252 3

2 3 4Transmit ports (nT)

Rece

ive p

orts

(nR

)

(c)


1 1

1

1

2

1

2

3

178

178 267

089

132

093

074

15

Rece

ive p

orts

(nR

)

(d)

Figure 12 GC contour plot for various pure and mixed scheme configurations (a) Overhead LVBPL (b) Overhead MVBPL(c) Underground LVBPL (d) Underground MVBPL

potential is achieved when transmit and receive portsare equalmdashthat is via the full MIMO scheme config-uration deploymentmdashvalidating the aforementionedapproximate GC increase by 08 per each extra degreeof full phasial freedom regardless of the power gridtype and power grid topology However there is asignificant trade-off that occurs between capacity and

system implementation complexity [28 43 46 130134]

(v) If CSI is available at transmitting end except for thealready mentioned optimal power allocation throughW-F algorithm [39 129 130] from the point of viewof fading reduction 119899119879 times 119899119877 and 119899119877 times 119899119879 MIMOBPL configurations present similar spectral behavior


Therefore the limit of MIMO capacity tends to be thesame for 119899119879 times 119899119877 and 119899119877 times 119899119879MIMOBPL configura-tions [130]

(vi) Through the aforementioned significant boost ofchannel capacity viaMIMO technology many broad-band technologies may rely on mainly undergroundMIMOLVBPL networked information systems so asto deliver a variety of modern broadband servicessuch as high-speed internet digital and HD televi-sion and digital phone Consequently the overarch-ing theme of this backbone emersion is a creationof an urban interconnected information network thatpermits the further urban backbone convergencebetween BPL and other well-established backbonecommunications technologies (eg FTTB technol-ogy) [28 135]

(vii) Concluding this review paper ofMIMOBPL technol-ogy it should be noted that until now due to differentinterests regarding BPL technologiesmdasheither indooror overhead or underground either LV or MV orHVmdashthe BPL system implementations and standard-ization pursue parallel paths to their developmentThe varying paces of BPL technology implementationhave resulted in a polyglot of capabilities that whiledifferent must still coexist among them and withother already established technologies [38 128 136]This coexistence potential becomes requisite as immi-nent SG will be supported by a heterogeneous set ofnetworking technologies Since no single solution fitsall scenarios overhead and underground LVBPL andMVBPL systems need to work in a compatible way(intraoperate) before BPL technology interoperateswith other broadband technologies such as wired(eg fiber and DSL) and wireless (eg Wi-Fi andWiMAX) Compatible frequencies equipment andsignaling green technology aspects adequate IPSDMlevels area coverage scalable capacityMIMOschemeconfigurations and evolution of coexistence stan-dards must be ensured taking the specific featuresof overhead and underground LVBPL and MVBPLsystems into account [13ndash19 38 137ndash140]

6 Conclusions

This reviewpaper has focused on the broadband transmissioncharacteristics the statistical performance metrics and thecapacity measures of overhead and underground MIMOLVBPL and MIMOMVBPL distribution power grids

As it concerns the broadband transmission characteristicsof overhead and underground LVBPL and MVBPL dis-tribution networks the well-known hybrid model has beenextended through UVD modal analysis expansion pack inorder to better cover the new MIMO features The broad-band transmission capabilities of such networks depend onthe frequency the power grid type the MIMO schemeconfiguration physical properties of the MTL configurationused the end-to-endmdashldquoLOSrdquomdashdistance and the number the

electrical length and the terminations of the branches alongthe end-to-end BPL signal propagation

The investigation of the statistical performance met-rics of overhead and underground MIMOLVBPL andMIMOMVBPL channels has revealed several fundamen-tal properties that characterize various wireline channelsnamely (i) the lognormal distribution of ACG RMS-DSand cumulative capacity (ii) the negative correlation betweenRMS-DS and ACG and (iii) the hyperbolic correlationbetween RMS-DS and CBMoreover the existing portfolio ofregression approaches has been enriched by the insertion ofmany regression approaches suitable for overhead and under-ground MIMOLVBPL and MIMOMVBPL channels inthe fields of (a) RMS-DSACG correlation (b) RMS-DSCBcorrelation and (c) cumulative capacity versus various single-and multiport scheme configurations

Finally based on the capacity metrics the incredi-ble capacity characteristics of overhead and undergroundMIMOLVBPL and MIMOMVBPL systems have beenunderlined More specifically depending on BPL systeminvestment budget technoeconomic and socioeconomiccharacteristics required system complexity power grid typepower grid topology IPSDM limits applied frequency bandassigned and BPL intraoperabilityinteroperability potentialdifferent pure and mixed MIMO scheme configurations canadaptively be applied improving the todayrsquos best SISOBPLsystem capacities by a factor of up to 365 This permitsthe deployment of BPL system capacities up to 8Gbps Thiscapacity breakthrough transforms the existing overhead andunderground LVBPL andMVBPL distribution networks toa quasi-fiber-optic backbone network an alternative to theFTTB technology These capacity characteristics determinethe deployment of modern broadband applications theperformance of various SG applications and the BPL systemconvergence in the oncoming SG network

References

[1] T A Papadopoulos C G Kaloudas A I Chrysochos and GK Papagiannis ldquoApplication of narrowband power-line com-munication in medium-voltage smart distribution gridsrdquo IEEETransactions on Power Delivery vol 28 no 2 pp 981ndash988 2013

[2] A G Lazaropoulos ldquoReview and progress towards the capacityboost of overhead and underground medium-voltage and low-voltage broadbandover power lines networks cooperative com-munications through two- and three-hop repeater systemsrdquoISRN Electronics vol 2013 Article ID 472190 19 pages 2013

[3] A G Lazaropoulos ldquoReview and progress towards the commonbroadband management of high-voltage transmission gridsmodel expansion and comparative modal analysisrdquo ISRN Elec-tronics vol 2012 Article ID 935286 18 pages 2012

[4] S Galli A Scaglione and Z Wang ldquoFor the grid and throughthe grid the role of power line communications in the smartgridrdquo Proceedings of the IEEE vol 99 no 6 pp 998ndash1027 2011

[5] A G Lazaropoulos ldquoTowards modal integration of overheadand underground low-voltage and medium-voltage power linecommunication channels in the smart grid landscape modelexpansion broadband signal transmission characteristics andstatistical performance metrics (Invited Paper)rdquo ISRN SignalProcessing vol 2012 Article ID 121628 17 pages 2012


[6] A G Lazaropoulos ldquoDeployment concepts for overhead highvoltage broadband over power Lines connections with two-hoprepeater system capacity countermeasures against aggravatedtopologies and high noise environmentsrdquo Progress in Electro-magnetics Research B vol 44 pp 283ndash307 2012

[7] N Pavlidou A J Han Vinck J Yazdani and B Honary ldquoPowerline communications state of the art and future trendsrdquo IEEECommunications Magazine vol 41 no 4 pp 34ndash40 2003

[8] M Gebhardt F Weinmann and K Dostert ldquoPhysical and reg-ulatory constraints for communication over the power supplygridrdquo IEEE Communications Magazine vol 41 no 5 pp 84ndash902003

[9] A G Lazaropoulos ldquoFactors influencing broadband transmis-sion characteristics of underground low-voltage distributionnetworksrdquo IET Communications vol 6 no 17 pp 2886ndash28932012

[10] A G Lazaropoulos ldquoTowards broadband over power lines sys-tems integration transmission characteristics of undergroundlow-voltage distribution power linesrdquo Progress in Electromag-netics Research B no 39 pp 89ndash114 2012

[11] G Jee R D Rao and Y Cern ldquoDemonstration of the technicalviability of PLC systems on medium- and low-voltage lines inthe United Statesrdquo IEEE Communications Magazine vol 41 no5 pp 108ndash112 2003

[12] A G Lazaropoulos ldquoBroadband transmission and statisticalperformance properties of overhead high-voltage transmissionnetworksrdquo Journal of Computer Networks and Communicationsvol 2012 Article ID 875632 16 pages 2012

[13] A G Lazaropoulos ldquoGreen overhead and undergroundmultiple-input multiple-output medium voltage broadbandover power lines networks energy-efficient power controlrdquo inJournal of Global Optimization vol 2012 pp 1ndash28 2012

[14] A G Lazaropoulos ldquoOverhead and underground MIMO lowvoltage broadband over power lines networks and EMI regula-tions towards greener capacity performancesrdquo Computers andElectrical Engineering 2013

[15] A G Lazaropoulos and P G Cottis ldquoTransmission character-istics of overhead medium-voltage power-line communicationchannelsrdquo IEEE Transactions on Power Delivery vol 24 no 3pp 1164ndash1173 2009

[16] A G Lazaropoulos and P G Cottis ldquoCapacity of overheadmedium voltage power line communication channelsrdquo IEEETransactions on Power Delivery vol 25 no 2 pp 723ndash733 2010

[17] A G Lazaropoulos and P G Cottis ldquoBroadband transmissionvia undergroundmedium-voltage power lines Part I transmis-sion characteristicsrdquo IEEE Transactions on Power Delivery vol25 no 4 pp 2414ndash2424 2010

[18] A G Lazaropoulos and P G Cottis ldquoBroadband transmissionvia undergroundmedium-voltage power lines Part II capacityrdquoIEEE Transactions on Power Delivery vol 25 no 4 pp 2425ndash2434 2010

[19] A G Lazaropoulos ldquoBroadband transmission characteristicsof overhead high-voltage power line communication channelsrdquoProgress In Electromagnetics Research B no 36 pp 373ndash3982011

[20] P S Henry ldquoInterference characteristics of broadband powerline communication systems using aerial medium voltagewiresrdquo IEEE Communications Magazine vol 43 no 4 pp 92ndash98 2005

[21] S Liu and L J Greenstein ldquoEmission characteristics and inter-ference constraint of overhead medium-voltage Broadband

Power Line (BPL) systemsrdquo in Proceedings of the IEEE GlobalTelecommunications Conference (GLOBECOM rsquo08) pp 2921ndash2925 New Orleans La USA December 2008

[22] D Fenton and P Brown ldquoModelling cumulative high frequencyradiated interference frompower line communication systemsrdquoin Proceedings of the IEEE International Conference on PowerLine Communications and Its Applications Athens GreeceMarch 2002

[23] M Gotz M Rapp and K Dostert ldquoPower line channel char-acteristics and their effect on communication system designrdquoIEEECommunicationsMagazine vol 42 no 4 pp 78ndash86 2004

[24] M Zimmermann and K Dostert ldquoA multipath model for thepowerline channelrdquo IEEE Transactions on Communications vol50 no 4 pp 553ndash559 2002

[25] L Lampe R Schober and S Yiu ldquoDistributed space-timecoding formultihop transmission in power line communicationnetworksrdquo IEEE Journal on Selected Areas in Communicationsvol 24 no 7 pp 1389ndash1400 2006

[26] A S Ibrahim A K Sadek SWeifeng andK J R Liu ldquoCooper-ative communications with relay-selection when to cooperateand whom to cooperate withrdquo IEEE Transactions on WirelessCommunications vol 7 no 7 pp 2814ndash2827 2008

[27] A Nosratinia T E Hunter and A Hedayat ldquoCooperativecommunication in wireless networksrdquo IEEE CommunicationsMagazine vol 42 no 10 pp 74ndash80 2004

[28] L Wang and L Hanzo ldquoDispensing with channel estimationdifferentiallymodulated cooperativewireless communicationsrdquoIEEE Communications Surveys and Tutorials vol 14 no 3 pp836ndash857 2012

[29] Y-W Hong W-J Huang F-H Chiu and C-C J KuoldquoCooperative communications in resource-constrainedwirelessnetworksrdquo IEEE Signal Processing Magazine vol 24 no 3 pp47ndash57 2007

[30] M Gastpar ldquoOn capacity under receive and spatial spectrum-sharing constraintsrdquo IEEE Transactions on Information Theoryvol 53 no 2 pp 471ndash487 2007

[31] F Versolatto and A M Tonello ldquoAn MTL theory approach forthe simulation ofMIMO power-line communication channelsrdquoIEEE Transactions on Power Delivery vol 26 no 3 pp 1710ndash1717 2011

[32] V Oksman and S Galli ldquoGhn the new ITU-T home network-ing standardrdquo IEEE Communications Magazine vol 47 no 10pp 138ndash145 2009

[33] D Schneider J Speidel L Stadelmeier andD Schill ldquoPrecodedspatial multiplexing MIMO for inhome power line Communi-cationsrdquo in Proceedings of the IEEE Global TelecommunicationsConference (GLOBECOM rsquo08) pp 2901ndash2905 NewOrleans LaUSA December 2008

[34] ldquoOPERA2 D24 document of conclusions about the differentservices tested over the new generation PLC network (M24definitive version)rdquo IST Integrated Project 026920 2008

[35] N Golmie N Chevrollier and O Rebala ldquoBluetooth andWLAN coexistence challenges and solutionsrdquo IEEE WirelessCommunications Magazine vol 10 no 6 pp 347ndash357 2003

[36] ldquoOPERA2 D51 white Paper OPERA technology (final ver-sion)rdquo IST Integrated Project 026920 2008

[37] M Kuhn S Berger I Hammerstrom and A WittnebenldquoPower line enhanced cooperative wireless communicationsrdquoIEEE Journal on Selected Areas in Communications vol 24 no7 pp 1401ndash1410 2006


[38] L Hanzo M El-Hajjar and O Alamri ldquoNear-capacity wirelesstransceivers and cooperative communications in the MIMOEra evolution of standards waveform design and futureperspectivesrdquo Proceedings of the IEEE vol 99 no 8 pp 1343ndash1385 2011

[39] A J Paulraj D A Gore R U Nabar and H Bolcskei ldquoAnoverview of MIMO communications a key to gigabit wirelessrdquoProceedings of the IEEE vol 92 no 2 pp 198ndash217 2004

[40] E Biglieri J Proakis and S Shamai ldquoFading channels informa-tion-theoretic and communications aspectsrdquo IEEE Transactionson Information Theory vol 44 no 6 pp 2619ndash2692 1998

[41] D Gesbert M Shafi D-S Shiu P J Smith and A NaguibldquoFrom theory to practice an overview of MIMO space-timecoded wireless systemsrdquo IEEE Journal on Selected Areas inCommunications vol 21 no 3 pp 281ndash302 2003

[42] A Goldsmith S A Jafar N Jindal and S Vishwanath ldquoCapac-ity limits of MIMO channelsrdquo IEEE Journal on Selected Areas inCommunications vol 21 no 5 pp 684ndash702 2003

[43] Q Du and X Zhang ldquoQoS-aware base-station selections fordistributedMIMO links in broadband wireless networksrdquo IEEEJournal on Selected Areas in Communications vol 29 no 6 pp1123ndash1138 2011

[44] S Vishwanath N Jindal and A Goldsmith ldquoDuality achiev-able rates and sum-rate capacity of Gaussian MIMO broadcastchannelsrdquo IEEE Transactions on InformationTheory vol 49 no10 pp 2658ndash2668 2003

[45] G Caire and S Shamai ldquoOn the achievable throughput of amultiantenna Gaussian broadcast channelrdquo IEEE Transactionson Information Theory vol 49 no 7 pp 1691ndash1706 2003

[46] S Sugiura S Chen and L Hanzo ldquoMIMO-aided near-capacityturbo transceivers taxonomy andperformance versus complex-ityrdquo IEEE Communications Surveys and Tutorials vol 14 no 2pp 421ndash442 2012

[47] M Tlich A Zeddam F Moulin and F Gauthier ldquoIndoorpower-line communications channel characterization up to 100MHz Part I one-parameter deterministic modelrdquo IEEE Trans-actions on Power Delivery vol 23 no 3 pp 1392ndash1401 2008

[48] R Hashmat P Pagani A Zeddam and T Chonave ldquoMIMOcommunications for inhome PLC Networks measurementsand results up to 100 MHzrdquo in Proceedings of the 14th AnnualInternational Symposium on Power Line Communications andits Applications (IEEE ISPLC rsquo10) pp 120ndash124 Rio de JaneiroBrazil March 2010

[49] A Canova N Benvenuto and P Bisaglia ldquoReceivers forMIMO-PLC channels throughput comparisonrdquo in Proceedingsof the 14th Annual International Symposium on Power LineCommunications and its Applications (IEEE ISPLC rsquo10) pp 114ndash119 Rio de Janeiro Brazil March 2010

[50] D Schneider A Schwager J Speidel and A Dilly ldquoImple-mentation and results of a MIMO PLC feasibility studyrdquo inProceedings of the IEEE International Symposium on Power LineCommunications and Its Applications (ISPLC rsquo11) pp 54ndash59Udine Italy April 2011

[51] M Biagi ldquoMIMO self-interference mitigation effects on PLCrelay networksrdquo in Proceedings of the IEEE International Sym-posium on Power Line Communications and Its Applications(ISPLC rsquo11) pp 182ndash186 Udine Italy April 2011

[52] M Biagi ldquoMIMO self-interference mitigation effects on powerline relay networksrdquo IEEE Communications Letters vol 15 no8 pp 866ndash868 2011

[53] D Schneider J Speidel L Stadelmeier D Schill and ASchwager ldquoPotential of MIMO for inhome power line commu-nicationsrdquo in Proceedings of the the ITGFachtagung DortmundGermany March 2009

[54] A Schwager D Schneider W Baschlin A Dilly and J SpeidelldquoMIMO PLC theory measurements and system setuprdquo inProceedings of the IEEE International Symposium on Power LineCommunications and Its Applications (ISPLC rsquo11) pp 48ndash53Udine Italy April 2011

[55] L Stadelmeier D Schneider D Schill A Schwager and JSpeidel ldquoMIMO for inhome power line communicationsrdquoin Proceedings of the International Conference on Source andChannel Coding Ulm Germany January 2008

[56] F Versolatto andAM Tonello ldquoAMIMOPLC random channelgenerator and capacity analysisrdquo in Proceedings of the IEEEInternational Symposium on Power Line Communications andIts Applications (ISPLC rsquo11) pp 66ndash71 Udine Italy April 2011

[57] PAmirshahi andMKavehrad ldquoHigh-frequency characteristicsof overhead multiconductor power lines for broadband com-municationsrdquo IEEE Journal on Selected Areas in Communica-tions vol 24 no 7 pp 1292ndash1302 2006

[58] T SartenaerMultiuser communications over frequency selectivewired channels and applications to the powerline access network[PhD thesis] Universite catholique de Louvain Louvain-la-Neuve Belgium 2004

[59] T Calliacoudas and F Issa ldquoMulticonductor transmission linesand cables solver an efficient simulation tool for plc channelnetworks developmentrdquo in Proceedings of the IEEE InternationalConference on Power Line Communications and Its ApplicationsAthens Greece March 2002

[60] C R PaulAnalysis ofMulticonductor Transmission LinesWileyNew York NY USA 1994

[61] J A B Faria Multiconductor Transmission-Line StructuresModal Analysis Techniques Wiley New York NY USA 1994

[62] T Sartenaer and P Delogne ldquoDeterministic modeling of the(shielded) outdoor power line channel based on the Multicon-ductor Transmission Line equationsrdquo IEEE Journal on SelectedAreas in Communications vol 24 no 7 pp 1277ndash1290 2006

[63] T Sartenaer and P Delogne ldquoPowerline cables modellingfor broadband communicationsrdquo in Proceedings of the IEEEInternational Conference on the Power Line Communicationsand Its Applications pp 331ndash337 Malmo Sweden April 2001

[64] S Galli and T C Banwell ldquoA deterministic frequency-domainmodel for the indoor power line transfer functionrdquo IEEE Journalon Selected Areas in Communications vol 24 no 7 pp 1304ndash1315 2006

[65] A Perez A M Sanchez J R Regue et al ldquoCircuital and modalcharacterization of the power-line network in the PLC bandrdquoIEEE Transactions on Power Delivery vol 24 no 3 pp 1182ndash1189 2009

[66] K Dostert Powerline Communications Prentice-Hall UpperSaddle River NJ USA 2001

[67] P Amirshahi Broadband access and home networking throughpowerline networks [PhD thesis] Pennsylvania StateUniversityUniversity Park Pa USA 2006

[68] H Meng S Chen Y L Guan et al ldquoModeling of transfercharacteristics for the broadband power line communicationchannelrdquo IEEE Transactions on Power Delivery vol 19 no 3 pp1057ndash1064 2004

[69] S Barmada A Musolino and M Raugi ldquoInnovative modelfor time-varying power line communication channel response


evaluationrdquo IEEE Journal on Selected Areas in Communicationsvol 24 no 7 pp 1317ndash1325 2006

[70] M DrsquoAmore and M S Sarto ldquoA new formulation of lossyground return parameters for transient analysis of multicon-ductor dissipative linesrdquo IEEE Transactions on Power Deliveryvol 12 no 1 pp 303ndash309 1997

[71] S Galli ldquoA novel approach to the statisticalmodeling of wirelinechannelsrdquo IEEE Transactions on Communications vol 59 no 5pp 1332ndash1345 2011

[72] S Galli ldquoA simplifiedmodel for the indoor power line channelrdquoin Proceedings of the IEEE International Symposium on PowerLine Communications and its Applications (ISPLC rsquo09) pp 13ndash19 Dresden Germany April 2009

[73] S Galli ldquoA simple two-tap statistical model for the power linechannelrdquo in Proceedings of the 14th Annual International Sympo-sium on Power Line Communications and its Applications (IEEEISPLC rsquo10) pp 242ndash248 Rio de Janeiro Brazil March 2010

[74] ldquoITU-T SG15Q4 Powerline channel datardquo ContributionNIPP-NAI-2007-107R1 2007

[75] B OrsquoMahony ldquoField testing of high-speed power line commu-nications in North American homesrdquo in Proceedings of the IEEEInternational SymposiumonPower LineCommunications and ItsApplications (ISPLC rsquo06) pp 155ndash159 Orlando Fla USAMarch2006

[76] F Versolatto and A M Tonello ldquoAnalysis of the PLC channelstatistics using a bottom-up random simulatorrdquo in Proceedingsof the 14th Annual International Symposium on Power LineCommunications and its Applications (IEEE ISPLC rsquo10) pp 236ndash241 Rio de Janeiro Brazil March 2010

