device discovery for d2d communication in in-band cellular … · 2018. 4. 3. · device discovery...

RESEARCH Open Access

Device discovery for D2D communicationin in-band cellular networks using spheredecoder like (SDL) algorithmO. Hayat1,2* , R. Ngah2 and Yasser Zahedi3

Abstract

In the fifth generation (5G), it is anticipated that device-to-device (D2D) operation will be locally incorporated as apart without any bounds. In D2D network, multiple devices coexisting is a challenging subject of device discovery.The device discovery is performed under a visually impaired situation such as channel information, location, andthe number of devices. In this paper, centralized device discovery is chosen due to power consumption and signalingoverhead of the distributed system. A distinctive approach for device discovery in an in-band cellular network, basedon the device’s power, is suggested with an efficient technique which enhances the implementation of D2Dcommunication and improves the accomplishment by alleviating the discovery issues. The group of devicesforms a lattice structure, and it is positioned in the coverage area. The hypersphere is constructed based onthe power knowledge of a discoverer device which helps for accurate and fast device discovery in a latticestructure. Besides, sphere decoder like (SDL) algorithm is applied for quick and precise discovery in the latticestructure. Simulation results present the performance of the proposed QR factorized lattice structure schemeregarding device power, enhanced in the number of discovered devices and controlled signaling overhead.

Keywords: Device-to-device (D2D) communication, Sphere decoder like (SDL) algorithm, Device discovery,Cellular network, End users (EU)

1 IntroductionDevice-to-device (D2D) communication alludes to directtransmission between two devices without passingthrough the base station. It has been broadly anticipatedto be an essential cornerstone to enhance systemperformance and bolster new amenities beyond 2020 infuture fifth generation (5G) systems [1]. In 5G networks,it is anticipated that controlled D2D communication of-fers the open door for short-distance communicationand local management and permits the isolation of localactivity from the global activity, for example, local dataoffloading. D2D communication evacuates the data traf-fic heap load on the backhaul and center systems anddecreases the vital exertion for managing data traffic atthe center system. Due to proximity services, D2D

communication is viewed as a promising remedy for en-hancing communication accomplishment and systemcapacity of long-term evolution-advanced (LTE-A) net-work. The potential enhancements in proximity servicesthat can be given by D2D are not entirely exploited yet.In the 5G network, such confinement does not exist anylonger, and it is anticipated that D2D operation in thein-band cellular network will be locally incorporated as acomponent without any bounds in the 5G network [2].The in-band cellular network is considered in this re-search because interference and resource allocation iscontrolled by the base station or center system [3].An important technique in deploying D2D communi-

cation is device discovery. Device discovery is character-ized into two categories which are distributed discoveryand centralized discovery. In the distributed discoverydesign, optimal resources and legitimate transmissionpower are apportioned with the consent of base station.However, this strategy involves complex signaling over-head and multifaceted nature of multiple transmitters to

* Correspondence: [email protected] of Engineering, National University of Modern Language(NUML), H-9, Islamabad, Pakistan2Wireless Communication Centre (WCC), Faculty of Electrical Engineering,Universiti Teknologi Malaysia, 81310 Johor Bahru, MalaysiaFull list of author information is available at the end of the article

© The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, andreproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link tothe Creative Commons license, and indicate if changes were made.

Hayat et al. EURASIP Journal on Wireless Communications and Networking (2018) 2018:74 https://doi.org/10.1186/s13638-018-1083-8

http://crossmark.crossref.org/dialog/?doi=10.1186/s13638-018-1083-8&domain=pdfhttp://orcid.org/0000-0002-5067-5845mailto:[email protected]://creativecommons.org/licenses/by/4.0/

be composed [4]. In contrast, centralized device discov-ery design is adequately controlled by the base stationand resources are managed by center system. Besides,the proximity services, for example, commercial an-nouncement, public transport and municipality informa-tion, local programs, and impulsive social and corporatecontacts infer decreased energy consumption and la-tency. Therefore, in D2D-enabled centralized networks,multiple devices coexisting is a challenging subject ofdevice discovery for D2D communication to initiate theproximity services. However, latest reviews have concen-trated on the D2D communication issues accepting thatdevice discovery issue [5–7] is the vital problem andneeds solution.Discovery signal is designed in [8], minimizes colli-