[77] A M Tonello F Versolatto and C Tornelli ldquoAnalysis of impul-sive UWB modulation on a real MV test networkrdquo in Proceed-ings of the IEEE International Symposium on Power Line Com-munications and Its Applications (ISPLC rsquo11) pp 18ndash23 UdineItaly April 2011

[78] M Antoniali A M Tonello M Lenardon and A QualizzaldquoMeasurements and analysis of PLC channels in a cruise shiprdquoin Proceedings of the IEEE International Symposium on PowerLine Communications and Its Applications (ISPLC rsquo11) pp 102ndash107 Udine Italy April 2011

[79] A M Tonello and F Versolatto ldquoBottom-up statistical PLCchannel modeling part II inferring the statisticsrdquo IEEE Trans-actions on Power Delivery vol 25 no 4 pp 2356ndash2363 2010

[80] F J Canete J A Cortes L Dıez and J T EntrambasaguasldquoA channel model proposal for indoor power line communica-tionsrdquo IEEE Communications Magazine vol 49 no 12 pp 166ndash174

[81] T S Rappaport Wireless Communications Prentice-Hall Up-per Saddle River NJ USA 2002

[82] M Tlich G Avril and A Zeddam ldquoCoherence bandwidth andits relationship with the rms delay spread for PLC channelsusingmeasurements up to 100MHzrdquo inHomeNetworking IFIPInternational Federation for Information Processing pp 129ndash142 2007

[83] ldquoDLC+VIT4IP D1 2 overall system architecture design DLCsystem architecturerdquo FP7 Integrated Project 247750 2010

[84] F Issa D Chaffanjon E P de la Bathie and A Pacaud ldquoAnefficient tool for modal analysis transmission lines for PLCnetworks developmentrdquo in Proceedings of the IEEE InternationalConference on Power Line Communications and Its ApplicationsAthens Greece March 2002

[85] J Anatory and N Theethayi ldquoOn the efficacy of usingground return in the broadband power-line communications atransmission-line analysisrdquo IEEE Transactions on Power Deliv-ery vol 23 no 1 pp 132ndash139 2008

[86] T A Papadopoulos B D Batalas A Radis and G K Papa-giannis ldquoMedium voltage network PLC modeling and signalpropagation analysisrdquo in Proceedings of the IEEE InternationalSymposium on Power Line Communications and Its Applications(ISPLC rsquo07) pp 284ndash289 Pisa Italy March 2007

[87] ldquoOPERA1D44 report presenting the architecture of plc systemthe electricity network topologies the operating modes and theequipment over which PLC access system will be installedrdquo ISTIntegrated Project 507667 2005

[88] M Tang and M Zhai ldquoResearch of transmission parametersof four-conductor cables for power line communicationrdquo inProceedings of the International Conference on Computer Scienceand Software Engineering (CSSE rsquo08) pp 1306ndash1309 WuhanChina December 2008

[89] MDrsquoAmore andM S Sarto ldquoSimulationmodels of a dissipativetransmission line above a lossy ground for a wide-frequencyrange Part I single conductor configurationrdquo IEEE Transac-tions on Electromagnetic Compatibility vol 38 no 2 pp 127ndash138 1996

[90] M D rsquoAmore and M S Sarto ldquoSimulation models of adissipative transmission line above a lossy ground for a wide-frequency range Part II multi-conductor configurationrdquo IEEETransactions on Electromagnetic Compatibility vol 38 no 2 pp139ndash149 1996

[91] P C J M van der Wielen On-line detection and location ofpartial discharges in medium-voltage power cables [PhD thesis]Eindhoven University of Technology Eindhoven the Nether-lands 2005

[92] ldquoOPERA1 D5 pathloss as a function of frequency distance andnetwork topology for various LV and MV European powerlinenetworksrdquo IST Integrated Project 507667 2005

[93] P C J M van derWielen E F Steennis and P A A FWoutersldquoFundamental aspects of excitation and propagation of on-line partial discharge signals in three-phase medium voltagecable systemsrdquo IEEE Transactions on Dielectrics and ElectricalInsulation vol 10 no 4 pp 678ndash688 2003

[94] J Veen On-line signal analysis of partial discharges in medium-voltage power cables [PhD thesis] Technische UniversiteitEindhoven Eindhoven the Netherlands 2005

[95] NTheethayi Electromagnetic interference in distributed outdoorelectrical systems with an emphasis on lightning interaction withelectrified railway network [PhD thesis] Uppsala UniversityUppsala Sweden 2005

[96] E F Vance Coupling To Cable Shields Wiley New York NYUSA 1978

[97] R Aquilue Power line communications for the electrical util-ity physical layer design and channel modeling [PhD thesis]Universitat Ramon Llull Enginyeria I Arquitectura La SalleBarcelona Spain 2008

[98] A Cataliotti A Daidone andG Tine ldquoPower line communica-tion inmediumvoltage systems characterization ofMVcablesrdquoIEEE Transactions on Power Delivery vol 23 no 4 pp 1896ndash1902 2008

[99] J Anatory NTheethayi RThottappillil M M Kissaka and NH Mvungi ldquoThe influence of load impedance line length andbranches on underground cable power-line communications(PLC) systemsrdquo IEEE Transactions on Power Delivery vol 23no 1 pp 180ndash187 2008


[100] D A Tsiamitros G K Papagiannis and P S DokopoulosldquoEarth return impedances of conductor arrangements in mul-tilayer soils Part I theoretical modelrdquo IEEE Transactions onPower Delivery vol 23 no 4 pp 2392ndash2400 2008

[101] D A Tsiamitros G K Papagiannis and P S DokopoulosldquoEarth return impedances of conductor arrangements in mul-tilayer soils Part II numerical resultsrdquo IEEE Transactions onPower Delivery vol 23 no 4 pp 2401ndash2408 2008

[102] F Rachidi and S V Tkachenko Electromagnetic Field Inter-action with Transmission Lines From Classical Theory to HFRadiation Effects WIT press Southampton UK 2008

[103] J R Carson ldquoWave propagation in overhead wires with groundreturnrdquo Bell System Technical Journal vol 5 pp 539ndash554 1926

[104] H Kikuchi ldquoWave propagation along an infinite wire aboveground at high frequenciesrdquo Electrotechnical Journal of Japanvol 2 pp 73ndash78 1956

[105] H Kikuchi ldquoOn the transition form a ground return circuit to asurface waveguiderdquo in Proceedings of the Congress on UltrahighFrequency Circuits Antennas pp 39ndash45 Paris France October1957

[106] S Galli and T Banwell ldquoA novel approach to the modeling ofthe indoor power line channel Part II transfer function and itspropertiesrdquo IEEE Transactions on Power Delivery vol 20 no 3pp 1869ndash1878 2005

[107] B S Yarman and A Fettweis ldquoComputer-aided double match-ing via parametric representation of Brune functionsrdquo IEEETransactions on Circuits and Systems vol 37 no 2 pp 212ndash2221990

[108] R Araneo S Celozzi G Lovat and F Maradei ldquoMulti-portimpedance matching technique for power line communica-tionsrdquo in Proceedings of the IEEE International Symposium onPower Line Communications and Its Applications (ISPLC rsquo11) pp96ndash101 Udine Italy April 2011

[109] L H-J Lampe and J B Huber ldquoBandwidth efficient power linecommunications based on OFDMrdquo AEU-Archiv fur Elektronikund Ubertragungstechnik vol 54 no 1 pp 2ndash12 2000

[110] M Crussiere J Y Baudais and J F Helard ldquoAdaptive spreadspectrum multicarrier multiple-access over wirelinesrdquo IEEEJournal on Selected Areas in Communications vol 24 no 7 pp1377ndash1388 2006

[111] P Achaichia M Le Bot and P Siohan ldquoOFDMOQAM asolution to efficiently increase the capacity of future PLCnetworksrdquo IEEE Transactions on Power Delivery vol 26 no 4pp 2443ndash2455 2011

[112] Y-H Kim and J-H Lee ldquoComparison of passband andbaseband transmission schemes for power-line communicationOFDM systemsrdquo IEEE Transactions on Power Delivery vol 26no 4 pp 2466ndash2475 2011

[113] J Anatory N Theethayi R Thottappillil M Kissaka andN Mvungi ldquoThe effects of load impedance line length andbranches in typical low-voltage channels of the BPLC systems ofdeveloping countries transmission-line analysesrdquo IEEE Trans-actions on Power Delivery vol 24 no 2 pp 621ndash629 2009

[114] J Anatory N Theethayi and R Thottappillil ldquoPower-linecommunication channel model for interconnected networksPart II multiconductor systemrdquo IEEE Transactions on PowerDelivery vol 24 no 1 pp 124ndash128 2009

[115] J Song C Pan Q Wu et al ldquoField trial of digital video trans-mission over medium-voltage powerline with time-domainsynchronous orthogonal frequency division multiplexing tech-nologyrdquo in Proceedings of the International Symposium on Power

Line Communications and Its Applications (ISPLC rsquo07) pp 559ndash564 Pisa Italy March 2007

[116] R Aquilue M Ribo J R Regue J L Pijoan and G SanchezldquoScattering parameters-based channel characterization andmodeling for underground medium-voltage power-line com-municationsrdquo IEEE Transactions on Power Delivery vol 24 no3 pp 1122ndash1131 2009

[117] H Liu J Song B Zhao and X Li ldquoChannel study for medium-voltage power networkrdquo in Proceedings of the IEEE InternationalSymposium on Power Line Communications and Its Applications(ISPLC rsquo06) pp 245ndash250 Orlando Fla USA March 2006

[118] Ofcom ldquoDS2 PLT Measurements in Crieffrdquo Tech Rep 793(Part 2) Ofcom 2005 httpwwwofcomorgukresearchtechnologyresearcharchivecetpowerlineds2pdf

[119] Ofcom ldquoAscom PLT Measurements in Winchesterrdquo Tech Rep793 (Part 1) Ofcom 2005

[120] P A A F Wouters P C J M van der Wielen J Veen P Wage-naars and E F Wagenaars ldquoEffect of cable load impedance oncoupling schemes for MV power line communicationrdquo IEEETransactions onPowerDelivery vol 20 no 2 pp 638ndash645 2005

[121] A Cataliotti V Cosentino DDi Cara andG Tine ldquoSimulationand laboratory experimental tests of a line to shield medium-voltage power-line communication systemrdquo IEEE Transactionson Power Delivery vol 26 no 4 pp 2829ndash2836 2011

[122] M Tlich A Zeddam F Moulin and F Gauthier ldquoIndoorpower-line communications channel characterization up to 100MHz Part II time-frequency analysisrdquo IEEE Transactions onPower Delivery vol 23 no 3 pp 1402ndash1409 2008

[123] Nexans deutschland industries GmbH and Co KG catalogueLV underground power cables distribution cables 2011

[124] Tele-Fonika Kable GmbH Cables and wires 2008[125] G J Anders Rating of Electric Power Cables IEEE Press New

York NY USA 1997[126] D Veronesi R Riva P Bisaglia et al ldquoCharacterization of in-

home MIMO power line channelsrdquo in Proceedings of the IEEEInternational Symposium on Power Line Communications andIts Applications (ISPLC rsquo11) pp 42ndash47 Udine Italy April 2011

[127] W Q Malik ldquoMIMO capacity convergence in frequency-selective channelsrdquo IEEE Transactions on Communications vol57 no 2 pp 353ndash356 2009

[128] K Peppas F Lazarakis C Skianis A A Alexandridis and KDangakis ldquoImpact of MIMO techniques on the interoperabilitybetween UMTS-HSDPA and WLAN wireless systemsrdquo IEEECommunications Surveys and Tutorials vol 13 no 4 pp 708ndash720 2011

[129] RGGallager InformationTheory andReliable CommunicationWiley New York NY USA 1968

[130] M A Khalighi K Raoof and G Jourdain ldquoCapacity of wirelesscommunication systems employing antenna arrays a tutorialstudyrdquoWireless Personal Communications vol 23 no 3 pp 321ndash352 2002

[131] J B Andersen ldquoArray gain and capacity for known randomchannels with multiple element arrays at both endsrdquo IEEEJournal on Selected Areas in Communications vol 18 no 11 pp2172ndash2178 2000

[132] P-S Kildal and K Rosengren ldquoCorrelation and capacity ofMIMO systems and mutual coupling radiation efficiency anddiversity gain of their antennas simulations and measurementsin a reverberation chamberrdquo IEEE Communications Magazinevol 42 no 12 pp 104ndash112 2004


[133] K Sulonen P Suvikunnas L Vuokko J Kivinen and PVainikainen ldquoComparison ofMIMO antenna configurations inpicocell and microcell environmentsrdquo IEEE Journal on SelectedAreas in Communications vol 21 no 5 pp 703ndash712 2003

[134] Y J Zhang and A M-C So ldquoOptimal spectrum sharing inMIMO cognitive radio networks via semidefinite program-mingrdquo IEEE Journal on Selected Areas in Communications vol29 no 2 pp 362ndash373 2011

[135] O Tipmongkolsilp S Zaghloul and A Jukan ldquoThe evolutionof cellular backhaul technologies current issues and futuretrendsrdquo IEEECommunications Surveys and Tutorials vol 13 no1 pp 97ndash113 2011

[136] P-D Arapoglou K Liolis M Bertinelli A Panagopoulos PCottis and R de Gaudenzi ldquoMIMO over satellite a reviewrdquoIEEE Communications Surveys and Tutorials vol 13 no 1 pp27ndash51 2011

[137] K Hooghe andM Guenach ldquoToward green copper broadbandaccess networksrdquo IEEE Communications Magazine vol 49 no8 pp 87ndash93 2011

[138] W Vereecken W van Heddeghem M Deruyck et al ldquoPowerconsumption in telecommunication networks overview andreduction strategiesrdquo IEEE Communications Magazine vol 49no 6 pp 62ndash69 2011

[139] M M Rahman C S Hong S Lee J Lee M A Razzaque andJ H Kim ldquoMedium access control for power line communica-tions an overview of the IEEE 1901 and ITU-T Ghn standardsrdquoIEEECommunicationsMagazine vol 49 no 6 pp 183ndash191 2011

[140] R Bolla F Davoli R Bruschi K Christensen F Cucchietti andS Singh ldquoThe potential impact of green technologies in next-generation wireline networks is there room for energy savingoptimizationrdquo IEEE Communications Magazine vol 49 no 8pp 80ndash86 2011

International Journal of

AerospaceEngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014


Active and Passive Electronic Components

Control Scienceand Engineering

Journal of



RotatingMachinery


Hindawi Publishing Corporation httpwwwhindawicom

Journal ofEngineeringVolume 2014

Submit your manuscripts athttpwwwhindawicom

VLSI Design



Shock and Vibration


Civil EngineeringAdvances in

Acoustics and VibrationAdvances in



Electrical and Computer Engineering

Journal of

Advances inOptoElectronics


Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

SensorsJournal of


Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014


Chemical EngineeringInternational Journal of Antennas and

Propagation




Navigation and Observation



DistributedSensor Networks



through the deployment of broadbandover power lines (BPL)networks [7ndash14]

Despite their great broadband promises distributionpower grids constitute a rather hostile medium for commu-nications signal transmission Attenuation multipath due tovarious reflections noise and electromagnetic interferenceare the main problematic issues [15ndash22] Since overhead andunderground LVBPL and MVBPL channels mix the nastybehavior of a power line with that of a communication chan-nel each of the aforementioned adverse factors significantlydegrades the overall performance of BPL networks [23ndash25]

Recently new interest arises due to developments regard-ing multiple-input multiple-output (MIMO) transmissionscheme configurations for BPLnetworks andBPL coexistencewith other broadband technologies [13 14 26ndash30] Actuallythe need of coexistence among overhead and undergroundLVBPL and MVBPL networks (intraoperability) will be anecessary prerequisite before investigating the cooperativecommunications between distribution BPL networks andother broadband technologies (interoperability) [28 31ndash38] Despite the urgent demand for systems intraoperabil-ityinteroperability todayrsquos distribution BPL network capac-ity performance seems significantly worse than that of othercompetitive broadband technologiesmdashwired such as fiberand DSL and wireless such as Wi-Fi and WiMAX Hencethere is a need of not only utilizing existing technical devel-opments but also inventing and exploiting novel ones suchas MIMO technology

Being prominent from the wireless world [38ndash46] andrecently applied in various BPL networks [13 14 33 47ndash56]the potential of integrating MIMO technology with distri-bution BPL networks is investigated Towards that directionMIMO capabilities of distribution BPL networks are com-pared with todayrsquos single-input single-output (SISO) oneseither in transmission or in spectral efficiency fields

For efficient MIMOBPL (either MIMOLVBPL orMIMOMVBPL) communications a thorough understand-ing of the MIMO signal transmission at high frequenciesalong the overhead and underground LV andMVpower linesis required To facilitate MIMO analysis the hybrid modelthat is usually employed to examine the behavior of BPLtransmission channels installed onmulticonductor transmis-sion line (MTL) structures is also used in this review paper[2 3 5 6 9 10 12ndash14] This hybrid method based on acombination of bottom-up [57ndash65] and top-down [8 15 1824 58 62 64 66ndash69] approaches receives as inputs accu-rately determined parameters such as the MIMO schemeimplementation the power grid type (either overhead orunderground either LV or MV) the MTL configuration thephysical properties of cables and the grid topology and deliv-ers as outputs accurate results in terms of channel attenuationActually exploiting the acquired MIMOBPL knowledge of[13 14] MIMO expansion of the hybrid method is firstpresented in order to (i) achieve a smooth expansion ofSISO to MIMO analysis and (ii) unify eigenvalue decom-position (EVD) and singular value decomposition (SVD)modal analyses under the aegis of the proposed unified valuedecomposition (UVD) modal analysis

Based on the outputs of UVDmodal analysis broadbandtransmission characteristics statistical performance metricsand capacity measures are investigated through numericalresults concerning simulated overhead and undergroundMIMOLVBPL and MIMOMVBPL networks [57 62 6467 70] The numerical results of UVD modal analysis arevalidated against already verified results of the accurate TM2method which are analytically presented in [13 14]

More specifically a set of important statistical perfor-mance metrics such as the average channel attenuation theaverage channel gain (ACG) the root-mean-square delayspread (RMS-DS) and the coherence bandwidth (CB) triesto elicit the commonand the different aspects of overhead andunderground MIMOLVBPL and MIMOMVBPL chan-nels In addition two fundamental properties of severalwireline channels (eg DSL coaxial and phone links) arefirst validated in overhead and undergroundMIMOLVBPLandMIMOMVBPL systems namely (i) ACG and RMS-DSare negatively correlated lognormal random variables beingin agreement with recently proposed statistical BPL channelmodels [71ndash79] A new approximationmdashUN1 approachmdashsuitable for the design of overhead and undergroundMIMOBPL channels is proposed and compared to existingdistributions and (ii) CB and RMS-DS correlation behaviorcan be described by hyperbolic functions By fitting thesimulation results a new approximationmdashUN2 approachmdashappropriate for MIMOBPL channels is proposed [78ndash82]Moreover lognormal distributions for the cumulative dis-tribution functions (CDFs) of ACG RMS-DS and cumu-lative capacitymdashsuitable for overhead and undergroundMIMOLVBPL andMIMOMVBPL channelsmdashare used soas to (a) be imposed by design in the upcoming BPL statisticalchannel modeling and (b) operate as evaluation measures ofthe statistical channel models proposed [71ndash73 76ndash79]

Apart from statistical performance metrics suitablecapacitymetrics such as the cumulative capacity the capacitycomplementary cumulative distribution functions (capacityCCDF) and the capacity gain (GC) of MIMOBPL systemsare applied It is unveiled thatMIMOBPL systems can definean alternative technological solution to the existing fiber-optic technology and especially to the fiber-to-the-building(FTTB) systems Depending on BPL system investmentbudget technoeconomic and socioeconomic characteristicsrequired system complexity power grid type power gridtopology IPSDM limits applied operation frequency bandand BPL intraoperabilityinteroperability requirements dif-ferent single- and multiport implementations may be appliedin improving todayrsquos system capacities By fully exploitingthe MIMO technology the increase of the capacity of bothoverhead and underground MIMOLVBPL and MIMOMVBPL systems is notable exceeding 8Gbps Using onlythe existing infrastructure and electromagnetic compatibilitylimits the imminent MIMOBPL technology implementa-tions promise better capacity results by a factor of up to365 in comparison with the best todayrsquos capacity results forgiven power grid type and power grid topology Hence theforthcoming MIMOBPL capacity boost indicates that theBPL technology should be examined as (i) an emerging


















Δ LV

Δ LVΔ LV

ΔMV ΔMVrMVp

120583o

120576g 120583o

hMV

120590g

hLV

(a) (b)

rMVs

rMVc

rMVΔMV

120579MV

x

y

z

Armor

Shield


Conductor 3

Armor insulation


(d)


rLVc

rLVp

rLVn

rLVs

rLV

(c)

120576o














[119881119898

1(119911) sdot sdot sdot 119881

119898


[119868119898

1(119911) sdot sdot sdot 119868

119898




60 61]

V (119911) = T119881 sdot V119898(119911) (1)

I (119911) = T119868 sdot I119898(119911) (2)



119894(119911) 119894 = 1 119899 and

119881119898


119894sdot 119894 = 1 119899

so that

V119898 (119911) = H119898 V119898 (0) (3)

where


119898

119899sdot (4)




V (119911) = H V (0) (5)

is determined from








+



V (0) = D119873119879119881sdot V+ (0) (7)

where D119873119879119881






1(119911) sdot sdot sdot 119881

+



V+ (119911) = D119873119877119881sdot V (119911) (8)

where D119873119877119881





119881 (9)


and D119873119877119881 UVD modal



H119898

sdot =

T119867


T119868

= (

T119867



119881sdotD119873119879119881sdot

T119868)

(10)

where

H119898

sdot

= [

[

diag

119867

119898



]

]

119894 = 1 min 119899119877 119899119879 (11)


119867

119898

119894sdot


T119881 and


H119898





119894119895sdot 119894 119895 = 1 119899




119894119895sdot 119894 119895 = 1 119899 of MIMO


119894119895119901= ℎ+

119894119895(119905 = 119901119879119904) 119901 = 0 119869 minus 1 119894 119895 = 1 119899


119867+

119894119895119902=

10038161003816100381610038161003816119867+

119894119895119902

10038161003816100381610038161003816119890119895120593+

119902 119902 = 0 119870 minus 1

0 119902 = 119870 119869 minus 1

= 119867+

119894119895(119891 = 119902119891119904) 119902 = 0 119870 minus 1

0 119902 = 119870 119869 minus 1 119894 119895 = 1 119899

(12)



119894119895119902| 119894 119895 = 1 119899 and 120601+

119894119895119902 119894 119895 =



119894119895|2 119894 119895 = 1 119899 can


10038161003816100381610038161003816119867+

119894119895

10038161003816100381610038161003816

2

=

119869minus1

sum

119901=0

10038161003816100381610038161003816ℎ+

119894119895119901

10038161003816100381610038161003816

2

=1

119869

119869minus1

sum

119902=0

10038161003816100381610038161003816119867+

119894119895119902

10038161003816100381610038161003816

2

119894 119895 = 1 119899 (13)


119894119895120591 119894 119895 =


120590+

119894119895120591= 119879119904

radic120583(2)

1198941198950minus (1205831198941198950)

2

119894 119895 = 1 119899 (14)


where

1205831198941198950 =

sum119869minus1

119901=011990110038161003816100381610038161003816ℎ+

119894119895119901

10038161003816100381610038161003816

2

sum119869minus1

119901=0

10038161003816100381610038161003816ℎ+

119894119895119901

10038161003816100381610038161003816

2 119894 119895 = 1 119899 (15a)

120583(2)

1198941198950=

sum119869minus1

119901=0119901210038161003816100381610038161003816ℎ+

119894119895119901

10038161003816100381610038161003816

2

sum119869minus1

119901=0

10038161003816100381610038161003816ℎ+

119894119895119901

10038161003816100381610038161003816

2 119894 119895 = 1 119899 (15b)


119861119894119895119888 (Δ119891) =

119864 119867+

119894119895119902sdot [119867+

119894119895(119891 = 119902119891119904 + Δ119891)]

lowast

119864 10038161003816100381610038161003816119867+

119894119895119902

10038161003816100381610038161003816

2

119902 = 0 1 lfloor(119869 minus 2) +Δ119891

119891119904

rfloor 119894 119895 = 1 119899

(16)



119862MIMO

= 119891119904

119869minus1

sum

119902=0

min119899119879119899119877

sum

119894=1

log21 +

SNR (119902119891119904)119899119879

sdot

1003816100381610038161003816100381610038161003816

119867

119898

119894(119902119891119904)

1003816100381610038161003816100381610038161003816

2

(17)

where

SNR (119891) =119901 (119891)

119873 (119891)(18)







































V+in

L2L1 Lk LN+1

A998400N

A998400kA998400

2A9984001


Vminusout


A1 A2 Ak AN




0

0

50

50

100

100

150

Frequency (MHz)

Chan

nel a

ttenu

atio

n (d

B)

(a)

0 50 100

Frequency (MHz)

0

50

100

150

Chan

nel a

ttenu

atio

n (d

B)

(b)


0

50

100

150



Suburban case

Chan

nel a

ttenu

atio

n (d

B)

(c)


0

50

100

150



Suburban case

Chan

nel a

ttenu

atio

n (d

B)

(d)






















minus1 0

0

1minus70

minus60

minus50

minus40

minus30

minus20

minus10

ACG

(dB)

Overhead XC

Overhead CC

(a)

1

08

06

04

02


ACG (dB)

Overhead XC

Overhead CC

ACG

CD

F (A

CGlt

absc

issa)

(b)

0

minus70

minus60

minus50

minus40

minus30

minus20

minus10

ACG

(dB)


Underground XC

Underground CC


(c)

1

08

06

04

02


ACG (dB)

Underground XC

Underground CC

ACG

CD

F (A

CGlt

absc

issa)


(d)





















Overhead XC

Overhead CC


0

05

15

2

1

minus2 0 2

RMS-

DS

(120583s)

(a)

RMS-DS (120583s)0 1 2

1

08

06

04

02

0

Overhead XC

Overhead CC

RMS-

DS

CDF

(RM

S-D

Slt

absc

issa)

(b)

0

05

15

2

1

RMS-

DS

(120583s)


Underground XC

Underground XC

Underground CC


(c)

1

08

06

04

02

0

RMS-

DS

CDF

(RM

S-D

Slt

absc

issa)

RMS-DS (120583s)0 1 2

Underground XC

Underground CC


(d)






LV 009 026 112 083 068 053 042 020MV 038 061 108 146 078 100 030 034

XCLV 011 011 192 078 106 050 066 024MV 011 056 184 159 104 106 063 037

AllLV 011 011 192 078 106 050 066 025MV 028 056 181 159 111 106 055 037





















0

05

15

2

1

RMS-

DS

(120583s)


ACG (dB)




(a)

0

05

15

2

1

RMS-

DS

(120583s)





ACG (dB)

(b)


0 10 20 30 40 50CB (MHz)



0

05

15

2

1

RMS-

DS

(120583s)

(a)

0 10 20 30 40 50CB (MHz)



0

05

15

2

1RM

S-D

S (120583

s)

(b)
















LV 070 080 4950 37 1182 1232 2114 1631MV 080 050 4760 560 1178 232 2016 228

XCLV 020 040 3150 48 700 1020 1372 2113MV 020 030 3220 880 716 206 1402 377

AllLV 020 040 3250 48 720 1020 1416 2113MV 020 030 27 880 576 206 1188 377














00

500

1000

2000

1500

Cum

ulat

ive c

apac

ity (M

bps)

Overhead LVXC

Overhead LVCC

(a)

0

500

1000

2000

1500

Cum

ulat

ive c

apac

ity (M

bps)


Overhead MVCC

Overhead MVXC

(b)

0

Capa

city

CCD

F (c

apac

itygt

absc

issa)

01

02

03

04

05

06

07

08

09

1

0 500 1000 1500 2000

Overhead LVCC

Overhead MVCC

Overhead MVXC

Overhead LVXC

Capacity (Mbps)

(c)

0

500

1000

2000

1500

Cum

ulat

ive c

apac

ity (M

bps)

UndergroundLVCC

UndergroundLVXC



Suburban case


0

(d)

0

500

1000

2000

1500

Cum

ulat

ive c

apac

ity (M

bps)

Underground MVXC

UndergroundMVCC



Suburban case


0

(e)

0 500 1000 1500 20000

01

02

03

04

05

06

07

08

09

1

UndergroundLVCC

UndergroundLVXC

UndergroundMVCC

Underground

Capacity (Mbps)

Capa

city

CCD

F (c

apac

itygt

absc

issa)

MVXC

(f)











0

500

1000

2000

3000

1500

2500

0

500

1000

2000

3000

1500

2500

Cum

ulat

ive c

apac

ity (M

bps)

0

500

1000

2000

3000

1500

2500

Cum

ulat

ive c

apac

ity (M

bps)

0

500

1000

2000

3000

1500

2500

Cum

ulat

ive c

apac

ity (M

bps)

0

500

1000

2000

3000

1500

2500

Cum

ulat

ive c

apac

ity (M

bps)

0

500

1000

2000

3000

1500

2500

Cum

ulat

ive c

apac

ity (M

bps)

Cum

ulat

ive c

apac

ity (M

bps)

00

1000 2000 3000

Capa

city

CCD

F (c

apac

itygt

absc

issa)

01

02

03

04

05

06

07

08

09

1

(g)Capacity (Mbps)

0 1000 2000 3000C (Mb )




4 times 1 MISOLV



Suburban case



Suburban case

(a) (b)

(c)

(e) (f)

(d)







2sdot presence in












0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)

00

1000

2000

2000

3000

4000

4000

5000

6000

6000

7000

8000

8000

Cum

ulat

ive c

apac

ity (M

bps)

0

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)






Capa

city

CCD

F (c

apac

itygt

absc

issa)

01

02

03

04

05

06

07

08

09


3 times 1 MISOMV

4 times 4 MIMOLV

3 times 3 MIMOMV

SISOCCMVSISOXCMV

(a) (b)

(c) (d)

(e) (f)

(g)Capacity (Mbps)



Suburban case



Suburban case

1 times 3 SIMOMV






LV 471 1634 904 2152 700 1928 168 200 4MV 459 664 892 1019 688 855 168 137 3

1 times 2 SIMOLV 345 1497 693 1999 530 1785 140 194 24MV 375 471 744 769 572 632 149 114 9

1 times 3 SIMOLV 476 1646 907 2161 704 1938 168 198 16MV 477 664 906 1019 705 855 168 137 3


2 times 1 MISOLV 286 1397 604 1896 453 1684 128 193 24MV 399 613 813 951 616 794 161 1303 9

3 times 1 MISOLV 370 1511 772 2026 579 1804 156 198 16MV 370 583 770 912 579 760 156 127 3


2 times 2 MIMOLV 537 2770 1114 3779 854 3349 241 390 36MV 602 828 1261 1381 949 1129 264 212 9

2 times 3 MIMOLV 536 2770 1137 3779 851 3349 240 390 24MV 736 1166 1538 1825 1155 1520 312 253 3


3 times 2 MIMOLV 530 2770 1135 3779 847 3349 243 390 24MV 732 1166 1535 1824 1151 1520 312 253 3

3 times 3 MIMOLV 774 3058 1591 4092 1201 3645 317 398 16MV 1082 1750 2288 2736 1710 2280 468 380 1



Table 4 Continued













119894119895 119894 isin 119873119879 119895 isin 119873119877 [130ndash133]




0

500

1000

2000

3000

1500

2500Cu

mul

ativ

e cap

acity

(Mbp

s)


1 times 4 SIMO

4 times 4 MIMO


3 times 2 MIMO

1 times 3 SIMO

SISOCCLV SISOCCMV

(a)

1

08

09

06

07

04

05

03

01

02

00 1000 2000 3000


SISOCCLV

SISOCCMV

Cum

ulat

ive c

apac

ity C

DF

(cum

ulat

ive c

apac

itylt

absc

issa)

4 times 4 MIMO

4 times 3 MIMO

1 times 3 SIMO

1 times 4 SIMO

3 times 2 MIMO

3 times 3 MIMO

(b)

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)


SISOCCLV

SISOCCMVminus1

4 times 4 MIMO

4 times 3 MIMO

3 times 3 MIMO

1 times 3 SIMO

1 times 4 SIMO

3 times 2 MIMO


(c)


08

09

1

06

07

04

05

03

01

02

0

SISOCCLV

SISOCCMV

Cum

ulat

ive c

apac

ity C

DF

(cum

ulat

ive c

apac

itylt

absc

issa)

3 times 3 MIMO

3 times 2 MIMO

1 times 4 SIMO

1 times 3 SIMO

4 times 3 MIMO

4 times 4 MIMO


(d)



11

1

1

11

2

2

2

3

3

4

4

079083064

12

12

119

312

171

15

25

171

121

101 165157

Rece

ive p

orts

(nR

)

Transmit ports (nT)

075

(a)

111 1 1

1

2

2

2

3

3

Transmit ports (nT)

12

12

14

14

16

22

18

167

102 248

082

089 084

136

167Re

ceiv

e por

ts (n

R)

(b)

11

1

2

3

4 365101

101

187184

174

174 174 174

189

093

088 092094

189

1

1

15

15

252 3


Rece

ive p

orts

(nR

)

(c)


1 1

1

1

2

1

2

3

178

178 267

089

132

093

074

15

Rece

ive p

orts

(nR

)

(d)









6 Conclusions






References























































































































































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation


























Δ LV

Δ LVΔ LV

ΔMV ΔMVrMVp

120583o

120576g 120583o

hMV

120590g

hLV

(a) (b)

rMVs

rMVc

rMVΔMV

120579MV

x

y

z

Armor

Shield


Conductor 3

Armor insulation


(d)


rLVc

rLVp

rLVn

rLVs

rLV

(c)

120576o














[119881119898

1(119911) sdot sdot sdot 119881

119898


[119868119898

1(119911) sdot sdot sdot 119868

119898




60 61]

V (119911) = T119881 sdot V119898(119911) (1)

I (119911) = T119868 sdot I119898(119911) (2)



119894(119911) 119894 = 1 119899 and

119881119898


119894sdot 119894 = 1 119899

so that

V119898 (119911) = H119898 V119898 (0) (3)

where


119898

119899sdot (4)




V (119911) = H V (0) (5)

is determined from








+



V (0) = D119873119879119881sdot V+ (0) (7)

where D119873119879119881






1(119911) sdot sdot sdot 119881

+



V+ (119911) = D119873119877119881sdot V (119911) (8)

where D119873119877119881





119881 (9)


and D119873119877119881 UVD modal



H119898

sdot =

T119867


T119868

= (

T119867



119881sdotD119873119879119881sdot

T119868)

(10)

where

H119898

sdot

= [

[

diag

119867

119898



]

]

119894 = 1 min 119899119877 119899119879 (11)


119867

119898

119894sdot


T119881 and


H119898





119894119895sdot 119894 119895 = 1 119899




119894119895sdot 119894 119895 = 1 119899 of MIMO


119894119895119901= ℎ+

119894119895(119905 = 119901119879119904) 119901 = 0 119869 minus 1 119894 119895 = 1 119899


119867+

119894119895119902=

10038161003816100381610038161003816119867+

119894119895119902

10038161003816100381610038161003816119890119895120593+

119902 119902 = 0 119870 minus 1

0 119902 = 119870 119869 minus 1

= 119867+

119894119895(119891 = 119902119891119904) 119902 = 0 119870 minus 1

0 119902 = 119870 119869 minus 1 119894 119895 = 1 119899

(12)



119894119895119902| 119894 119895 = 1 119899 and 120601+

119894119895119902 119894 119895 =



119894119895|2 119894 119895 = 1 119899 can


10038161003816100381610038161003816119867+

119894119895

10038161003816100381610038161003816

2

=

119869minus1

sum

119901=0

10038161003816100381610038161003816ℎ+

119894119895119901

10038161003816100381610038161003816

2

=1

119869

119869minus1

sum

119902=0

10038161003816100381610038161003816119867+

119894119895119902

10038161003816100381610038161003816

2

119894 119895 = 1 119899 (13)


119894119895120591 119894 119895 =


120590+

119894119895120591= 119879119904

radic120583(2)

1198941198950minus (1205831198941198950)

2

119894 119895 = 1 119899 (14)


where

1205831198941198950 =

sum119869minus1

119901=011990110038161003816100381610038161003816ℎ+

119894119895119901

10038161003816100381610038161003816

2

sum119869minus1

119901=0

10038161003816100381610038161003816ℎ+

119894119895119901

10038161003816100381610038161003816

2 119894 119895 = 1 119899 (15a)

120583(2)

1198941198950=

sum119869minus1

119901=0119901210038161003816100381610038161003816ℎ+

119894119895119901

10038161003816100381610038161003816

2

sum119869minus1

119901=0

10038161003816100381610038161003816ℎ+

119894119895119901

10038161003816100381610038161003816

2 119894 119895 = 1 119899 (15b)


119861119894119895119888 (Δ119891) =

119864 119867+

119894119895119902sdot [119867+

119894119895(119891 = 119902119891119904 + Δ119891)]

lowast

119864 10038161003816100381610038161003816119867+

119894119895119902

10038161003816100381610038161003816

2

119902 = 0 1 lfloor(119869 minus 2) +Δ119891

119891119904

rfloor 119894 119895 = 1 119899

(16)



119862MIMO

= 119891119904

119869minus1

sum

119902=0

min119899119879119899119877

sum

119894=1

log21 +

SNR (119902119891119904)119899119879

sdot

1003816100381610038161003816100381610038161003816

119867

119898

119894(119902119891119904)

1003816100381610038161003816100381610038161003816

2

(17)

where

SNR (119891) =119901 (119891)

119873 (119891)(18)







































V+in

L2L1 Lk LN+1

A998400N

A998400kA998400

2A9984001


Vminusout


A1 A2 Ak AN




0

0

50

50

100

100

150

Frequency (MHz)

Chan

nel a

ttenu

atio

n (d

B)

(a)

0 50 100

Frequency (MHz)

0

50

100

150

Chan

nel a

ttenu

atio

n (d

B)

(b)


0

50

100

150



Suburban case

Chan

nel a

ttenu

atio

n (d

B)

(c)


0

50

100

150



Suburban case

Chan

nel a

ttenu

atio

n (d

B)

(d)






















minus1 0

0

1minus70

minus60

minus50

minus40

minus30

minus20

minus10

ACG

(dB)

Overhead XC

Overhead CC

(a)

1

08

06

04

02


ACG (dB)

Overhead XC

Overhead CC

ACG

CD

F (A

CGlt

absc

issa)

(b)

0

minus70

minus60

minus50

minus40

minus30

minus20

minus10

ACG

(dB)


Underground XC

Underground CC


(c)

1

08

06

04

02


ACG (dB)

Underground XC

Underground CC

ACG

CD

F (A

CGlt

absc

issa)


(d)





















Overhead XC

Overhead CC


0

05

15

2

1

minus2 0 2

RMS-

DS

(120583s)

(a)

RMS-DS (120583s)0 1 2

1

08

06

04

02

0

Overhead XC

Overhead CC

RMS-

DS

CDF

(RM

S-D

Slt

absc

issa)

(b)

0

05

15

2

1

RMS-

DS

(120583s)


Underground XC

Underground XC

Underground CC


(c)

1

08

06

04

02

0

RMS-

DS

CDF

(RM

S-D

Slt

absc

issa)

RMS-DS (120583s)0 1 2

Underground XC

Underground CC


(d)






LV 009 026 112 083 068 053 042 020MV 038 061 108 146 078 100 030 034

XCLV 011 011 192 078 106 050 066 024MV 011 056 184 159 104 106 063 037

AllLV 011 011 192 078 106 050 066 025MV 028 056 181 159 111 106 055 037





















0

05

15

2

1

RMS-

DS

(120583s)


ACG (dB)




(a)

0

05

15

2

1

RMS-

DS

(120583s)





ACG (dB)

(b)


0 10 20 30 40 50CB (MHz)



0

05

15

2

1

RMS-

DS

(120583s)

(a)

0 10 20 30 40 50CB (MHz)



0

05

15

2

1RM

S-D

S (120583

s)

(b)
















LV 070 080 4950 37 1182 1232 2114 1631MV 080 050 4760 560 1178 232 2016 228

XCLV 020 040 3150 48 700 1020 1372 2113MV 020 030 3220 880 716 206 1402 377

AllLV 020 040 3250 48 720 1020 1416 2113MV 020 030 27 880 576 206 1188 377














00

500

1000

2000

1500

Cum

ulat

ive c

apac

ity (M

bps)

Overhead LVXC

Overhead LVCC

(a)

0

500

1000

2000

1500

Cum

ulat

ive c

apac

ity (M

bps)


Overhead MVCC

Overhead MVXC

(b)

0

Capa

city

CCD

F (c

apac

itygt

absc

issa)

01

02

03

04

05

06

07

08

09

1

0 500 1000 1500 2000

Overhead LVCC

Overhead MVCC

Overhead MVXC

Overhead LVXC

Capacity (Mbps)

(c)

0

500

1000

2000

1500

Cum

ulat

ive c

apac

ity (M

bps)

UndergroundLVCC

UndergroundLVXC



Suburban case


0

(d)

0

500

1000

2000

1500

Cum

ulat

ive c

apac

ity (M

bps)

Underground MVXC

UndergroundMVCC



Suburban case


0

(e)

0 500 1000 1500 20000

01

02

03

04

05

06

07

08

09

1

UndergroundLVCC

UndergroundLVXC

UndergroundMVCC

Underground

Capacity (Mbps)

Capa

city

CCD

F (c

apac

itygt

absc

issa)

MVXC

(f)











0

500

1000

2000

3000

1500

2500

0

500

1000

2000

3000

1500

2500

Cum

ulat

ive c

apac

ity (M

bps)

0

500

1000

2000

3000

1500

2500

Cum

ulat

ive c

apac

ity (M

bps)

0

500

1000

2000

3000

1500

2500

Cum

ulat

ive c

apac

ity (M

bps)

0

500

1000

2000

3000

1500

2500

Cum

ulat

ive c

apac

ity (M

bps)

0

500

1000

2000

3000

1500

2500

Cum

ulat

ive c

apac

ity (M

bps)

Cum

ulat

ive c

apac

ity (M

bps)

00

1000 2000 3000

Capa

city

CCD

F (c

apac

itygt

absc

issa)

01

02

03

04

05

06

07

08

09

1

(g)Capacity (Mbps)

0 1000 2000 3000C (Mb )




4 times 1 MISOLV



Suburban case



Suburban case

(a) (b)

(c)

(e) (f)

(d)







2sdot presence in












0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)

00

1000

2000

2000

3000

4000

4000

5000

6000

6000

7000

8000

8000

Cum

ulat

ive c

apac

ity (M

bps)

0

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)






Capa

city

CCD

F (c

apac

itygt

absc

issa)

01

02

03

04

05

06

07

08

09


3 times 1 MISOMV

4 times 4 MIMOLV

3 times 3 MIMOMV

SISOCCMVSISOXCMV

(a) (b)

(c) (d)

(e) (f)

(g)Capacity (Mbps)



Suburban case



Suburban case

1 times 3 SIMOMV






LV 471 1634 904 2152 700 1928 168 200 4MV 459 664 892 1019 688 855 168 137 3

1 times 2 SIMOLV 345 1497 693 1999 530 1785 140 194 24MV 375 471 744 769 572 632 149 114 9

1 times 3 SIMOLV 476 1646 907 2161 704 1938 168 198 16MV 477 664 906 1019 705 855 168 137 3


2 times 1 MISOLV 286 1397 604 1896 453 1684 128 193 24MV 399 613 813 951 616 794 161 1303 9