sions, and tries to improve the discovery process. Aslight portion of resources used for device discovery anddiscovery signal which can be broadcasted with a mini-mum delay is proposed in [9, 10]. Distinguishing be-tween different discovery signals is difficult which leadsto an increment in power consumption. Compressivesensing method has been suggested in [11], in whichuser detection decreases the collision for device discov-ery. However, when many devices are involved, activecongestion will occur, and actual discovery will also beproblematic. Device beaconing system is offered in [12];it makes a discovery in the background of cellular traffic.If the devices are moving, then beaconing design forhigh-speed moving devices is much complicated. Neigh-bour discovery in LTE network, where the distributedorthogonal frequency multiple access (OFDMA) radioresources are used as user identities, is projected in [13],in which device discovery in high dense areas is notconsidered. Recommended technique in [14] uses thebase station for discovery in which signaling flowfrom the base station can be used for discovery. Inthe large number of devices, the signaling overheadand discovery interference become problematic. Bio-inspired and Firefly algorithm used for direct devicediscovery are advised in [15]. Due to the computa-tional complexity, probability of misdetection is high.A device discovery scheme based on clustering hasbeen introduced in [16] for a heterogeneous network,but the deficiency in the work is the sluggish discov-ery, and needs some techniques to decide clustering.All the previous discussions are guided to design anovel algorithm for device discovery which performwell in all aspects.Therefore, in this article, the device’s power-based

sphere decipherer discovery scheme is proposed inwhich the device’s power is incorporated for device dis-covery procedure. Power efficient, minimum collision,and low signaling overhead characterized discovery areachieved based on the utilization of a sphere decoder

like (SDL) algorithm. The primary commitments of thisarticle are encapsulated as follows:

� We propose a device’s power-based device discoveryto enhance the power efficiency and minimize thesignaling overhead of typical D2D discovery.

� We propose disseminated collision fortitude algorithmand scheme which can take care of the D2Dimpairment like device power issue happeningduring the discovery phase.

� We evaluate the performance of the proposedalgorithm by fusing the suppositions in recent LTE-Aspecialized reports [6].

The rest of the paper is organized as follows: Section 2provides the overview of discovery resources which areused in discovery procedure, while Section 3 consists ofthe D2D discovery system model. In Section 4, powerscheme for device discovery is elaborated, and Section 5contains the results’ analysis. In the end, the paper isconcluded in Section 6.

2 Resources for device discoveryIn general, devices in D2D communication may misuseboth downlink and uplink resources for cellular commu-nication [2]. In this work, we expect that device discov-ery uses uplink resources based on reusing the uplinktransmission chain [17] Therefore, to ensure the excel-lent performance of device discovery, shared radio re-sources are suggested alternately with dedicated radioresources. Accordingly, D2D devices, taking an interestin discovery, will choose one radio resource blocks(RRB) among the intermittent discovery radio resources.A case of discovery period and RRB is exhibited inFig. 1. The D2D devices can transmit own discoverysignal on their chosen RRB one time and tune for re-ception of discovery signals from other D2D devices.During the discovery period, each D2D device takesan interest in the discovery procedure only, and othersorts of communications are not permitted.Each device surveys all RRB’s received power level and

selects the RRB which has the most reduced power level[18]. Appropriately, numerous devices situated far awaymay pick a similar resource. On the other hand, everydevice randomly chooses RRB resources for discoverysignal transmission. We concentrate on a random choicedue to the human mobility pattern [19]. There are twoscenarios for device discovery that depend on mobility,haphazard walk scenarios, and velocity scenarios inwhich discovery is computed. Haphazard walk modeldoes not much depend on environment changes, whilemobility depends on context and velocity, which mightbe unknown or partially known or measurable by somemodels [20]. The sensing-based determination is wasteful

Hayat et al. EURASIP Journal on Wireless Communications and Networking (2018) 2018:74 Page 2 of 7

when the sensing outcomes are obsolete rapidly, for ex-ample, under high-mobility situation. When two or moredevices reuse similar discovery resources in the vicinity, acollision may happen because of the asynchronous trans-mission [21]. Accordingly, these neighboring D2D devicescan neither distinguish each other nor be recognized be-cause of the mutual interference.

3 D2D device discovery system modelIn this section, we present a short preface for the D2Ddevice discovery system model. A system model for de-vice discovery where R is the radius of a sphere made bythe discoverer device is shown in Fig. 2. In our analysis,we confine our extension of synchronous D2D devicediscovery, for example, all D2D devices are in timesynchronization and coverage area reference can be ac-quired from the base station downlink transmission.This devours significantly less energy and discovery timecompared with asynchronous. For synchronous devicediscovery, each D2D device can be dynamic within a

predefined discovery time, which shows up intermit-tently, such as D2D device occasionally awakes to ac-complish the discovery process utilizing the D2D RRB.After finishing the periodic discovery, D2D devices startdozing until subsequent discovery period starts. Whena D2D device has discovered a fancied target D2Ddevice by accepting a signal, it can build up a D2Dinterface for direct communication. In cellular net-work topology, the transmission of mobile devices isrelied upon base station.Appropriately, if various cellular devices have a place

with various cells, reuse a common resource in a celledge, a cellular device signal interferes with the neigh-boring base station. Then again, in D2D systems, therecoexist numerous D2D devices which can be both re-ceiver and transmitter. Under this topology, radiatedsignals from various transmitting D2D devices will reachproximal D2D device’s receiver. Note that various re-ceivers are possibly presented to endure high interfer-ence by numerous D2D links. Accepting the quantity ofk D2D devices in D2D systems, the greatest number ofD2D links is k(k − 1), which has a polynomial ratio [18].