3 times 1 MISOLV 370 1511 772 2026 579 1804 156 198 16MV 370 583 770 912 579 760 156 127 3


2 times 2 MIMOLV 537 2770 1114 3779 854 3349 241 390 36MV 602 828 1261 1381 949 1129 264 212 9

2 times 3 MIMOLV 536 2770 1137 3779 851 3349 240 390 24MV 736 1166 1538 1825 1155 1520 312 253 3


3 times 2 MIMOLV 530 2770 1135 3779 847 3349 243 390 24MV 732 1166 1535 1824 1151 1520 312 253 3

3 times 3 MIMOLV 774 3058 1591 4092 1201 3645 317 398 16MV 1082 1750 2288 2736 1710 2280 468 380 1



Table 4 Continued













119894119895 119894 isin 119873119879 119895 isin 119873119877 [130ndash133]




0

500

1000

2000

3000

1500

2500Cu

mul

ativ

e cap

acity

(Mbp

s)


1 times 4 SIMO

4 times 4 MIMO


3 times 2 MIMO

1 times 3 SIMO

SISOCCLV SISOCCMV

(a)

1

08

09

06

07

04

05

03

01

02

00 1000 2000 3000


SISOCCLV

SISOCCMV

Cum

ulat

ive c

apac

ity C

DF

(cum

ulat

ive c

apac

itylt

absc

issa)

4 times 4 MIMO

4 times 3 MIMO

1 times 3 SIMO

1 times 4 SIMO

3 times 2 MIMO

3 times 3 MIMO

(b)

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)


SISOCCLV

SISOCCMVminus1

4 times 4 MIMO

4 times 3 MIMO

3 times 3 MIMO

1 times 3 SIMO

1 times 4 SIMO

3 times 2 MIMO


(c)


08

09

1

06

07

04

05

03

01

02

0

SISOCCLV

SISOCCMV

Cum

ulat

ive c

apac

ity C

DF

(cum

ulat

ive c

apac

itylt

absc

issa)

3 times 3 MIMO

3 times 2 MIMO

1 times 4 SIMO

1 times 3 SIMO

4 times 3 MIMO

4 times 4 MIMO


(d)



11

1

1

11

2

2

2

3

3

4

4

079083064

12

12

119

312

171

15

25

171

121

101 165157

Rece

ive p

orts

(nR

)

Transmit ports (nT)

075

(a)

111 1 1

1

2

2

2

3

3

Transmit ports (nT)

12

12

14

14

16

22

18

167

102 248

082

089 084

136

167Re

ceiv

e por

ts (n

R)

(b)

11

1

2

3

4 365101

101

187184

174

174 174 174

189

093

088 092094

189

1

1

15

15

252 3


Rece

ive p

orts

(nR

)

(c)


1 1

1

1

2

1

2

3

178

178 267

089

132

093

074

15

Rece

ive p

orts

(nR

)

(d)









6 Conclusions






References























































































































































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation
















[119881119898

1(119911) sdot sdot sdot 119881

119898


[119868119898

1(119911) sdot sdot sdot 119868

119898




60 61]

V (119911) = T119881 sdot V119898(119911) (1)

I (119911) = T119868 sdot I119898(119911) (2)



119894(119911) 119894 = 1 119899 and

119881119898


119894sdot 119894 = 1 119899

so that

V119898 (119911) = H119898 V119898 (0) (3)

where


119898

119899sdot (4)




V (119911) = H V (0) (5)

is determined from








+



V (0) = D119873119879119881sdot V+ (0) (7)

where D119873119879119881






1(119911) sdot sdot sdot 119881

+



V+ (119911) = D119873119877119881sdot V (119911) (8)

where D119873119877119881





119881 (9)


and D119873119877119881 UVD modal



H119898

sdot =

T119867


T119868

= (

T119867



119881sdotD119873119879119881sdot

T119868)

(10)

where

H119898

sdot

= [

[

diag

119867

119898



]

]

119894 = 1 min 119899119877 119899119879 (11)


119867

119898

119894sdot


T119881 and


H119898





119894119895sdot 119894 119895 = 1 119899




119894119895sdot 119894 119895 = 1 119899 of MIMO


119894119895119901= ℎ+

119894119895(119905 = 119901119879119904) 119901 = 0 119869 minus 1 119894 119895 = 1 119899


119867+

119894119895119902=

10038161003816100381610038161003816119867+

119894119895119902

10038161003816100381610038161003816119890119895120593+

119902 119902 = 0 119870 minus 1

0 119902 = 119870 119869 minus 1

= 119867+

119894119895(119891 = 119902119891119904) 119902 = 0 119870 minus 1

0 119902 = 119870 119869 minus 1 119894 119895 = 1 119899

(12)



119894119895119902| 119894 119895 = 1 119899 and 120601+

119894119895119902 119894 119895 =



119894119895|2 119894 119895 = 1 119899 can


10038161003816100381610038161003816119867+

119894119895

10038161003816100381610038161003816

2

=

119869minus1

sum

119901=0

10038161003816100381610038161003816ℎ+

119894119895119901

10038161003816100381610038161003816

2

=1

119869

119869minus1

sum

119902=0

10038161003816100381610038161003816119867+

119894119895119902

10038161003816100381610038161003816

2

119894 119895 = 1 119899 (13)


119894119895120591 119894 119895 =


120590+

119894119895120591= 119879119904

radic120583(2)

1198941198950minus (1205831198941198950)

2

119894 119895 = 1 119899 (14)


where

1205831198941198950 =

sum119869minus1

119901=011990110038161003816100381610038161003816ℎ+

119894119895119901

10038161003816100381610038161003816

2

sum119869minus1

119901=0

10038161003816100381610038161003816ℎ+

119894119895119901

10038161003816100381610038161003816

2 119894 119895 = 1 119899 (15a)

120583(2)

1198941198950=

sum119869minus1

119901=0119901210038161003816100381610038161003816ℎ+

119894119895119901

10038161003816100381610038161003816

2

sum119869minus1

119901=0

10038161003816100381610038161003816ℎ+

119894119895119901

10038161003816100381610038161003816

2 119894 119895 = 1 119899 (15b)


119861119894119895119888 (Δ119891) =

119864 119867+

119894119895119902sdot [119867+

119894119895(119891 = 119902119891119904 + Δ119891)]

lowast

119864 10038161003816100381610038161003816119867+

119894119895119902

10038161003816100381610038161003816

2

119902 = 0 1 lfloor(119869 minus 2) +Δ119891

119891119904

rfloor 119894 119895 = 1 119899

(16)



119862MIMO

= 119891119904

119869minus1

sum

119902=0

min119899119879119899119877

sum

119894=1

log21 +

SNR (119902119891119904)119899119879

sdot

1003816100381610038161003816100381610038161003816

119867

119898

119894(119902119891119904)

1003816100381610038161003816100381610038161003816

2

(17)

where

SNR (119891) =119901 (119891)

119873 (119891)(18)







































V+in

L2L1 Lk LN+1

A998400N

A998400kA998400

2A9984001


Vminusout


A1 A2 Ak AN




0

0

50

50

100

100

150

Frequency (MHz)

Chan

nel a

ttenu

atio

n (d

B)

(a)

0 50 100

Frequency (MHz)

0

50

100

150

Chan

nel a

ttenu

atio

n (d

B)

(b)


0

50

100

150



Suburban case

Chan

nel a

ttenu

atio

n (d

B)

(c)


0

50

100

150



Suburban case

Chan

nel a

ttenu

atio

n (d

B)

(d)






















minus1 0

0

1minus70

minus60

minus50

minus40

minus30

minus20

minus10

ACG

(dB)

Overhead XC

Overhead CC

(a)

1

08

06

04

02


ACG (dB)

Overhead XC

Overhead CC

ACG

CD

F (A

CGlt

absc

issa)

(b)

0

minus70

minus60

minus50

minus40

minus30

minus20

minus10

ACG

(dB)


Underground XC

Underground CC


(c)

1

08

06

04

02


ACG (dB)

Underground XC

Underground CC

ACG

CD

F (A

CGlt

absc

issa)


(d)





















Overhead XC

Overhead CC


0

05

15

2

1

minus2 0 2

RMS-

DS

(120583s)

(a)

RMS-DS (120583s)0 1 2

1

08

06

04

02

0

Overhead XC

Overhead CC

RMS-

DS

CDF

(RM

S-D

Slt

absc

issa)

(b)

0

05

15

2

1

RMS-

DS

(120583s)


Underground XC

Underground XC

Underground CC


(c)

1

08

06

04

02

0

RMS-

DS

CDF

(RM

S-D

Slt

absc

issa)

RMS-DS (120583s)0 1 2

Underground XC

Underground CC


(d)






LV 009 026 112 083 068 053 042 020MV 038 061 108 146 078 100 030 034

XCLV 011 011 192 078 106 050 066 024MV 011 056 184 159 104 106 063 037

AllLV 011 011 192 078 106 050 066 025MV 028 056 181 159 111 106 055 037





















0

05

15

2

1

RMS-

DS

(120583s)


ACG (dB)




(a)

0

05

15

2

1

RMS-

DS

(120583s)





ACG (dB)

(b)


0 10 20 30 40 50CB (MHz)



0

05

15

2

1

RMS-

DS

(120583s)

(a)

0 10 20 30 40 50CB (MHz)



0

05

15

2

1RM

S-D

S (120583

s)

(b)
















LV 070 080 4950 37 1182 1232 2114 1631MV 080 050 4760 560 1178 232 2016 228

XCLV 020 040 3150 48 700 1020 1372 2113MV 020 030 3220 880 716 206 1402 377

AllLV 020 040 3250 48 720 1020 1416 2113MV 020 030 27 880 576 206 1188 377














00

500

1000

2000

1500

Cum

ulat

ive c

apac

ity (M

bps)

Overhead LVXC

Overhead LVCC

(a)

0

500

1000

2000

1500

Cum

ulat

ive c

apac

ity (M

bps)


Overhead MVCC

Overhead MVXC

(b)

0

Capa

city

CCD

F (c

apac

itygt

absc

issa)

01

02

03

04

05

06

07

08

09

1

0 500 1000 1500 2000

Overhead LVCC

Overhead MVCC

Overhead MVXC

Overhead LVXC

Capacity (Mbps)

(c)

0

500

1000

2000

1500

Cum

ulat

ive c

apac

ity (M

bps)

UndergroundLVCC

UndergroundLVXC



Suburban case


0

(d)

0

500

1000

2000

1500

Cum

ulat

ive c

apac

ity (M

bps)

Underground MVXC

UndergroundMVCC



Suburban case


0

(e)

0 500 1000 1500 20000

01

02

03

04

05

06

07

08

09

1

UndergroundLVCC

UndergroundLVXC

UndergroundMVCC

Underground

Capacity (Mbps)

Capa

city

CCD

F (c

apac

itygt

absc

issa)

MVXC

(f)











0

500

1000

2000

3000

1500

2500

0

500

1000

2000

3000

1500

2500

Cum

ulat

ive c

apac

ity (M

bps)

0

500

1000

2000

3000

1500

2500

Cum

ulat

ive c

apac

ity (M

bps)

0

500

1000

2000

3000

1500

2500

Cum

ulat

ive c

apac

ity (M

bps)

0

500

1000

2000

3000

1500

2500

Cum

ulat

ive c

apac

ity (M

bps)

0

500

1000

2000

3000

1500

2500

Cum

ulat

ive c

apac

ity (M

bps)

Cum

ulat

ive c

apac

ity (M

bps)

00

1000 2000 3000

Capa

city

CCD

F (c

apac

itygt

absc

issa)

01

02

03

04

05

06

07

08

09

1

(g)Capacity (Mbps)

0 1000 2000 3000C (Mb )




4 times 1 MISOLV



Suburban case



Suburban case

(a) (b)

(c)

(e) (f)

(d)







2sdot presence in












0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)

00

1000

2000

2000

3000

4000

4000

5000

6000

6000

7000

8000

8000

Cum

ulat

ive c

apac

ity (M

bps)

0

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)






Capa

city

CCD

F (c

apac

itygt

absc

issa)

01

02

03

04

05

06

07

08

09


3 times 1 MISOMV

4 times 4 MIMOLV

3 times 3 MIMOMV

SISOCCMVSISOXCMV

(a) (b)

(c) (d)

(e) (f)

(g)Capacity (Mbps)



Suburban case



Suburban case

1 times 3 SIMOMV






LV 471 1634 904 2152 700 1928 168 200 4MV 459 664 892 1019 688 855 168 137 3

1 times 2 SIMOLV 345 1497 693 1999 530 1785 140 194 24MV 375 471 744 769 572 632 149 114 9

1 times 3 SIMOLV 476 1646 907 2161 704 1938 168 198 16MV 477 664 906 1019 705 855 168 137 3


2 times 1 MISOLV 286 1397 604 1896 453 1684 128 193 24MV 399 613 813 951 616 794 161 1303 9

3 times 1 MISOLV 370 1511 772 2026 579 1804 156 198 16MV 370 583 770 912 579 760 156 127 3


2 times 2 MIMOLV 537 2770 1114 3779 854 3349 241 390 36MV 602 828 1261 1381 949 1129 264 212 9

2 times 3 MIMOLV 536 2770 1137 3779 851 3349 240 390 24MV 736 1166 1538 1825 1155 1520 312 253 3


3 times 2 MIMOLV 530 2770 1135 3779 847 3349 243 390 24MV 732 1166 1535 1824 1151 1520 312 253 3

3 times 3 MIMOLV 774 3058 1591 4092 1201 3645 317 398 16MV 1082 1750 2288 2736 1710 2280 468 380 1



Table 4 Continued













119894119895 119894 isin 119873119879 119895 isin 119873119877 [130ndash133]




0

500

1000

2000

3000

1500

2500Cu

mul

ativ

e cap

acity

(Mbp

s)


1 times 4 SIMO

4 times 4 MIMO


3 times 2 MIMO

1 times 3 SIMO

SISOCCLV SISOCCMV

(a)

1

08

09

06

07

04

05

03

01

02

00 1000 2000 3000


SISOCCLV

SISOCCMV

Cum

ulat

ive c

apac

ity C

DF

(cum

ulat

ive c

apac

itylt

absc

issa)

4 times 4 MIMO

4 times 3 MIMO

1 times 3 SIMO

1 times 4 SIMO

3 times 2 MIMO

3 times 3 MIMO

(b)

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)


SISOCCLV

SISOCCMVminus1

4 times 4 MIMO

4 times 3 MIMO

3 times 3 MIMO

1 times 3 SIMO

1 times 4 SIMO

3 times 2 MIMO


(c)


08

09

1

06

07

04

05

03

01

02

0

SISOCCLV

SISOCCMV

Cum

ulat

ive c

apac

ity C

DF

(cum

ulat

ive c

apac

itylt

absc

issa)

3 times 3 MIMO

3 times 2 MIMO

1 times 4 SIMO

1 times 3 SIMO

4 times 3 MIMO

4 times 4 MIMO


(d)



11

1

1

11

2

2

2

3

3

4

4

079083064

12

12

119

312

171

15

25

171

121

101 165157

Rece

ive p

orts

(nR

)

Transmit ports (nT)

075

(a)

111 1 1

1

2

2

2

3

3

Transmit ports (nT)

12

12

14

14

16

22

18

167

102 248

082

089 084

136

167Re

ceiv

e por

ts (n

R)

(b)

11

1

2

3

4 365101

101

187184

174

174 174 174

189

093

088 092094

189

1

1

15

15

252 3


Rece

ive p

orts

(nR

)

(c)


1 1

1

1

2

1

2

3

178

178 267

089

132

093

074

15

Rece

ive p

orts

(nR

)

(d)









6 Conclusions






References























































































































































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation













1(119911) sdot sdot sdot 119881

+



V+ (119911) = D119873119877119881sdot V (119911) (8)

where D119873119877119881





119881 (9)


and D119873119877119881 UVD modal



H119898

sdot =

T119867


T119868

= (

T119867



119881sdotD119873119879119881sdot

T119868)

(10)

where

H119898

sdot

= [

[

diag

119867

119898



]

]

119894 = 1 min 119899119877 119899119879 (11)


119867

119898

119894sdot


T119881 and


H119898





119894119895sdot 119894 119895 = 1 119899




119894119895sdot 119894 119895 = 1 119899 of MIMO


119894119895119901= ℎ+

119894119895(119905 = 119901119879119904) 119901 = 0 119869 minus 1 119894 119895 = 1 119899


119867+

119894119895119902=

10038161003816100381610038161003816119867+

119894119895119902

10038161003816100381610038161003816119890119895120593+

119902 119902 = 0 119870 minus 1

0 119902 = 119870 119869 minus 1

= 119867+

119894119895(119891 = 119902119891119904) 119902 = 0 119870 minus 1

0 119902 = 119870 119869 minus 1 119894 119895 = 1 119899

(12)



119894119895119902| 119894 119895 = 1 119899 and 120601+

119894119895119902 119894 119895 =



119894119895|2 119894 119895 = 1 119899 can


10038161003816100381610038161003816119867+

119894119895

10038161003816100381610038161003816

2

=

119869minus1

sum

119901=0

10038161003816100381610038161003816ℎ+

119894119895119901

10038161003816100381610038161003816

2

=1

119869

119869minus1

sum

119902=0

10038161003816100381610038161003816119867+

119894119895119902

10038161003816100381610038161003816

2

119894 119895 = 1 119899 (13)


119894119895120591 119894 119895 =


120590+

119894119895120591= 119879119904

radic120583(2)

1198941198950minus (1205831198941198950)

2

119894 119895 = 1 119899 (14)


where

1205831198941198950 =

sum119869minus1

119901=011990110038161003816100381610038161003816ℎ+

119894119895119901

10038161003816100381610038161003816

2

sum119869minus1

119901=0

10038161003816100381610038161003816ℎ+

119894119895119901

10038161003816100381610038161003816

2 119894 119895 = 1 119899 (15a)

120583(2)

1198941198950=

sum119869minus1

119901=0119901210038161003816100381610038161003816ℎ+

119894119895119901

10038161003816100381610038161003816

2

sum119869minus1

119901=0

10038161003816100381610038161003816ℎ+

119894119895119901

10038161003816100381610038161003816

2 119894 119895 = 1 119899 (15b)


119861119894119895119888 (Δ119891) =

119864 119867+

119894119895119902sdot [119867+

119894119895(119891 = 119902119891119904 + Δ119891)]

lowast

119864 10038161003816100381610038161003816119867+

119894119895119902

10038161003816100381610038161003816

2

119902 = 0 1 lfloor(119869 minus 2) +Δ119891

119891119904

rfloor 119894 119895 = 1 119899

(16)



119862MIMO

= 119891119904

119869minus1

sum

119902=0

min119899119879119899119877

sum

119894=1

log21 +

SNR (119902119891119904)119899119879

sdot

1003816100381610038161003816100381610038161003816

119867

119898

119894(119902119891119904)

1003816100381610038161003816100381610038161003816

2

(17)

where

SNR (119891) =119901 (119891)

119873 (119891)(18)







































V+in

L2L1 Lk LN+1

A998400N

A998400kA998400

2A9984001


Vminusout


A1 A2 Ak AN




0

0

50

50

100

100

150

Frequency (MHz)

Chan

nel a

ttenu

atio

n (d

B)

(a)

0 50 100

Frequency (MHz)

0

50

100

150

Chan

nel a

ttenu

atio

n (d

B)

(b)


0

50

100

150



Suburban case

Chan

nel a

ttenu

atio

n (d

B)

(c)


0

50

100

150



Suburban case

Chan

nel a

ttenu

atio

n (d

B)

(d)






















minus1 0

0

1minus70

minus60

minus50

minus40

minus30

minus20

minus10

ACG

(dB)

Overhead XC

Overhead CC

(a)

1

08

06

04

02


ACG (dB)

Overhead XC

Overhead CC

ACG

CD

F (A

CGlt

absc

issa)

(b)

0

minus70

minus60

minus50

minus40

minus30

minus20

minus10

ACG

(dB)


Underground XC

Underground CC


(c)

1

08

06

04

02


ACG (dB)

Underground XC

Underground CC

ACG

CD

F (A

CGlt

absc

issa)


(d)





















Overhead XC

Overhead CC


0

05

15

2

1

minus2 0 2

RMS-

DS

(120583s)

(a)

RMS-DS (120583s)0 1 2

1

08

06

04

02

0

Overhead XC

Overhead CC

RMS-

DS

CDF

(RM

S-D

Slt

absc

issa)

(b)

0

05

15

2

1

RMS-

DS

(120583s)


Underground XC

Underground XC

Underground CC


(c)

1

08

06

04

02

0

RMS-

DS

CDF

(RM

S-D

Slt

absc

issa)

RMS-DS (120583s)0 1 2

Underground XC

Underground CC


(d)






LV 009 026 112 083 068 053 042 020MV 038 061 108 146 078 100 030 034

XCLV 011 011 192 078 106 050 066 024MV 011 056 184 159 104 106 063 037

AllLV 011 011 192 078 106 050 066 025MV 028 056 181 159 111 106 055 037





















0

05

15

2

1

RMS-

DS

(120583s)


ACG (dB)




(a)

0

05

15

2

1

RMS-

DS

(120583s)





ACG (dB)

(b)


0 10 20 30 40 50CB (MHz)



0

05

15

2

1

RMS-

DS

(120583s)

(a)

0 10 20 30 40 50CB (MHz)



0

05

15

2

1RM

S-D

S (120583

s)

(b)
















LV 070 080 4950 37 1182 1232 2114 1631MV 080 050 4760 560 1178 232 2016 228

XCLV 020 040 3150 48 700 1020 1372 2113MV 020 030 3220 880 716 206 1402 377

AllLV 020 040 3250 48 720 1020 1416 2113MV 020 030 27 880 576 206 1188 377














00

500

1000

2000

1500

Cum

ulat

ive c

apac

ity (M

bps)

Overhead LVXC

Overhead LVCC

(a)

0

500

1000

2000

1500

Cum

ulat

ive c

apac

ity (M

bps)


Overhead MVCC

Overhead MVXC

(b)

0

Capa

city

CCD

F (c

apac

itygt

absc

issa)

01

02

03

04

05

06

07

08

09

1

0 500 1000 1500 2000

Overhead LVCC

Overhead MVCC

Overhead MVXC

Overhead LVXC

Capacity (Mbps)

(c)

0

500

1000

2000

1500

Cum

ulat

ive c

apac

ity (M

bps)

UndergroundLVCC

UndergroundLVXC



Suburban case


0

(d)

0

500

1000

2000

1500

Cum

ulat

ive c

apac

ity (M

bps)

Underground MVXC

UndergroundMVCC



Suburban case


0

(e)

0 500 1000 1500 20000

01

02

03

04

05

06

07

08

09

1

UndergroundLVCC

UndergroundLVXC

UndergroundMVCC

Underground

Capacity (Mbps)

Capa

city

CCD

F (c

apac

itygt

absc

issa)

MVXC

(f)











0

500

1000

2000

3000

1500

2500

0

500

1000

2000

3000

1500

2500

Cum

ulat

ive c

apac

ity (M

bps)

0

500

1000

2000

3000

1500

2500

Cum

ulat

ive c

apac

ity (M

bps)

0

500

1000

2000

3000

1500

2500

Cum

ulat

ive c

apac

ity (M

bps)

0

500

1000

2000

3000

1500

2500

Cum

ulat

ive c

apac

ity (M

bps)

0

500

1000

2000

3000

1500

2500

Cum

ulat

ive c

apac

ity (M

bps)

Cum

ulat

ive c

apac

ity (M

bps)

00

1000 2000 3000

Capa

city

CCD

F (c

apac

itygt

absc

issa)

01

02

03

04

05

06

07

08

09

1

(g)Capacity (Mbps)

0 1000 2000 3000C (Mb )




4 times 1 MISOLV



Suburban case



Suburban case

(a) (b)

(c)

(e) (f)

(d)







2sdot presence in












0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)

00

1000

2000

2000

3000

4000

4000

5000

6000

6000

7000

8000

8000

Cum

ulat

ive c

apac

ity (M

bps)

0

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)






Capa

city

CCD

F (c

apac

itygt

absc

issa)

01

02

03

04

05

06

07

08

09


3 times 1 MISOMV

4 times 4 MIMOLV

3 times 3 MIMOMV

SISOCCMVSISOXCMV

(a) (b)

(c) (d)

(e) (f)

(g)Capacity (Mbps)



Suburban case



Suburban case

1 times 3 SIMOMV






LV 471 1634 904 2152 700 1928 168 200 4MV 459 664 892 1019 688 855 168 137 3

1 times 2 SIMOLV 345 1497 693 1999 530 1785 140 194 24MV 375 471 744 769 572 632 149 114 9

1 times 3 SIMOLV 476 1646 907 2161 704 1938 168 198 16MV 477 664 906 1019 705 855 168 137 3


2 times 1 MISOLV 286 1397 604 1896 453 1684 128 193 24MV 399 613 813 951 616 794 161 1303 9

3 times 1 MISOLV 370 1511 772 2026 579 1804 156 198 16MV 370 583 770 912 579 760 156 127 3


2 times 2 MIMOLV 537 2770 1114 3779 854 3349 241 390 36MV 602 828 1261 1381 949 1129 264 212 9

2 times 3 MIMOLV 536 2770 1137 3779 851 3349 240 390 24MV 736 1166 1538 1825 1155 1520 312 253 3


3 times 2 MIMOLV 530 2770 1135 3779 847 3349 243 390 24MV 732 1166 1535 1824 1151 1520 312 253 3

3 times 3 MIMOLV 774 3058 1591 4092 1201 3645 317 398 16MV 1082 1750 2288 2736 1710 2280 468 380 1



Table 4 Continued













119894119895 119894 isin 119873119879 119895 isin 119873119877 [130ndash133]




0

500

1000

2000

3000

1500

2500Cu

mul

ativ

e cap

acity

(Mbp

s)


1 times 4 SIMO

4 times 4 MIMO


3 times 2 MIMO

1 times 3 SIMO

SISOCCLV SISOCCMV

(a)

1

08

09

06

07

04

05

03

01

02

00 1000 2000 3000


SISOCCLV

SISOCCMV

Cum

ulat

ive c

apac

ity C

DF

(cum

ulat

ive c

apac

itylt

absc

issa)

4 times 4 MIMO

4 times 3 MIMO

1 times 3 SIMO

1 times 4 SIMO

3 times 2 MIMO

3 times 3 MIMO

(b)

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)


SISOCCLV

SISOCCMVminus1

4 times 4 MIMO

4 times 3 MIMO

3 times 3 MIMO

1 times 3 SIMO

1 times 4 SIMO

3 times 2 MIMO


(c)


08

09

1

06

07

04

05

03

01

02

0

SISOCCLV

SISOCCMV

Cum

ulat

ive c

apac

ity C

DF

(cum

ulat

ive c

apac

itylt

absc

issa)

3 times 3 MIMO

3 times 2 MIMO

1 times 4 SIMO

1 times 3 SIMO

4 times 3 MIMO

4 times 4 MIMO


(d)



11

1

1

11

2

2

2

3

3

4

4

079083064

12

12

119

312

171

15

25

171

121

101 165157

Rece

ive p

orts

(nR

)

Transmit ports (nT)

075

(a)

111 1 1

1

2

2

2

3

3

Transmit ports (nT)

12

12

14

14

16

22

18

167

102 248

082

089 084

136

167Re

ceiv

e por

ts (n

R)

(b)

11

1

2

3

4 365101

101

187184

174

174 174 174

189

093

088 092094

189

1

1

15

15

252 3


Rece

ive p

orts

(nR

)

(c)


1 1

1

1

2

1

2

3

178

178 267

089

132

093

074

15

Rece

ive p

orts

(nR

)

(d)









6 Conclusions






References























































































































































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










where

1205831198941198950 =

sum119869minus1

119901=011990110038161003816100381610038161003816ℎ+

119894119895119901

10038161003816100381610038161003816

2

sum119869minus1

119901=0

10038161003816100381610038161003816ℎ+

119894119895119901

10038161003816100381610038161003816

2 119894 119895 = 1 119899 (15a)

120583(2)

1198941198950=

sum119869minus1

119901=0119901210038161003816100381610038161003816ℎ+

119894119895119901

10038161003816100381610038161003816

2

sum119869minus1

119901=0

10038161003816100381610038161003816ℎ+

119894119895119901

10038161003816100381610038161003816

2 119894 119895 = 1 119899 (15b)


119861119894119895119888 (Δ119891) =

119864 119867+

119894119895119902sdot [119867+

119894119895(119891 = 119902119891119904 + Δ119891)]

lowast

119864 10038161003816100381610038161003816119867+

119894119895119902

10038161003816100381610038161003816

2

119902 = 0 1 lfloor(119869 minus 2) +Δ119891

119891119904

rfloor 119894 119895 = 1 119899

(16)



119862MIMO

= 119891119904

119869minus1

sum

119902=0

min119899119879119899119877

sum

119894=1

log21 +

SNR (119902119891119904)119899119879

sdot

1003816100381610038161003816100381610038161003816

119867

119898

119894(119902119891119904)

1003816100381610038161003816100381610038161003816

2

(17)

where

SNR (119891) =119901 (119891)

119873 (119891)(18)







































V+in

L2L1 Lk LN+1

A998400N

A998400kA998400

2A9984001


Vminusout


A1 A2 Ak AN




0

0

50

50

100

100

150

Frequency (MHz)

Chan

nel a

ttenu

atio

n (d

B)

(a)

0 50 100

Frequency (MHz)

0

50

100

150

Chan

nel a

ttenu

atio

n (d

B)

(b)


0

50

100

150



Suburban case

Chan

nel a

ttenu

atio

n (d

B)

(c)


0

50

100

150



Suburban case

Chan

nel a

ttenu

atio

n (d

B)

(d)






















minus1 0

0

1minus70

minus60

minus50

minus40

minus30

minus20

minus10

ACG

(dB)

Overhead XC

Overhead CC

(a)

1

08

06

04

02


ACG (dB)

Overhead XC

Overhead CC

ACG

CD

F (A

CGlt

absc

issa)

(b)

0

minus70

minus60

minus50

minus40

minus30

minus20

minus10

ACG

(dB)


Underground XC

Underground CC


(c)

1

08

06

04

02


ACG (dB)

Underground XC

Underground CC

ACG

CD

F (A

CGlt

absc

issa)


(d)





















Overhead XC

Overhead CC


0

05

15

2

1

minus2 0 2

RMS-

DS

(120583s)

(a)

RMS-DS (120583s)0 1 2

1

08

06

04

02

0

Overhead XC

Overhead CC

RMS-

DS

CDF

(RM

S-D

Slt

absc

issa)

(b)

0

05

15

2

1

RMS-

DS

(120583s)


Underground XC

Underground XC

Underground CC


(c)

1

08

06

04

02

0

RMS-

DS

CDF

(RM

S-D

Slt

absc

issa)

RMS-DS (120583s)0 1 2

Underground XC

Underground CC


(d)






LV 009 026 112 083 068 053 042 020MV 038 061 108 146 078 100 030 034

XCLV 011 011 192 078 106 050 066 024MV 011 056 184 159 104 106 063 037

AllLV 011 011 192 078 106 050 066 025MV 028 056 181 159 111 106 055 037





















0

05

15

2

1

RMS-

DS

(120583s)


ACG (dB)




(a)

0

05

15

2

1

RMS-

DS

(120583s)





ACG (dB)

(b)


0 10 20 30 40 50CB (MHz)



0

05

15

2

1

RMS-

DS

(120583s)

(a)

0 10 20 30 40 50CB (MHz)



0

05

15

2

1RM

S-D

S (120583

s)

(b)
















LV 070 080 4950 37 1182 1232 2114 1631MV 080 050 4760 560 1178 232 2016 228

XCLV 020 040 3150 48 700 1020 1372 2113MV 020 030 3220 880 716 206 1402 377

AllLV 020 040 3250 48 720 1020 1416 2113MV 020 030 27 880 576 206 1188 377














00

500

1000

2000

1500

Cum

ulat

ive c

apac

ity (M

bps)

Overhead LVXC

Overhead LVCC

(a)

0

500

1000

2000

1500

Cum

ulat

ive c

apac

ity (M

bps)


Overhead MVCC

Overhead MVXC

(b)

0

Capa

city

CCD

F (c

apac

itygt

absc

issa)

01

02

03

04

05

06

07

08

09

1

0 500 1000 1500 2000

Overhead LVCC

Overhead MVCC

Overhead MVXC

Overhead LVXC

Capacity (Mbps)

(c)

0

500

1000

2000

1500

Cum

ulat

ive c

apac

ity (M

bps)

UndergroundLVCC

UndergroundLVXC



Suburban case


0

(d)

0

500

1000

2000

1500

Cum

ulat

ive c

apac

ity (M

bps)

Underground MVXC

UndergroundMVCC



Suburban case


0

(e)

0 500 1000 1500 20000

01

02

03

04

05

06

07

08

09

1

UndergroundLVCC

UndergroundLVXC

UndergroundMVCC

Underground

Capacity (Mbps)

Capa

city

CCD

F (c

apac

itygt

absc

issa)

MVXC

(f)











0

500

1000

2000

3000

1500

2500

0

500

1000

2000

3000

1500

2500

Cum

ulat

ive c

apac

ity (M

bps)

0

500

1000

2000

3000

1500

2500

Cum

ulat

ive c

apac

ity (M

bps)

0

500

1000

2000

3000

1500

2500

Cum

ulat

ive c

apac

ity (M

bps)

0

500

1000

2000

3000

1500

2500

Cum

ulat

ive c

apac

ity (M

bps)

0

500

1000

2000

3000

1500

2500

Cum

ulat

ive c

apac

ity (M

bps)

Cum

ulat

ive c

apac

ity (M

bps)

00

1000 2000 3000

Capa

city

CCD

F (c

apac

itygt

absc

issa)

01

02

03

04

05

06

07

08

09

1

(g)Capacity (Mbps)

0 1000 2000 3000C (Mb )




4 times 1 MISOLV



Suburban case



Suburban case

(a) (b)

(c)

(e) (f)

(d)







2sdot presence in












0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)

00

1000

2000

2000

3000

4000

4000

5000

6000

6000

7000

8000

8000

Cum

ulat

ive c

apac

ity (M

bps)

0

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)






Capa

city

CCD

F (c

apac

itygt

absc

issa)

01

02

03

04

05

06

07

08

09


3 times 1 MISOMV

4 times 4 MIMOLV

3 times 3 MIMOMV

SISOCCMVSISOXCMV

(a) (b)

(c) (d)

(e) (f)

(g)Capacity (Mbps)



Suburban case



Suburban case

1 times 3 SIMOMV






LV 471 1634 904 2152 700 1928 168 200 4MV 459 664 892 1019 688 855 168 137 3

1 times 2 SIMOLV 345 1497 693 1999 530 1785 140 194 24MV 375 471 744 769 572 632 149 114 9

1 times 3 SIMOLV 476 1646 907 2161 704 1938 168 198 16MV 477 664 906 1019 705 855 168 137 3


2 times 1 MISOLV 286 1397 604 1896 453 1684 128 193 24MV 399 613 813 951 616 794 161 1303 9

3 times 1 MISOLV 370 1511 772 2026 579 1804 156 198 16MV 370 583 770 912 579 760 156 127 3


2 times 2 MIMOLV 537 2770 1114 3779 854 3349 241 390 36MV 602 828 1261 1381 949 1129 264 212 9

2 times 3 MIMOLV 536 2770 1137 3779 851 3349 240 390 24MV 736 1166 1538 1825 1155 1520 312 253 3


3 times 2 MIMOLV 530 2770 1135 3779 847 3349 243 390 24MV 732 1166 1535 1824 1151 1520 312 253 3

3 times 3 MIMOLV 774 3058 1591 4092 1201 3645 317 398 16MV 1082 1750 2288 2736 1710 2280 468 380 1



Table 4 Continued













119894119895 119894 isin 119873119879 119895 isin 119873119877 [130ndash133]




0

500

1000

2000

3000

1500

2500Cu

mul

ativ

e cap

acity

(Mbp

s)


1 times 4 SIMO

4 times 4 MIMO


3 times 2 MIMO

1 times 3 SIMO

SISOCCLV SISOCCMV

(a)

1

08

09

06

07

04

05

03

01

02

00 1000 2000 3000


SISOCCLV

SISOCCMV

Cum

ulat

ive c

apac

ity C

DF

(cum

ulat

ive c

apac

itylt

absc

issa)

4 times 4 MIMO

4 times 3 MIMO

1 times 3 SIMO

1 times 4 SIMO

3 times 2 MIMO

3 times 3 MIMO

(b)

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)


SISOCCLV

SISOCCMVminus1

4 times 4 MIMO

4 times 3 MIMO

3 times 3 MIMO

1 times 3 SIMO

1 times 4 SIMO

3 times 2 MIMO


(c)


08

09

1

06

07

04

05

03

01

02

0

SISOCCLV

SISOCCMV

Cum

ulat

ive c

apac

ity C

DF

(cum

ulat

ive c

apac

itylt

absc

issa)

3 times 3 MIMO

3 times 2 MIMO

1 times 4 SIMO

1 times 3 SIMO

4 times 3 MIMO

4 times 4 MIMO


(d)



11

1

1

11

2

2

2

3

3

4

4

079083064

12

12

119

312

171

15

25

171

121

101 165157

Rece

ive p

orts

(nR

)

Transmit ports (nT)

075

(a)

111 1 1

1

2

2

2

3

3

Transmit ports (nT)

12

12

14

14

16

22

18

167

102 248

082

089 084

136

167Re

ceiv

e por

ts (n

R)

(b)

11

1

2

3

4 365101

101

187184

174

174 174 174

189

093

088 092094

189

1

1

15

15

252 3


Rece

ive p

orts

(nR

)

(c)


1 1

1

1

2

1

2

3

178

178 267

089

132

093

074

15

Rece

ive p

orts

(nR

)

(d)









6 Conclusions






References























































































































































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation
































V+in

L2L1 Lk LN+1

A998400N

A998400kA998400

2A9984001


Vminusout


A1 A2 Ak AN




0

0

50

50

100

100

150

Frequency (MHz)

Chan

nel a

ttenu

atio

n (d

B)

(a)

0 50 100

Frequency (MHz)

0

50

100

150

Chan

nel a

ttenu

atio

n (d

B)

(b)


0

50

100

150



Suburban case

Chan

nel a

ttenu

atio

n (d

B)

(c)


0

50

100

150



Suburban case

Chan

nel a

ttenu

atio

n (d

B)

(d)






















minus1 0

0

1minus70

minus60

minus50

minus40

minus30

minus20

minus10

ACG

(dB)

Overhead XC

Overhead CC

(a)

1

08

06

04

02


ACG (dB)

Overhead XC

Overhead CC

ACG

CD

F (A

CGlt

absc

issa)

(b)

0

minus70

minus60

minus50

minus40

minus30

minus20

minus10

ACG

(dB)


Underground XC

Underground CC


(c)

1

08

06

04

02


ACG (dB)

Underground XC

Underground CC

ACG

CD

F (A

CGlt

absc

issa)


(d)





















Overhead XC

Overhead CC


0

05

15

2

1

minus2 0 2

RMS-

DS

(120583s)

(a)

RMS-DS (120583s)0 1 2

1

08

06

04

02

0

Overhead XC

Overhead CC

RMS-

DS

CDF

(RM

S-D

Slt

absc

issa)

(b)

0

05

15

2

1

RMS-

DS

(120583s)


Underground XC

Underground XC

Underground CC


(c)

1

08

06

04

02

0

RMS-

DS

CDF

(RM

S-D

Slt

absc

issa)

RMS-DS (120583s)0 1 2

Underground XC

Underground CC


(d)






LV 009 026 112 083 068 053 042 020MV 038 061 108 146 078 100 030 034

XCLV 011 011 192 078 106 050 066 024MV 011 056 184 159 104 106 063 037

AllLV 011 011 192 078 106 050 066 025MV 028 056 181 159 111 106 055 037





















0

05

15

2

1

RMS-

DS

(120583s)


ACG (dB)




(a)

0

05

15

2

1

RMS-

DS

(120583s)





ACG (dB)

(b)


0 10 20 30 40 50CB (MHz)



0

05

15

2

1

RMS-

DS

(120583s)

(a)

0 10 20 30 40 50CB (MHz)



0

05

15

2

1RM

S-D

S (120583

s)

(b)
















LV 070 080 4950 37 1182 1232 2114 1631MV 080 050 4760 560 1178 232 2016 228

XCLV 020 040 3150 48 700 1020 1372 2113MV 020 030 3220 880 716 206 1402 377

AllLV 020 040 3250 48 720 1020 1416 2113MV 020 030 27 880 576 206 1188 377














00

500

1000

2000

1500

Cum

ulat

ive c

apac

ity (M

bps)

Overhead LVXC

Overhead LVCC

(a)

0

500

1000

2000

1500

Cum

ulat

ive c

apac

ity (M

bps)


Overhead MVCC

Overhead MVXC

(b)

0

Capa

city

CCD

F (c

apac

itygt

absc

issa)

01

02

03

04

05

06

07

08

09

1

0 500 1000 1500 2000

Overhead LVCC

Overhead MVCC

Overhead MVXC

Overhead LVXC

Capacity (Mbps)

(c)

0

500

1000

2000

1500

Cum

ulat

ive c

apac

ity (M

bps)

UndergroundLVCC

UndergroundLVXC



Suburban case


0

(d)

0

500

1000

2000

1500

Cum

ulat

ive c

apac

ity (M

bps)

Underground MVXC

UndergroundMVCC



Suburban case


0

(e)

0 500 1000 1500 20000

01

02

03

04

05

06

07

08

09

1

UndergroundLVCC

UndergroundLVXC

UndergroundMVCC

Underground

Capacity (Mbps)

Capa

city

CCD

F (c

apac

itygt

absc

issa)

MVXC

(f)











0

500

1000

2000

3000

1500

2500

0

500

1000

2000

3000

1500

2500

Cum

ulat

ive c

apac

ity (M

bps)

0

500

1000

2000

3000

1500

2500

Cum

ulat

ive c

apac

ity (M

bps)

0

500

1000

2000

3000

1500

2500

Cum

ulat

ive c

apac

ity (M

bps)

0

500

1000

2000

3000

1500

2500

Cum

ulat

ive c

apac

ity (M

bps)

0

500

1000

2000

3000

1500

2500

Cum

ulat

ive c

apac

ity (M

bps)

Cum

ulat

ive c

apac

ity (M

bps)

00

1000 2000 3000

Capa

city

CCD

F (c

apac

itygt

absc

issa)

01

02

03

04

05

06

07

08

09

1

(g)Capacity (Mbps)

0 1000 2000 3000C (Mb )




4 times 1 MISOLV



Suburban case



Suburban case

(a) (b)

(c)

(e) (f)

(d)







2sdot presence in












0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)

00

1000

2000

2000

3000

4000

4000

5000

6000

6000

7000

8000

8000

Cum

ulat

ive c

apac

ity (M

bps)

0

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)






Capa

city

CCD

F (c

apac

itygt

absc

issa)

01

02

03

04

05

06

07

08

09


3 times 1 MISOMV

4 times 4 MIMOLV

3 times 3 MIMOMV

SISOCCMVSISOXCMV

(a) (b)

(c) (d)

(e) (f)

(g)Capacity (Mbps)



Suburban case



Suburban case

1 times 3 SIMOMV






LV 471 1634 904 2152 700 1928 168 200 4MV 459 664 892 1019 688 855 168 137 3

1 times 2 SIMOLV 345 1497 693 1999 530 1785 140 194 24MV 375 471 744 769 572 632 149 114 9

1 times 3 SIMOLV 476 1646 907 2161 704 1938 168 198 16MV 477 664 906 1019 705 855 168 137 3


2 times 1 MISOLV 286 1397 604 1896 453 1684 128 193 24MV 399 613 813 951 616 794 161 1303 9

3 times 1 MISOLV 370 1511 772 2026 579 1804 156 198 16MV 370 583 770 912 579 760 156 127 3


2 times 2 MIMOLV 537 2770 1114 3779 854 3349 241 390 36MV 602 828 1261 1381 949 1129 264 212 9

2 times 3 MIMOLV 536 2770 1137 3779 851 3349 240 390 24MV 736 1166 1538 1825 1155 1520 312 253 3


3 times 2 MIMOLV 530 2770 1135 3779 847 3349 243 390 24MV 732 1166 1535 1824 1151 1520 312 253 3

3 times 3 MIMOLV 774 3058 1591 4092 1201 3645 317 398 16MV 1082 1750 2288 2736 1710 2280 468 380 1



Table 4 Continued













119894119895 119894 isin 119873119879 119895 isin 119873119877 [130ndash133]




0

500

1000

2000

3000

1500

2500Cu

mul

ativ

e cap

acity

(Mbp

s)


1 times 4 SIMO

4 times 4 MIMO


3 times 2 MIMO

1 times 3 SIMO

SISOCCLV SISOCCMV

(a)

1

08

09

06

07

04

05

03

01

02

00 1000 2000 3000


SISOCCLV

SISOCCMV

Cum

ulat

ive c

apac

ity C

DF

(cum

ulat

ive c

apac

itylt

absc

issa)

4 times 4 MIMO

4 times 3 MIMO

1 times 3 SIMO

1 times 4 SIMO

3 times 2 MIMO

3 times 3 MIMO

(b)

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)


SISOCCLV

SISOCCMVminus1

4 times 4 MIMO

4 times 3 MIMO

3 times 3 MIMO

1 times 3 SIMO

1 times 4 SIMO

3 times 2 MIMO


(c)


08

09

1

06

07

04

05

03

01

02

0

SISOCCLV

SISOCCMV

Cum

ulat

ive c

apac

ity C

DF

(cum

ulat

ive c

apac

itylt

absc

issa)

3 times 3 MIMO

3 times 2 MIMO

1 times 4 SIMO

1 times 3 SIMO

4 times 3 MIMO

4 times 4 MIMO


(d)



11

1

1

11

2

2

2

3

3

4

4

079083064

12

12

119

312

171

15

25

171

121

101 165157

Rece

ive p

orts

(nR

)

Transmit ports (nT)

075

(a)

111 1 1

1

2

2

2

3

3

Transmit ports (nT)

12

12

14

14

16

22

18

167

102 248

082

089 084

136

167Re

ceiv

e por

ts (n

R)

(b)

11

1

2

3

4 365101

101

187184

174

174 174 174

189

093

088 092094

189

1

1

15

15

252 3


Rece

ive p

orts

(nR

)

(c)


1 1

1

1

2

1

2

3

178

178 267

089

132

093

074

15

Rece

ive p

orts

(nR

)

(d)









6 Conclusions






References























































































































































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation





























minus1 0

0

1minus70

minus60

minus50

minus40

minus30

minus20

minus10

ACG

(dB)

Overhead XC

Overhead CC

(a)

1

08

06

04

02


ACG (dB)

Overhead XC

Overhead CC

ACG

CD

F (A

CGlt

absc

issa)

(b)

0

minus70

minus60

minus50

minus40

minus30

minus20

minus10

ACG

(dB)


Underground XC

Underground CC


(c)

1

08

06

04

02


ACG (dB)

Underground XC

Underground CC

ACG

CD

F (A

CGlt

absc

issa)


(d)





















Overhead XC

Overhead CC


0

05

15

2

1

minus2 0 2

RMS-

DS

(120583s)

(a)

RMS-DS (120583s)0 1 2

1

08

06

04

02

0

Overhead XC

Overhead CC

RMS-

DS

CDF

(RM

S-D

Slt

absc

issa)

(b)

0

05

15

2

1

RMS-

DS

(120583s)


Underground XC

Underground XC

Underground CC


(c)

1

08

06

04

02

0

RMS-

DS

CDF

(RM

S-D

Slt

absc

issa)

RMS-DS (120583s)0 1 2

Underground XC

Underground CC


(d)






LV 009 026 112 083 068 053 042 020MV 038 061 108 146 078 100 030 034

XCLV 011 011 192 078 106 050 066 024MV 011 056 184 159 104 106 063 037

AllLV 011 011 192 078 106 050 066 025MV 028 056 181 159 111 106 055 037





















0

05

15

2

1

RMS-

DS

(120583s)


ACG (dB)




(a)

0

05

15

2

1

RMS-

DS

(120583s)





ACG (dB)

(b)


0 10 20 30 40 50CB (MHz)



0

05

15

2

1

RMS-

DS

(120583s)

(a)

0 10 20 30 40 50CB (MHz)



0

05

15

2

1RM

S-D

S (120583

s)

(b)
















LV 070 080 4950 37 1182 1232 2114 1631MV 080 050 4760 560 1178 232 2016 228

XCLV 020 040 3150 48 700 1020 1372 2113MV 020 030 3220 880 716 206 1402 377

AllLV 020 040 3250 48 720 1020 1416 2113MV 020 030 27 880 576 206 1188 377














00

500

1000

2000

1500

Cum

ulat

ive c

apac

ity (M

bps)

Overhead LVXC

Overhead LVCC

(a)

0

500

1000

2000

1500

Cum

ulat

ive c

apac

ity (M

bps)


Overhead MVCC

Overhead MVXC

(b)

0

Capa

city

CCD

F (c

apac

itygt

absc

issa)

01

02

03

04

05

06

07

08

09

1

0 500 1000 1500 2000

Overhead LVCC

Overhead MVCC

Overhead MVXC

Overhead LVXC

Capacity (Mbps)

(c)

0

500

1000

2000

1500

Cum

ulat

ive c

apac

ity (M

bps)

UndergroundLVCC

UndergroundLVXC



Suburban case


0

(d)

0

500

1000

2000

1500

Cum

ulat

ive c

apac

ity (M

bps)

Underground MVXC

UndergroundMVCC



Suburban case


0

(e)

0 500 1000 1500 20000

01

02

03

04

05

06

07

08

09

1

UndergroundLVCC

UndergroundLVXC

UndergroundMVCC

Underground

Capacity (Mbps)

Capa

city

CCD

F (c

apac

itygt

absc

issa)

MVXC

(f)











0

500

1000

2000

3000

1500

2500

0

500

1000

2000

3000

1500

2500

Cum

ulat

ive c

apac

ity (M

bps)

0

500

1000

2000

3000

1500

2500

Cum

ulat

ive c

apac

ity (M

bps)

0

500

1000

2000

3000

1500

2500

Cum

ulat

ive c

apac

ity (M

bps)

0

500

1000

2000

3000

1500

2500

Cum

ulat

ive c

apac

ity (M

bps)

0

500

1000

2000

3000

1500

2500

Cum

ulat

ive c

apac

ity (M

bps)

Cum

ulat

ive c

apac

ity (M

bps)

00

1000 2000 3000

Capa

city

CCD

F (c

apac

itygt

absc

issa)

01

02

03

04

05

06

07

08

09

1

(g)Capacity (Mbps)

0 1000 2000 3000C (Mb )




4 times 1 MISOLV



Suburban case



Suburban case

(a) (b)

(c)

(e) (f)

(d)







2sdot presence in












0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)

00

1000

2000

2000

3000

4000

4000

5000

6000

6000

7000

8000

8000

Cum

ulat

ive c

apac

ity (M

bps)

0

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)






Capa

city

CCD

F (c

apac

itygt

absc

issa)

01

02

03

04

05

06

07

08

09


3 times 1 MISOMV

4 times 4 MIMOLV

3 times 3 MIMOMV

SISOCCMVSISOXCMV

(a) (b)

(c) (d)

(e) (f)

(g)Capacity (Mbps)



Suburban case



Suburban case

1 times 3 SIMOMV






LV 471 1634 904 2152 700 1928 168 200 4MV 459 664 892 1019 688 855 168 137 3

1 times 2 SIMOLV 345 1497 693 1999 530 1785 140 194 24MV 375 471 744 769 572 632 149 114 9

1 times 3 SIMOLV 476 1646 907 2161 704 1938 168 198 16MV 477 664 906 1019 705 855 168 137 3


2 times 1 MISOLV 286 1397 604 1896 453 1684 128 193 24MV 399 613 813 951 616 794 161 1303 9

3 times 1 MISOLV 370 1511 772 2026 579 1804 156 198 16MV 370 583 770 912 579 760 156 127 3


2 times 2 MIMOLV 537 2770 1114 3779 854 3349 241 390 36MV 602 828 1261 1381 949 1129 264 212 9

2 times 3 MIMOLV 536 2770 1137 3779 851 3349 240 390 24MV 736 1166 1538 1825 1155 1520 312 253 3


3 times 2 MIMOLV 530 2770 1135 3779 847 3349 243 390 24MV 732 1166 1535 1824 1151 1520 312 253 3

3 times 3 MIMOLV 774 3058 1591 4092 1201 3645 317 398 16MV 1082 1750 2288 2736 1710 2280 468 380 1



Table 4 Continued













119894119895 119894 isin 119873119879 119895 isin 119873119877 [130ndash133]




0

500

1000

2000

3000

1500

2500Cu

mul

ativ

e cap

acity

(Mbp

s)


1 times 4 SIMO

4 times 4 MIMO


3 times 2 MIMO

1 times 3 SIMO

SISOCCLV SISOCCMV

(a)

1

08

09

06

07

04

05

03

01

02

00 1000 2000 3000


SISOCCLV

SISOCCMV

Cum

ulat

ive c

apac

ity C

DF

(cum

ulat

ive c

apac

itylt

absc

issa)

4 times 4 MIMO

4 times 3 MIMO

1 times 3 SIMO

1 times 4 SIMO

3 times 2 MIMO

3 times 3 MIMO

(b)

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)


SISOCCLV

SISOCCMVminus1

4 times 4 MIMO

4 times 3 MIMO

3 times 3 MIMO

1 times 3 SIMO

1 times 4 SIMO

3 times 2 MIMO


(c)


08

09

1

06

07

04

05

03

01

02

0

SISOCCLV

SISOCCMV

Cum

ulat

ive c

apac

ity C

DF

(cum

ulat

ive c

apac

itylt

absc

issa)

3 times 3 MIMO

3 times 2 MIMO

1 times 4 SIMO

1 times 3 SIMO

4 times 3 MIMO

4 times 4 MIMO


(d)



11

1

1

11

2

2

2

3

3

4

4

079083064

12

12

119

312

171

15

25

171

121

101 165157

Rece

ive p

orts

(nR

)

Transmit ports (nT)

075

(a)

111 1 1

1

2

2

2

3

3

Transmit ports (nT)

12

12

14

14

16

22

18

167

102 248

082

089 084

136

167Re

ceiv

e por

ts (n

R)

(b)

11

1

2

3

4 365101

101

187184

174

174 174 174

189

093

088 092094

189

1

1

15

15

252 3


Rece

ive p

orts

(nR

)

(c)


1 1

1

1

2

1

2

3

178

178 267

089

132

093

074

15

Rece

ive p

orts

(nR

)

(d)









6 Conclusions






References























































































































































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation

























Overhead XC

Overhead CC


0

05

15

2

1

minus2 0 2

RMS-

DS

(120583s)

(a)

RMS-DS (120583s)0 1 2

1

08

06

04

02

0

Overhead XC

Overhead CC

RMS-

DS

CDF

(RM

S-D

Slt

absc

issa)

(b)

0

05

15

2

1

RMS-

DS

(120583s)


Underground XC

Underground XC

Underground CC


(c)

1

08

06

04

02

0

RMS-

DS

CDF

(RM

S-D

Slt

absc

issa)

RMS-DS (120583s)0 1 2

Underground XC

Underground CC


(d)






LV 009 026 112 083 068 053 042 020MV 038 061 108 146 078 100 030 034

XCLV 011 011 192 078 106 050 066 024MV 011 056 184 159 104 106 063 037

AllLV 011 011 192 078 106 050 066 025MV 028 056 181 159 111 106 055 037





















0

05

15

2

1

RMS-

DS

(120583s)


ACG (dB)




(a)

0

05

15

2

1

RMS-

DS

(120583s)





ACG (dB)

(b)


0 10 20 30 40 50CB (MHz)



0

05

15

2

1

RMS-

DS

(120583s)

(a)

0 10 20 30 40 50CB (MHz)



0

05

15

2

1RM

S-D

S (120583

s)

(b)
















LV 070 080 4950 37 1182 1232 2114 1631MV 080 050 4760 560 1178 232 2016 228

XCLV 020 040 3150 48 700 1020 1372 2113MV 020 030 3220 880 716 206 1402 377

AllLV 020 040 3250 48 720 1020 1416 2113MV 020 030 27 880 576 206 1188 377














00

500

1000

2000

1500

Cum

ulat

ive c

apac

ity (M

bps)

Overhead LVXC

Overhead LVCC

(a)

0

500

1000

2000

1500

Cum

ulat

ive c

apac

ity (M

bps)


Overhead MVCC

Overhead MVXC

(b)

0

Capa

city

CCD

F (c

apac

itygt

absc

issa)

01

02

03

04

05

06

07

08

09

1

0 500 1000 1500 2000

Overhead LVCC

Overhead MVCC

Overhead MVXC

Overhead LVXC

Capacity (Mbps)

(c)

0

500

1000

2000

1500

Cum

ulat

ive c

apac

ity (M

bps)

UndergroundLVCC

UndergroundLVXC



Suburban case


0

(d)

0

500

1000

2000

1500

Cum

ulat

ive c

apac

ity (M

bps)

Underground MVXC

UndergroundMVCC



Suburban case


0

(e)

0 500 1000 1500 20000

01

02

03

04

05

06

07

08

09

1

UndergroundLVCC

UndergroundLVXC

UndergroundMVCC

Underground

Capacity (Mbps)

Capa

city

CCD

F (c

apac

itygt

absc

issa)

MVXC

(f)











0

500

1000

2000

3000

1500

2500

0

500

1000

2000

3000

1500

2500

Cum

ulat

ive c

apac

ity (M

bps)

0

500

1000

2000

3000

1500

2500

Cum

ulat

ive c

apac

ity (M

bps)

0

500

1000

2000

3000

1500

2500

Cum

ulat

ive c

apac

ity (M

bps)

0

500

1000

2000

3000

1500

2500

Cum

ulat

ive c

apac

ity (M

bps)

0

500

1000

2000

3000

1500

2500

Cum

ulat

ive c

apac

ity (M

bps)

Cum

ulat

ive c

apac

ity (M

bps)

00

1000 2000 3000

Capa

city

CCD

F (c

apac

itygt

absc

issa)

01

02

03

04

05

06

07

08

09

1

(g)Capacity (Mbps)

0 1000 2000 3000C (Mb )




4 times 1 MISOLV



Suburban case



Suburban case

(a) (b)

(c)

(e) (f)

(d)







2sdot presence in












0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)

00

1000

2000

2000

3000

4000

4000

5000

6000

6000

7000

8000

8000

Cum

ulat

ive c

apac

ity (M

bps)

0

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)






Capa

city

CCD

F (c

apac

itygt

absc

issa)

01

02

03

04

05

06

07

08

09


3 times 1 MISOMV

4 times 4 MIMOLV

3 times 3 MIMOMV

SISOCCMVSISOXCMV

(a) (b)

(c) (d)

(e) (f)

(g)Capacity (Mbps)



Suburban case



Suburban case

1 times 3 SIMOMV






LV 471 1634 904 2152 700 1928 168 200 4MV 459 664 892 1019 688 855 168 137 3

1 times 2 SIMOLV 345 1497 693 1999 530 1785 140 194 24MV 375 471 744 769 572 632 149 114 9

1 times 3 SIMOLV 476 1646 907 2161 704 1938 168 198 16MV 477 664 906 1019 705 855 168 137 3


2 times 1 MISOLV 286 1397 604 1896 453 1684 128 193 24MV 399 613 813 951 616 794 161 1303 9

3 times 1 MISOLV 370 1511 772 2026 579 1804 156 198 16MV 370 583 770 912 579 760 156 127 3


2 times 2 MIMOLV 537 2770 1114 3779 854 3349 241 390 36MV 602 828 1261 1381 949 1129 264 212 9

2 times 3 MIMOLV 536 2770 1137 3779 851 3349 240 390 24MV 736 1166 1538 1825 1155 1520 312 253 3


3 times 2 MIMOLV 530 2770 1135 3779 847 3349 243 390 24MV 732 1166 1535 1824 1151 1520 312 253 3

3 times 3 MIMOLV 774 3058 1591 4092 1201 3645 317 398 16MV 1082 1750 2288 2736 1710 2280 468 380 1



Table 4 Continued













119894119895 119894 isin 119873119879 119895 isin 119873119877 [130ndash133]




0

500

1000

2000

3000

1500

2500Cu

mul

ativ

e cap

acity

(Mbp

s)


1 times 4 SIMO

4 times 4 MIMO


3 times 2 MIMO

1 times 3 SIMO

SISOCCLV SISOCCMV

(a)

1

08

09

06

07

04

05

03

01

02

00 1000 2000 3000


SISOCCLV

SISOCCMV

Cum

ulat

ive c

apac

ity C

DF

(cum

ulat

ive c

apac

itylt

absc

issa)

4 times 4 MIMO

4 times 3 MIMO

1 times 3 SIMO

1 times 4 SIMO

3 times 2 MIMO

3 times 3 MIMO

(b)

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)


SISOCCLV

SISOCCMVminus1

4 times 4 MIMO

4 times 3 MIMO

3 times 3 MIMO

1 times 3 SIMO

1 times 4 SIMO

3 times 2 MIMO


(c)


08

09

1

06

07

04

05

03

01

02

0

SISOCCLV

SISOCCMV

Cum

ulat

ive c

apac

ity C

DF

(cum

ulat

ive c

apac

itylt

absc

issa)

3 times 3 MIMO

3 times 2 MIMO

1 times 4 SIMO

1 times 3 SIMO

4 times 3 MIMO

4 times 4 MIMO


(d)



11

1

1

11

2

2

2

3

3

4

4

079083064

12

12

119

312

171

15

25

171

121

101 165157

Rece

ive p

orts

(nR

)

Transmit ports (nT)

075

(a)

111 1 1

1

2

2

2

3

3

Transmit ports (nT)

12

12

14

14

16

22

18

167

102 248

082

089 084

136

167Re

ceiv

e por

ts (n

R)

(b)

11

1

2

3

4 365101

101

187184

174

174 174 174

189

093

088 092094

189

1

1

15

15

252 3


Rece

ive p

orts

(nR

)

(c)


1 1

1

1

2

1

2

3

178

178 267

089

132

093

074

15

Rece

ive p

orts

(nR

)

(d)









6 Conclusions






References























































































































































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation













LV 009 026 112 083 068 053 042 020MV 038 061 108 146 078 100 030 034

XCLV 011 011 192 078 106 050 066 024MV 011 056 184 159 104 106 063 037

AllLV 011 011 192 078 106 050 066 025MV 028 056 181 159 111 106 055 037





















0

05

15

2

1

RMS-

DS

(120583s)


ACG (dB)




(a)

0

05

15

2

1

RMS-

DS

(120583s)





ACG (dB)

(b)


0 10 20 30 40 50CB (MHz)



0

05

15

2

1

RMS-

DS

(120583s)

(a)

0 10 20 30 40 50CB (MHz)



0

05

15

2

1RM

S-D

S (120583

s)

(b)
















LV 070 080 4950 37 1182 1232 2114 1631MV 080 050 4760 560 1178 232 2016 228

XCLV 020 040 3150 48 700 1020 1372 2113MV 020 030 3220 880 716 206 1402 377

AllLV 020 040 3250 48 720 1020 1416 2113MV 020 030 27 880 576 206 1188 377














00

500

1000

2000

1500

Cum

ulat

ive c

apac

ity (M

bps)

Overhead LVXC

Overhead LVCC

(a)

0

500

1000

2000

1500

Cum

ulat

ive c

apac

ity (M

bps)


Overhead MVCC

Overhead MVXC

(b)

0

Capa

city

CCD

F (c

apac

itygt

absc

issa)

01

02

03

04

05

06

07

08

09

1

0 500 1000 1500 2000

Overhead LVCC

Overhead MVCC

Overhead MVXC

Overhead LVXC

Capacity (Mbps)

(c)

0

500

1000

2000

1500

Cum

ulat

ive c

apac

ity (M

bps)

UndergroundLVCC

UndergroundLVXC



Suburban case


0

(d)

0

500

1000

2000

1500

Cum

ulat

ive c

apac

ity (M

bps)

Underground MVXC

UndergroundMVCC



Suburban case


0

(e)

0 500 1000 1500 20000

01

02

03

04

05

06

07

08

09

1

UndergroundLVCC

UndergroundLVXC

UndergroundMVCC

Underground

Capacity (Mbps)

Capa

city

CCD

F (c

apac

itygt

absc

issa)

MVXC

(f)











0

500

1000

2000

3000

1500

2500

0

500

1000

2000

3000

1500

2500

Cum

ulat

ive c

apac

ity (M

bps)

0

500

1000

2000

3000

1500

2500

Cum

ulat

ive c

apac

ity (M

bps)

0

500

1000

2000

3000

1500

2500

Cum

ulat

ive c

apac

ity (M

bps)

0

500

1000

2000

3000

1500

2500

Cum

ulat

ive c

apac

ity (M

bps)

0

500

1000

2000

3000

1500

2500

Cum

ulat

ive c

apac

ity (M

bps)

Cum

ulat

ive c

apac

ity (M

bps)

00

1000 2000 3000

Capa

city

CCD

F (c

apac

itygt

absc

issa)

01

02

03

04

05

06

07

08

09

1

(g)Capacity (Mbps)

0 1000 2000 3000C (Mb )




4 times 1 MISOLV



Suburban case



Suburban case

(a) (b)

(c)

(e) (f)

(d)







2sdot presence in












0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)

00

1000

2000

2000

3000

4000

4000

5000

6000

6000

7000

8000

8000

Cum

ulat

ive c

apac

ity (M

bps)

0

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)






Capa

city

CCD

F (c

apac

itygt

absc

issa)

01

02

03

04

05

06

07

08

09


3 times 1 MISOMV

4 times 4 MIMOLV

3 times 3 MIMOMV

SISOCCMVSISOXCMV

(a) (b)

(c) (d)

(e) (f)

(g)Capacity (Mbps)



Suburban case



Suburban case

1 times 3 SIMOMV






LV 471 1634 904 2152 700 1928 168 200 4MV 459 664 892 1019 688 855 168 137 3

1 times 2 SIMOLV 345 1497 693 1999 530 1785 140 194 24MV 375 471 744 769 572 632 149 114 9

1 times 3 SIMOLV 476 1646 907 2161 704 1938 168 198 16MV 477 664 906 1019 705 855 168 137 3


2 times 1 MISOLV 286 1397 604 1896 453 1684 128 193 24MV 399 613 813 951 616 794 161 1303 9

3 times 1 MISOLV 370 1511 772 2026 579 1804 156 198 16MV 370 583 770 912 579 760 156 127 3


2 times 2 MIMOLV 537 2770 1114 3779 854 3349 241 390 36MV 602 828 1261 1381 949 1129 264 212 9

2 times 3 MIMOLV 536 2770 1137 3779 851 3349 240 390 24MV 736 1166 1538 1825 1155 1520 312 253 3


3 times 2 MIMOLV 530 2770 1135 3779 847 3349 243 390 24MV 732 1166 1535 1824 1151 1520 312 253 3

3 times 3 MIMOLV 774 3058 1591 4092 1201 3645 317 398 16MV 1082 1750 2288 2736 1710 2280 468 380 1



Table 4 Continued













119894119895 119894 isin 119873119879 119895 isin 119873119877 [130ndash133]




0

500

1000

2000

3000

1500

2500Cu

mul

ativ

e cap

acity

(Mbp

s)


1 times 4 SIMO

4 times 4 MIMO


3 times 2 MIMO

1 times 3 SIMO

SISOCCLV SISOCCMV

(a)

1

08

09

06

07

04

05

03

01

02

00 1000 2000 3000


SISOCCLV

SISOCCMV

Cum

ulat

ive c

apac

ity C

DF

(cum

ulat

ive c

apac

itylt

absc

issa)

4 times 4 MIMO

4 times 3 MIMO

1 times 3 SIMO

1 times 4 SIMO

3 times 2 MIMO

3 times 3 MIMO

(b)

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)


SISOCCLV

SISOCCMVminus1

4 times 4 MIMO

4 times 3 MIMO

3 times 3 MIMO

1 times 3 SIMO

1 times 4 SIMO

3 times 2 MIMO


(c)


08

09

1

06

07

04

05

03

01

02

0

SISOCCLV

SISOCCMV

Cum

ulat

ive c

apac

ity C

DF

(cum

ulat

ive c

apac

itylt

absc

issa)

3 times 3 MIMO

3 times 2 MIMO

1 times 4 SIMO

1 times 3 SIMO

4 times 3 MIMO

4 times 4 MIMO


(d)



11

1

1

11

2

2

2

3

3

4

4

079083064

12

12

119

312

171

15

25

171

121

101 165157

Rece

ive p

orts

(nR

)

Transmit ports (nT)

075

(a)

111 1 1

1

2

2

2

3

3

Transmit ports (nT)

12

12

14

14

16

22

18

167

102 248

082

089 084

136

167Re

ceiv

e por

ts (n

R)

(b)

11

1

2

3

4 365101

101

187184

174

174 174 174

189

093

088 092094

189

1

1

15

15

252 3


Rece

ive p

orts

(nR

)

(c)


1 1

1

1

2

1

2

3

178

178 267

089

132

093

074

15

Rece

ive p

orts

(nR

)

(d)









6 Conclusions






References























































































































































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










0

05

15

2

1

RMS-

DS

(120583s)


ACG (dB)




(a)

0

05

15

2

1

RMS-

DS

(120583s)





ACG (dB)

(b)


0 10 20 30 40 50CB (MHz)



0

05

15

2

1

RMS-

DS

(120583s)

(a)

0 10 20 30 40 50CB (MHz)



0

05

15

2

1RM

S-D

S (120583

s)

(b)
















LV 070 080 4950 37 1182 1232 2114 1631MV 080 050 4760 560 1178 232 2016 228

XCLV 020 040 3150 48 700 1020 1372 2113MV 020 030 3220 880 716 206 1402 377

AllLV 020 040 3250 48 720 1020 1416 2113MV 020 030 27 880 576 206 1188 377














00

500

1000

2000

1500

Cum

ulat

ive c

apac

ity (M

bps)

Overhead LVXC

Overhead LVCC

(a)

0

500

1000

2000

1500

Cum

ulat

ive c

apac

ity (M

bps)


Overhead MVCC

Overhead MVXC

(b)

0

Capa

city

CCD

F (c

apac

itygt

absc

issa)

01

02

03

04

05

06

07

08

09

1

0 500 1000 1500 2000

Overhead LVCC

Overhead MVCC

Overhead MVXC

Overhead LVXC

Capacity (Mbps)

(c)

0

500

1000

2000

1500

Cum

ulat

ive c

apac

ity (M

bps)

UndergroundLVCC

UndergroundLVXC



Suburban case


0

(d)

0

500

1000

2000

1500

Cum

ulat

ive c

apac

ity (M

bps)

Underground MVXC

UndergroundMVCC



Suburban case


0

(e)

0 500 1000 1500 20000

01

02

03

04

05

06

07

08

09

1

UndergroundLVCC

UndergroundLVXC

UndergroundMVCC

Underground

Capacity (Mbps)

Capa

city

CCD

F (c

apac

itygt

absc

issa)

MVXC

(f)











0

500

1000

2000

3000

1500

2500

0

500

1000

2000

3000

1500

2500

Cum

ulat

ive c

apac

ity (M

bps)

0

500

1000

2000

3000

1500

2500

Cum

ulat

ive c

apac

ity (M

bps)

0

500

1000

2000

3000

1500

2500

Cum

ulat

ive c

apac

ity (M

bps)

0

500

1000

2000

3000

1500

2500

Cum

ulat

ive c

apac

ity (M

bps)

0

500

1000

2000

3000

1500

2500

Cum

ulat

ive c

apac

ity (M

bps)

Cum

ulat

ive c

apac

ity (M

bps)

00

1000 2000 3000

Capa

city

CCD

F (c

apac

itygt

absc

issa)

01

02

03

04

05

06

07

08

09

1

(g)Capacity (Mbps)

0 1000 2000 3000C (Mb )




4 times 1 MISOLV



Suburban case



Suburban case

(a) (b)

(c)

(e) (f)

(d)







2sdot presence in












0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)

00

1000

2000

2000

3000

4000

4000

5000

6000

6000

7000

8000

8000

Cum

ulat

ive c

apac

ity (M

bps)

0

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)






Capa

city

CCD

F (c

apac

itygt

absc

issa)

01

02

03

04

05

06

07

08

09


3 times 1 MISOMV

4 times 4 MIMOLV

3 times 3 MIMOMV

SISOCCMVSISOXCMV

(a) (b)

(c) (d)

(e) (f)

(g)Capacity (Mbps)



Suburban case



Suburban case

1 times 3 SIMOMV






LV 471 1634 904 2152 700 1928 168 200 4MV 459 664 892 1019 688 855 168 137 3

1 times 2 SIMOLV 345 1497 693 1999 530 1785 140 194 24MV 375 471 744 769 572 632 149 114 9

1 times 3 SIMOLV 476 1646 907 2161 704 1938 168 198 16MV 477 664 906 1019 705 855 168 137 3


2 times 1 MISOLV 286 1397 604 1896 453 1684 128 193 24MV 399 613 813 951 616 794 161 1303 9

3 times 1 MISOLV 370 1511 772 2026 579 1804 156 198 16MV 370 583 770 912 579 760 156 127 3


2 times 2 MIMOLV 537 2770 1114 3779 854 3349 241 390 36MV 602 828 1261 1381 949 1129 264 212 9

2 times 3 MIMOLV 536 2770 1137 3779 851 3349 240 390 24MV 736 1166 1538 1825 1155 1520 312 253 3


3 times 2 MIMOLV 530 2770 1135 3779 847 3349 243 390 24MV 732 1166 1535 1824 1151 1520 312 253 3

3 times 3 MIMOLV 774 3058 1591 4092 1201 3645 317 398 16MV 1082 1750 2288 2736 1710 2280 468 380 1



Table 4 Continued













119894119895 119894 isin 119873119879 119895 isin 119873119877 [130ndash133]




0

500

1000

2000

3000

1500

2500Cu

mul

ativ

e cap

acity

(Mbp

s)


1 times 4 SIMO

4 times 4 MIMO


3 times 2 MIMO

1 times 3 SIMO

SISOCCLV SISOCCMV

(a)

1

08

09

06

07

04

05

03

01

02

00 1000 2000 3000


SISOCCLV

SISOCCMV

Cum

ulat

ive c

apac

ity C

DF

(cum

ulat

ive c

apac

itylt

absc

issa)

4 times 4 MIMO

4 times 3 MIMO

1 times 3 SIMO

1 times 4 SIMO

3 times 2 MIMO

3 times 3 MIMO

(b)

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)


SISOCCLV

SISOCCMVminus1

4 times 4 MIMO

4 times 3 MIMO

3 times 3 MIMO

1 times 3 SIMO

1 times 4 SIMO

3 times 2 MIMO


(c)


08

09

1

06

07

04

05

03

01

02

0

SISOCCLV

SISOCCMV

Cum

ulat

ive c

apac

ity C

DF

(cum

ulat

ive c

apac

itylt

absc

issa)

3 times 3 MIMO

3 times 2 MIMO

1 times 4 SIMO

1 times 3 SIMO

4 times 3 MIMO

4 times 4 MIMO


(d)



11

1

1

11

2

2

2

3

3

4

4

079083064

12

12

119

312

171

15

25

171

121

101 165157

Rece

ive p

orts

(nR

)

Transmit ports (nT)

075

(a)

111 1 1

1

2

2

2

3

3

Transmit ports (nT)

12

12

14

14

16

22

18

167

102 248

082

089 084

136

167Re

ceiv

e por

ts (n

R)

(b)

11

1

2

3

4 365101

101

187184

174

174 174 174

189

093

088 092094

189

1

1

15

15

252 3


Rece

ive p

orts

(nR

)

(c)


1 1

1

1

2

1

2

3

178

178 267

089

132

093

074

15

Rece

ive p

orts

(nR

)

(d)









6 Conclusions






References























































































































































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation













LV 070 080 4950 37 1182 1232 2114 1631MV 080 050 4760 560 1178 232 2016 228

XCLV 020 040 3150 48 700 1020 1372 2113MV 020 030 3220 880 716 206 1402 377

AllLV 020 040 3250 48 720 1020 1416 2113MV 020 030 27 880 576 206 1188 377














00

500

1000

2000

1500

Cum

ulat

ive c

apac

ity (M

bps)

Overhead LVXC

Overhead LVCC

(a)

0

500

1000

2000

1500

Cum

ulat

ive c

apac

ity (M

bps)


Overhead MVCC

Overhead MVXC

(b)

0

Capa

city

CCD

F (c

apac

itygt

absc

issa)

01

02

03

04

05

06

07

08

09

1

0 500 1000 1500 2000

Overhead LVCC

Overhead MVCC

Overhead MVXC

Overhead LVXC

Capacity (Mbps)

(c)

0

500

1000

2000

1500

Cum

ulat

ive c

apac

ity (M

bps)

UndergroundLVCC

UndergroundLVXC



Suburban case


0

(d)

0

500

1000

2000

1500

Cum

ulat

ive c

apac

ity (M

bps)

Underground MVXC

UndergroundMVCC



Suburban case


0

(e)

0 500 1000 1500 20000

01

02

03

04

05

06

07

08

09

1

UndergroundLVCC

UndergroundLVXC

UndergroundMVCC

Underground

Capacity (Mbps)

Capa

city

CCD

F (c

apac

itygt

absc

issa)

MVXC

(f)











0

500

1000

2000

3000

1500

2500

0

500

1000

2000

3000

1500

2500

Cum

ulat

ive c

apac

ity (M

bps)

0

500

1000

2000

3000

1500

2500

Cum

ulat

ive c

apac

ity (M

bps)

0

500

1000

2000

3000

1500

2500

Cum

ulat

ive c

apac

ity (M

bps)

0

500

1000

2000

3000

1500

2500

Cum

ulat

ive c

apac

ity (M

bps)

0

500

1000

2000

3000

1500

2500

Cum

ulat

ive c

apac

ity (M

bps)

Cum

ulat

ive c

apac

ity (M

bps)

00

1000 2000 3000

Capa

city

CCD

F (c

apac

itygt

absc

issa)

01

02

03

04

05

06

07

08

09

1

(g)Capacity (Mbps)

0 1000 2000 3000C (Mb )




4 times 1 MISOLV



Suburban case



Suburban case

(a) (b)

(c)

(e) (f)

(d)







2sdot presence in












0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)

00

1000

2000

2000

3000

4000

4000

5000

6000

6000

7000

8000

8000

Cum

ulat

ive c

apac

ity (M

bps)

0

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)






Capa

city

CCD

F (c

apac

itygt

absc

issa)

01

02

03

04

05

06

07

08

09


3 times 1 MISOMV

4 times 4 MIMOLV

3 times 3 MIMOMV

SISOCCMVSISOXCMV

(a) (b)

(c) (d)

(e) (f)

(g)Capacity (Mbps)



Suburban case



Suburban case

1 times 3 SIMOMV






LV 471 1634 904 2152 700 1928 168 200 4MV 459 664 892 1019 688 855 168 137 3

1 times 2 SIMOLV 345 1497 693 1999 530 1785 140 194 24MV 375 471 744 769 572 632 149 114 9

1 times 3 SIMOLV 476 1646 907 2161 704 1938 168 198 16MV 477 664 906 1019 705 855 168 137 3


2 times 1 MISOLV 286 1397 604 1896 453 1684 128 193 24MV 399 613 813 951 616 794 161 1303 9

3 times 1 MISOLV 370 1511 772 2026 579 1804 156 198 16MV 370 583 770 912 579 760 156 127 3


2 times 2 MIMOLV 537 2770 1114 3779 854 3349 241 390 36MV 602 828 1261 1381 949 1129 264 212 9

2 times 3 MIMOLV 536 2770 1137 3779 851 3349 240 390 24MV 736 1166 1538 1825 1155 1520 312 253 3


3 times 2 MIMOLV 530 2770 1135 3779 847 3349 243 390 24MV 732 1166 1535 1824 1151 1520 312 253 3

3 times 3 MIMOLV 774 3058 1591 4092 1201 3645 317 398 16MV 1082 1750 2288 2736 1710 2280 468 380 1



Table 4 Continued













119894119895 119894 isin 119873119879 119895 isin 119873119877 [130ndash133]




0

500

1000

2000

3000

1500

2500Cu

mul

ativ

e cap

acity

(Mbp

s)


1 times 4 SIMO

4 times 4 MIMO


3 times 2 MIMO

1 times 3 SIMO

SISOCCLV SISOCCMV

(a)

1

08

09

06

07

04

05

03

01

02

00 1000 2000 3000


SISOCCLV

SISOCCMV

Cum

ulat

ive c

apac

ity C

DF

(cum

ulat

ive c

apac

itylt

absc

issa)

4 times 4 MIMO

4 times 3 MIMO

1 times 3 SIMO

1 times 4 SIMO

3 times 2 MIMO

3 times 3 MIMO

(b)

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)


SISOCCLV

SISOCCMVminus1

4 times 4 MIMO

4 times 3 MIMO

3 times 3 MIMO

1 times 3 SIMO

1 times 4 SIMO

3 times 2 MIMO


(c)


08

09

1

06

07

04

05

03

01

02

0

SISOCCLV

SISOCCMV

Cum

ulat

ive c

apac

ity C

DF

(cum

ulat

ive c

apac

itylt

absc

issa)

3 times 3 MIMO

3 times 2 MIMO

1 times 4 SIMO

1 times 3 SIMO

4 times 3 MIMO

4 times 4 MIMO


(d)



11

1

1

11

2

2

2

3

3

4

4

079083064

12

12

119

312

171

15

25

171

121

101 165157

Rece

ive p

orts

(nR

)

Transmit ports (nT)

075

(a)

111 1 1

1

2

2

2

3

3

Transmit ports (nT)

12

12

14

14

16

22

18

167

102 248

082

089 084

136

167Re

ceiv

e por

ts (n

R)

(b)

11

1

2

3

4 365101

101

187184

174

174 174 174

189

093

088 092094

189

1

1

15

15

252 3


Rece

ive p

orts

(nR

)

(c)


1 1

1

1

2

1

2

3

178

178 267

089

132

093

074

15

Rece

ive p

orts

(nR

)

(d)









6 Conclusions






References























































































































































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation











00

500

1000

2000

1500

Cum

ulat

ive c

apac

ity (M

bps)

Overhead LVXC

Overhead LVCC

(a)

0

500

1000

2000

1500

Cum

ulat

ive c

apac

ity (M

bps)


Overhead MVCC

Overhead MVXC

(b)

0

Capa

city

CCD

F (c

apac

itygt

absc

issa)

01

02

03

04

05

06

07

08

09

1

0 500 1000 1500 2000

Overhead LVCC

Overhead MVCC

Overhead MVXC

Overhead LVXC

Capacity (Mbps)

(c)

0

500

1000

2000

1500

Cum

ulat

ive c

apac

ity (M

bps)

UndergroundLVCC

UndergroundLVXC



Suburban case


0

(d)

0

500

1000

2000

1500

Cum

ulat

ive c

apac

ity (M

bps)

Underground MVXC

UndergroundMVCC



Suburban case


0

(e)

0 500 1000 1500 20000

01

02

03

04

05

06

07

08

09

1

UndergroundLVCC

UndergroundLVXC

UndergroundMVCC

Underground

Capacity (Mbps)

Capa

city

CCD

F (c

apac

itygt

absc

issa)

MVXC

(f)











0

500

1000

2000

3000

1500

2500

0

500

1000

2000

3000

1500

2500

Cum

ulat

ive c

apac

ity (M

bps)

0

500

1000

2000

3000

1500

2500

Cum

ulat

ive c

apac

ity (M

bps)

0

500

1000

2000

3000

1500

2500

Cum

ulat

ive c

apac

ity (M

bps)

0

500

1000

2000

3000

1500

2500

Cum

ulat

ive c

apac

ity (M

bps)

0

500

1000

2000

3000

1500

2500

Cum

ulat

ive c

apac

ity (M

bps)

Cum

ulat

ive c

apac

ity (M

bps)

00

1000 2000 3000

Capa

city

CCD

F (c

apac

itygt

absc

issa)

01

02

03

04

05

06

07

08

09

1

(g)Capacity (Mbps)

0 1000 2000 3000C (Mb )




4 times 1 MISOLV



Suburban case



Suburban case

(a) (b)

(c)

(e) (f)

(d)







2sdot presence in












0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)

00

1000

2000

2000

3000

4000

4000

5000

6000

6000

7000

8000

8000

Cum

ulat

ive c

apac

ity (M

bps)

0

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)






Capa

city

CCD

F (c

apac

itygt

absc

issa)

01

02

03

04

05

06

07

08

09


3 times 1 MISOMV

4 times 4 MIMOLV

3 times 3 MIMOMV

SISOCCMVSISOXCMV

(a) (b)

(c) (d)

(e) (f)

(g)Capacity (Mbps)



Suburban case



Suburban case

1 times 3 SIMOMV






LV 471 1634 904 2152 700 1928 168 200 4MV 459 664 892 1019 688 855 168 137 3

1 times 2 SIMOLV 345 1497 693 1999 530 1785 140 194 24MV 375 471 744 769 572 632 149 114 9

1 times 3 SIMOLV 476 1646 907 2161 704 1938 168 198 16MV 477 664 906 1019 705 855 168 137 3


2 times 1 MISOLV 286 1397 604 1896 453 1684 128 193 24MV 399 613 813 951 616 794 161 1303 9

3 times 1 MISOLV 370 1511 772 2026 579 1804 156 198 16MV 370 583 770 912 579 760 156 127 3


2 times 2 MIMOLV 537 2770 1114 3779 854 3349 241 390 36MV 602 828 1261 1381 949 1129 264 212 9

2 times 3 MIMOLV 536 2770 1137 3779 851 3349 240 390 24MV 736 1166 1538 1825 1155 1520 312 253 3


3 times 2 MIMOLV 530 2770 1135 3779 847 3349 243 390 24MV 732 1166 1535 1824 1151 1520 312 253 3

3 times 3 MIMOLV 774 3058 1591 4092 1201 3645 317 398 16MV 1082 1750 2288 2736 1710 2280 468 380 1



Table 4 Continued













119894119895 119894 isin 119873119879 119895 isin 119873119877 [130ndash133]




0

500

1000

2000

3000

1500

2500Cu

mul

ativ

e cap

acity

(Mbp

s)


1 times 4 SIMO

4 times 4 MIMO


3 times 2 MIMO

1 times 3 SIMO

SISOCCLV SISOCCMV

(a)

1

08

09

06

07

04

05

03

01

02

00 1000 2000 3000


SISOCCLV

SISOCCMV

Cum

ulat

ive c

apac

ity C

DF

(cum

ulat

ive c

apac

itylt

absc

issa)

4 times 4 MIMO

4 times 3 MIMO

1 times 3 SIMO

1 times 4 SIMO

3 times 2 MIMO

3 times 3 MIMO

(b)

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)


SISOCCLV

SISOCCMVminus1

4 times 4 MIMO

4 times 3 MIMO

3 times 3 MIMO

1 times 3 SIMO

1 times 4 SIMO

3 times 2 MIMO


(c)


08

09

1

06

07

04

05

03

01

02

0

SISOCCLV

SISOCCMV

Cum

ulat

ive c

apac

ity C

DF

(cum

ulat

ive c

apac

itylt

absc

issa)

3 times 3 MIMO

3 times 2 MIMO

1 times 4 SIMO

1 times 3 SIMO

4 times 3 MIMO

4 times 4 MIMO


(d)



11

1

1

11

2

2

2

3

3

4

4

079083064

12

12

119

312

171

15

25

171

121

101 165157

Rece

ive p

orts

(nR

)

Transmit ports (nT)

075

(a)

111 1 1

1

2

2

2

3

3

Transmit ports (nT)

12

12

14

14

16

22

18

167

102 248

082

089 084

136

167Re

ceiv

e por

ts (n

R)

(b)

11

1

2

3

4 365101

101

187184

174

174 174 174

189

093

088 092094

189

1

1

15

15

252 3


Rece

ive p

orts

(nR

)

(c)


1 1

1

1

2

1

2

3

178

178 267

089

132

093

074

15

Rece

ive p

orts

(nR

)

(d)









6 Conclusions






References























































































































































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation
















0

500

1000

2000

3000

1500

2500

0

500

1000

2000

3000

1500

2500

Cum

ulat

ive c

apac

ity (M

bps)

0

500

1000

2000

3000

1500

2500

Cum

ulat

ive c

apac

ity (M

bps)

0

500

1000

2000

3000

1500

2500

Cum

ulat

ive c

apac

ity (M

bps)

0

500

1000

2000

3000

1500

2500

Cum

ulat

ive c

apac

ity (M

bps)

0

500

1000

2000

3000

1500

2500

Cum

ulat

ive c

apac

ity (M

bps)

Cum

ulat

ive c

apac

ity (M

bps)

00

1000 2000 3000

Capa

city

CCD

F (c

apac

itygt

absc

issa)

01

02

03

04

05

06

07

08

09

1

(g)Capacity (Mbps)

0 1000 2000 3000C (Mb )




4 times 1 MISOLV



Suburban case



Suburban case

(a) (b)

(c)

(e) (f)

(d)







2sdot presence in












0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)

00

1000

2000

2000

3000

4000

4000

5000

6000

6000

7000

8000

8000

Cum

ulat

ive c

apac

ity (M

bps)

0

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)






Capa

city

CCD

F (c

apac

itygt

absc

issa)

01

02

03

04

05

06

07

08

09


3 times 1 MISOMV

4 times 4 MIMOLV

3 times 3 MIMOMV

SISOCCMVSISOXCMV

(a) (b)

(c) (d)

(e) (f)

(g)Capacity (Mbps)



Suburban case



Suburban case

1 times 3 SIMOMV






LV 471 1634 904 2152 700 1928 168 200 4MV 459 664 892 1019 688 855 168 137 3

1 times 2 SIMOLV 345 1497 693 1999 530 1785 140 194 24MV 375 471 744 769 572 632 149 114 9

1 times 3 SIMOLV 476 1646 907 2161 704 1938 168 198 16MV 477 664 906 1019 705 855 168 137 3


2 times 1 MISOLV 286 1397 604 1896 453 1684 128 193 24MV 399 613 813 951 616 794 161 1303 9

3 times 1 MISOLV 370 1511 772 2026 579 1804 156 198 16MV 370 583 770 912 579 760 156 127 3


2 times 2 MIMOLV 537 2770 1114 3779 854 3349 241 390 36MV 602 828 1261 1381 949 1129 264 212 9

2 times 3 MIMOLV 536 2770 1137 3779 851 3349 240 390 24MV 736 1166 1538 1825 1155 1520 312 253 3


3 times 2 MIMOLV 530 2770 1135 3779 847 3349 243 390 24MV 732 1166 1535 1824 1151 1520 312 253 3

3 times 3 MIMOLV 774 3058 1591 4092 1201 3645 317 398 16MV 1082 1750 2288 2736 1710 2280 468 380 1



Table 4 Continued













119894119895 119894 isin 119873119879 119895 isin 119873119877 [130ndash133]




0

500

1000

2000

3000

1500

2500Cu

mul

ativ

e cap

acity

(Mbp

s)


1 times 4 SIMO

4 times 4 MIMO


3 times 2 MIMO

1 times 3 SIMO

SISOCCLV SISOCCMV

(a)

1

08

09

06

07

04

05

03

01

02

00 1000 2000 3000


SISOCCLV

SISOCCMV

Cum

ulat

ive c

apac

ity C

DF

(cum

ulat

ive c

apac

itylt

absc

issa)

4 times 4 MIMO

4 times 3 MIMO

1 times 3 SIMO

1 times 4 SIMO

3 times 2 MIMO

3 times 3 MIMO

(b)

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)


SISOCCLV

SISOCCMVminus1

4 times 4 MIMO

4 times 3 MIMO

3 times 3 MIMO

1 times 3 SIMO

1 times 4 SIMO

3 times 2 MIMO


(c)


08

09

1

06

07

04

05

03

01

02

0

SISOCCLV

SISOCCMV

Cum

ulat

ive c

apac

ity C

DF

(cum

ulat

ive c

apac

itylt

absc

issa)

3 times 3 MIMO

3 times 2 MIMO

1 times 4 SIMO

1 times 3 SIMO

4 times 3 MIMO

4 times 4 MIMO


(d)



11

1

1

11

2

2

2

3

3

4

4

079083064

12

12

119

312

171

15

25

171

121

101 165157

Rece

ive p

orts

(nR

)

Transmit ports (nT)

075

(a)

111 1 1

1

2

2

2

3

3

Transmit ports (nT)

12

12

14

14

16

22

18

167

102 248

082

089 084

136

167Re

ceiv

e por

ts (n

R)

(b)

11

1

2

3

4 365101

101

187184

174

174 174 174

189

093

088 092094

189

1

1

15

15

252 3


Rece

ive p

orts

(nR

)

(c)


1 1

1

1

2

1

2

3

178

178 267

089

132

093

074

15

Rece

ive p

orts

(nR

)

(d)









6 Conclusions






References























































































































































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation














2sdot presence in












0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)

00

1000

2000

2000

3000

4000

4000

5000

6000

6000

7000

8000

8000

Cum

ulat

ive c

apac

ity (M

bps)

0

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)






Capa

city

CCD

F (c

apac

itygt

absc

issa)

01

02

03

04

05

06

07

08

09


3 times 1 MISOMV

4 times 4 MIMOLV

3 times 3 MIMOMV

SISOCCMVSISOXCMV

(a) (b)

(c) (d)

(e) (f)

(g)Capacity (Mbps)



Suburban case



Suburban case

1 times 3 SIMOMV






LV 471 1634 904 2152 700 1928 168 200 4MV 459 664 892 1019 688 855 168 137 3

1 times 2 SIMOLV 345 1497 693 1999 530 1785 140 194 24MV 375 471 744 769 572 632 149 114 9

1 times 3 SIMOLV 476 1646 907 2161 704 1938 168 198 16MV 477 664 906 1019 705 855 168 137 3


2 times 1 MISOLV 286 1397 604 1896 453 1684 128 193 24MV 399 613 813 951 616 794 161 1303 9

3 times 1 MISOLV 370 1511 772 2026 579 1804 156 198 16MV 370 583 770 912 579 760 156 127 3


2 times 2 MIMOLV 537 2770 1114 3779 854 3349 241 390 36MV 602 828 1261 1381 949 1129 264 212 9

2 times 3 MIMOLV 536 2770 1137 3779 851 3349 240 390 24MV 736 1166 1538 1825 1155 1520 312 253 3


3 times 2 MIMOLV 530 2770 1135 3779 847 3349 243 390 24MV 732 1166 1535 1824 1151 1520 312 253 3

3 times 3 MIMOLV 774 3058 1591 4092 1201 3645 317 398 16MV 1082 1750 2288 2736 1710 2280 468 380 1



Table 4 Continued













119894119895 119894 isin 119873119879 119895 isin 119873119877 [130ndash133]




0

500

1000

2000

3000

1500

2500Cu

mul

ativ

e cap

acity

(Mbp

s)


1 times 4 SIMO

4 times 4 MIMO


3 times 2 MIMO

1 times 3 SIMO

SISOCCLV SISOCCMV

(a)

1

08

09

06

07

04

05

03

01

02

00 1000 2000 3000


SISOCCLV

SISOCCMV

Cum

ulat

ive c

apac

ity C

DF

(cum

ulat

ive c

apac

itylt

absc

issa)

4 times 4 MIMO

4 times 3 MIMO

1 times 3 SIMO

1 times 4 SIMO

3 times 2 MIMO

3 times 3 MIMO

(b)

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)


SISOCCLV

SISOCCMVminus1

4 times 4 MIMO

4 times 3 MIMO

3 times 3 MIMO

1 times 3 SIMO

1 times 4 SIMO

3 times 2 MIMO


(c)


08

09

1

06

07

04

05

03

01

02

0

SISOCCLV

SISOCCMV

Cum

ulat

ive c

apac

ity C

DF

(cum

ulat

ive c

apac

itylt

absc

issa)

3 times 3 MIMO

3 times 2 MIMO

1 times 4 SIMO

1 times 3 SIMO

4 times 3 MIMO

4 times 4 MIMO


(d)



11

1

1

11

2

2

2

3

3

4

4

079083064

12

12

119

312

171

15

25

171

121

101 165157

Rece

ive p

orts

(nR

)

Transmit ports (nT)

075

(a)

111 1 1

1

2

2

2

3

3

Transmit ports (nT)

12

12

14

14

16

22

18

167

102 248

082

089 084

136

167Re

ceiv

e por

ts (n

R)

(b)

11

1

2

3

4 365101

101

187184

174

174 174 174

189

093

088 092094

189

1

1

15

15

252 3


Rece

ive p

orts

(nR

)

(c)


1 1

1

1

2

1

2

3

178

178 267

089

132

093

074

15

Rece

ive p

orts

(nR

)

(d)









6 Conclusions






References























































































































































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)

00

1000

2000

2000

3000

4000

4000

5000

6000

6000

7000

8000

8000

Cum

ulat

ive c

apac

ity (M

bps)

0

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)






Capa

city

CCD

F (c

apac

itygt

absc

issa)

01

02

03

04

05

06

07

08

09


3 times 1 MISOMV

4 times 4 MIMOLV

3 times 3 MIMOMV

SISOCCMVSISOXCMV

(a) (b)

(c) (d)

(e) (f)

(g)Capacity (Mbps)



Suburban case



Suburban case

1 times 3 SIMOMV






LV 471 1634 904 2152 700 1928 168 200 4MV 459 664 892 1019 688 855 168 137 3

1 times 2 SIMOLV 345 1497 693 1999 530 1785 140 194 24MV 375 471 744 769 572 632 149 114 9

1 times 3 SIMOLV 476 1646 907 2161 704 1938 168 198 16MV 477 664 906 1019 705 855 168 137 3


2 times 1 MISOLV 286 1397 604 1896 453 1684 128 193 24MV 399 613 813 951 616 794 161 1303 9

3 times 1 MISOLV 370 1511 772 2026 579 1804 156 198 16MV 370 583 770 912 579 760 156 127 3


2 times 2 MIMOLV 537 2770 1114 3779 854 3349 241 390 36MV 602 828 1261 1381 949 1129 264 212 9

2 times 3 MIMOLV 536 2770 1137 3779 851 3349 240 390 24MV 736 1166 1538 1825 1155 1520 312 253 3


3 times 2 MIMOLV 530 2770 1135 3779 847 3349 243 390 24MV 732 1166 1535 1824 1151 1520 312 253 3

3 times 3 MIMOLV 774 3058 1591 4092 1201 3645 317 398 16MV 1082 1750 2288 2736 1710 2280 468 380 1



Table 4 Continued













119894119895 119894 isin 119873119879 119895 isin 119873119877 [130ndash133]




0

500

1000

2000

3000

1500

2500Cu

mul

ativ

e cap

acity

(Mbp

s)


1 times 4 SIMO

4 times 4 MIMO


3 times 2 MIMO

1 times 3 SIMO

SISOCCLV SISOCCMV

(a)

1

08

09

06

07

04

05

03

01

02

00 1000 2000 3000


SISOCCLV

SISOCCMV

Cum

ulat

ive c

apac

ity C

DF

(cum

ulat

ive c

apac

itylt

absc

issa)

4 times 4 MIMO

4 times 3 MIMO

1 times 3 SIMO

1 times 4 SIMO

3 times 2 MIMO

3 times 3 MIMO

(b)

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)


SISOCCLV

SISOCCMVminus1

4 times 4 MIMO

4 times 3 MIMO

3 times 3 MIMO

1 times 3 SIMO

1 times 4 SIMO

3 times 2 MIMO


(c)


08

09

1

06

07

04

05

03

01

02

0

SISOCCLV

SISOCCMV

Cum

ulat

ive c

apac

ity C

DF

(cum

ulat

ive c

apac

itylt

absc

issa)

3 times 3 MIMO

3 times 2 MIMO

1 times 4 SIMO

1 times 3 SIMO

4 times 3 MIMO

4 times 4 MIMO


(d)



11

1

1

11

2

2

2

3

3

4

4

079083064

12

12

119

312

171

15

25

171

121

101 165157

Rece

ive p

orts

(nR

)

Transmit ports (nT)

075

(a)

111 1 1

1

2

2

2

3

3

Transmit ports (nT)

12

12

14

14

16

22

18

167

102 248

082

089 084

136

167Re

ceiv

e por

ts (n

R)

(b)

11

1

2

3

4 365101

101

187184

174

174 174 174

189

093

088 092094

189

1

1

15

15

252 3


Rece

ive p

orts

(nR

)

(c)


1 1

1

1

2

1

2

3

178

178 267

089

132

093

074

15

Rece

ive p

orts

(nR

)

(d)









6 Conclusions






References























































































































































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation













LV 471 1634 904 2152 700 1928 168 200 4MV 459 664 892 1019 688 855 168 137 3

1 times 2 SIMOLV 345 1497 693 1999 530 1785 140 194 24MV 375 471 744 769 572 632 149 114 9

1 times 3 SIMOLV 476 1646 907 2161 704 1938 168 198 16MV 477 664 906 1019 705 855 168 137 3


2 times 1 MISOLV 286 1397 604 1896 453 1684 128 193 24MV 399 613 813 951 616 794 161 1303 9

3 times 1 MISOLV 370 1511 772 2026 579 1804 156 198 16MV 370 583 770 912 579 760 156 127 3


2 times 2 MIMOLV 537 2770 1114 3779 854 3349 241 390 36MV 602 828 1261 1381 949 1129 264 212 9

2 times 3 MIMOLV 536 2770 1137 3779 851 3349 240 390 24MV 736 1166 1538 1825 1155 1520 312 253 3


3 times 2 MIMOLV 530 2770 1135 3779 847 3349 243 390 24MV 732 1166 1535 1824 1151 1520 312 253 3

3 times 3 MIMOLV 774 3058 1591 4092 1201 3645 317 398 16MV 1082 1750 2288 2736 1710 2280 468 380 1



Table 4 Continued













119894119895 119894 isin 119873119879 119895 isin 119873119877 [130ndash133]




0

500

1000

2000

3000

1500

2500Cu

mul

ativ

e cap

acity

(Mbp

s)


1 times 4 SIMO

4 times 4 MIMO


3 times 2 MIMO

1 times 3 SIMO

SISOCCLV SISOCCMV

(a)

1

08

09

06

07

04

05

03

01

02

00 1000 2000 3000


SISOCCLV

SISOCCMV

Cum

ulat

ive c

apac

ity C

DF

(cum

ulat

ive c

apac

itylt

absc

issa)

4 times 4 MIMO

4 times 3 MIMO

1 times 3 SIMO

1 times 4 SIMO

3 times 2 MIMO

3 times 3 MIMO

(b)

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)


SISOCCLV

SISOCCMVminus1

4 times 4 MIMO

4 times 3 MIMO

3 times 3 MIMO

1 times 3 SIMO

1 times 4 SIMO

3 times 2 MIMO


(c)


08

09

1

06

07

04

05

03

01

02

0

SISOCCLV

SISOCCMV

Cum

ulat

ive c

apac

ity C

DF

(cum

ulat

ive c

apac

itylt

absc

issa)

3 times 3 MIMO

3 times 2 MIMO

1 times 4 SIMO

1 times 3 SIMO

4 times 3 MIMO

4 times 4 MIMO


(d)



11

1

1

11

2

2

2

3

3

4

4

079083064

12

12

119

312

171

15

25

171

121

101 165157

Rece

ive p

orts

(nR

)

Transmit ports (nT)

075

(a)

111 1 1

1

2

2

2

3

3

Transmit ports (nT)

12

12

14

14

16

22

18

167

102 248

082

089 084

136

167Re

ceiv

e por

ts (n

R)

(b)

11

1

2

3

4 365101

101

187184

174

174 174 174

189

093

088 092094

189

1

1

15

15

252 3


Rece

ive p

orts

(nR

)

(c)


1 1

1

1

2

1

2

3

178

178 267

089

132

093

074

15

Rece

ive p

orts

(nR

)

(d)









6 Conclusions






References























































































































































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










Table 4 Continued













119894119895 119894 isin 119873119879 119895 isin 119873119877 [130ndash133]




0

500

1000

2000

3000

1500

2500Cu

mul

ativ

e cap

acity

(Mbp

s)


1 times 4 SIMO

4 times 4 MIMO


3 times 2 MIMO

1 times 3 SIMO

SISOCCLV SISOCCMV

(a)

1

08

09

06

07

04

05

03

01

02

00 1000 2000 3000


SISOCCLV

SISOCCMV

Cum

ulat

ive c

apac

ity C

DF

(cum

ulat

ive c

apac

itylt

absc

issa)

4 times 4 MIMO

4 times 3 MIMO

1 times 3 SIMO

1 times 4 SIMO

3 times 2 MIMO

3 times 3 MIMO

(b)

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)


SISOCCLV

SISOCCMVminus1

4 times 4 MIMO

4 times 3 MIMO

3 times 3 MIMO

1 times 3 SIMO

1 times 4 SIMO

3 times 2 MIMO


(c)


08

09

1

06

07

04

05

03

01

02

0

SISOCCLV

SISOCCMV

Cum

ulat

ive c

apac

ity C

DF

(cum

ulat

ive c

apac

itylt

absc

issa)

3 times 3 MIMO

3 times 2 MIMO

1 times 4 SIMO

1 times 3 SIMO

4 times 3 MIMO

4 times 4 MIMO


(d)



11

1

1

11

2

2

2

3

3

4

4

079083064

12

12

119

312

171

15

25

171

121

101 165157

Rece

ive p

orts

(nR

)

Transmit ports (nT)

075

(a)

111 1 1

1

2

2

2

3

3

Transmit ports (nT)

12

12

14

14

16

22

18

167

102 248

082

089 084

136

167Re

ceiv

e por

ts (n

R)

(b)

11

1

2

3

4 365101

101

187184

174

174 174 174

189

093

088 092094

189

1

1

15

15

252 3


Rece

ive p

orts

(nR

)

(c)


1 1

1

1

2

1

2

3

178

178 267

089

132

093

074

15

Rece

ive p

orts

(nR

)

(d)









6 Conclusions






References























































































































































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










0

500

1000

2000

3000

1500

2500Cu

mul

ativ

e cap

acity

(Mbp

s)


1 times 4 SIMO

4 times 4 MIMO


3 times 2 MIMO

1 times 3 SIMO

SISOCCLV SISOCCMV

(a)

1

08

09

06

07

04

05

03

01

02

00 1000 2000 3000


SISOCCLV

SISOCCMV

Cum

ulat

ive c

apac

ity C

DF

(cum

ulat

ive c

apac

itylt

absc

issa)

4 times 4 MIMO

4 times 3 MIMO

1 times 3 SIMO

1 times 4 SIMO

3 times 2 MIMO

3 times 3 MIMO

(b)

0

1000

2000

3000

4000

5000

6000

7000

8000

Cum

ulat

ive c

apac

ity (M

bps)


SISOCCLV

SISOCCMVminus1

4 times 4 MIMO

4 times 3 MIMO

3 times 3 MIMO

1 times 3 SIMO

1 times 4 SIMO

3 times 2 MIMO


(c)


08

09

1

06

07

04

05

03

01

02

0

SISOCCLV

SISOCCMV

Cum

ulat

ive c

apac

ity C

DF

(cum

ulat

ive c

apac

itylt

absc

issa)

3 times 3 MIMO

3 times 2 MIMO

1 times 4 SIMO

1 times 3 SIMO

4 times 3 MIMO

4 times 4 MIMO


(d)



11

1

1

11

2

2

2

3

3

4

4

079083064

12

12

119

312

171

15

25

171

121

101 165157

Rece

ive p

orts

(nR

)

Transmit ports (nT)

075

(a)

111 1 1

1

2

2

2

3

3

Transmit ports (nT)

12

12

14

14

16

22

18

167

102 248

082

089 084

136

167Re

ceiv

e por

ts (n

R)

(b)

11

1

2

3

4 365101

101

187184

174

174 174 174

189

093

088 092094

189

1

1

15

15

252 3


Rece

ive p

orts

(nR

)

(c)


1 1

1

1

2

1

2

3

178

178 267

089

132

093

074

15

Rece

ive p

orts

(nR

)

(d)









6 Conclusions






References























































































































































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










11

1

1

11

2

2

2

3

3

4

4

079083064

12

12

119

312

171

15

25

171

121

101 165157

Rece

ive p

orts

(nR

)

Transmit ports (nT)

075

(a)

111 1 1

1

2

2

2

3

3

Transmit ports (nT)

12

12

14

14

16

22

18

167

102 248

082

089 084

136

167Re

ceiv

e por

ts (n

R)

(b)

11

1

2

3

4 365101

101

187184

174

174 174 174

189

093

088 092094

189

1

1

15

15

252 3


Rece

ive p

orts

(nR

)

(c)


1 1

1

1

2

1

2

3

178

178 267

089

132

093

074

15

Rece

ive p

orts

(nR

)

(d)









6 Conclusions






References























































































































































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation













6 Conclusions






References























































































































































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation


























































































































































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation









review article broadband over power lines systems...

Documents