4 Power scheme for discoveryIn the LTE-A, D2D users and cellular users are multi-plexed at the same uplink channel orthogonal frequencydivision multiplexing (OFDM) physical RRB. The re-ceived signal from the k cellular devices or D2D devicescan be modelled as [22]

yk ¼ αk;kHk;kT kxk þX

j≠kαk; jHk; jT jx j þ nk ð1Þ

αk; j ¼ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiP jd

−ρk; jχk; j=Nt

qis a scalar depending upon the

total transmitting power Pj for j user, χk, j is shadowing fading,and dk, j is the distance between k receivers and j transmitterswith path loss ρ. xk∈ℂNt�1 normalized data vector with zero

Fig. 1 RRB for discovery Tdis is the total time for discovery and T is the distributed time for each device. Legend: receiving time of reference signal andtransmission time of neighboring devices

Fig. 2 System model for device discovery where R is the radius of asphere made by the discoverer device. Legend: D2D signal andcontrol signal


mean and uncorrelated Εðxkx†kÞ ¼ INt . Hk, j implies (Nr×Nt)channel matrix with Nr which receive antennas andNt which transmit antennas, and Tk refers to kthuser’s diagonal power matrix. To keep the total powerconstant, Tk must satisfy the following condition:

trace T kT†k

� � ¼ XNti¼1 T

i; ið Þk

�� 2 ¼ Nt ∀ k ð2Þnk is additive white Gaussian noise (AWGN) with zero

mean and covariance is

Rnk ¼ Ε nkn†k� � ¼ σ2n INr ∀ k ð3Þ

Signal model can be written as

yk ¼ αk;kHk;kT kxk þ zk þ nk ð4Þzk ¼

Pj≠kαk; jHk; jT jx j is the interference matrix for

D2D and cellular users with covariance

Rzk ¼ Ε zkz†k� � ¼ X

j≠kα2k; jHk; jT jT

†jH

†k; j ð5Þ

The power of the j device is

P j ¼ PtH t; jγdrP

j≠kHk; j−

σ2nPj≠kH t; j

ð6Þ

Ht, j is the channel response between a base stationand j device. Hk, j are channel responses between k de-vices and j devices. γdr is the discovery signal to interfer-ence noise ratio (SINR) at the receiver device. Successfuldevice discovery is only achieved when SINR of de-coding signal is greater than the threshold value ofSINR γthr ; therefore,

Pt ≥γtrX

j≠k

P JHk; jH t; j

þ σ2n

H t; j

� ð7Þ

From (7), P j≥γthr ð 1βPt−σ2nÞ, where β is the ratio ofH t; jHk; j

δt ¼ 0 γdr < γ

th

1 γdr ≥γth

ð8Þ

where δt ∈ {0, 1}. If δt = 0 no discovery, δt = 1 device has beendiscovered and γthr ¼ sIþn in which s, I, and n are the signalpower, interference, and noise power respectively. From (1)

yk ¼ Hk; jxk þ nk ð9ÞBy taking the QR decomposition of Hk, j from (9)

R ¼r11 r12⋯ r1 j0 r22⋯ r2 j⋮ ⋱ ⋮0 ⋯ rjj

2664

3775 ð10Þ

QHyk ¼ Rxk þQHnk

y1y2⋮yk

264

375 ¼

r11 r12⋯ r1 j0 r22⋯ r2 j⋮ ⋱ ⋮0 ⋯ rjj

2664

3775

x1x2⋮xk

264

375þQHnk ð11Þ

The SDL algorithm can reduce the complexity bysearching for the closest device among the possible lat-tice devices that lie within a hypersphere of radius Raround the discoverer device x. Mathematically, the SDLalgorithm solves the problem as:

x̂ ¼ arg y−Hxk k2si∈OM j : y−Hsk k2≤R2

min

ð12Þwhere H∈Rm�n; x∈Rn�1; and ONT show the m-di-mensional lattice points. O shows the device’s constella-tion. The SDL algorithm only searches for the maximumlikelihood devices from among the lattice devices that liewithin a hypersphere of radius R around the discoveryreceived signal as presented in Fig. 3.Calculating the distance between each lattice device to

the discoverer reduces the exhaustive search. The SDLcan cleverly find the set of lattice devices within thehypersphere. This is the simple set of lattice devicesin an interval around the discoverer. Similarly, if weknow the set of lattice devices in a k-dimensionalspace that lies within a hypersphere, then the possiblevalues of the (k + 1)th coordinate for the set of latticedevices [23]. We can thus recursively find and checkall lattice devices in the hypersphere. Mathematically,it can be expressed as:

H ¼ Q�

RN j�N j0 Nk−N jð Þ�N j

�ð13Þ

Fig. 3 Hypersphere of the radius R


where the unitary matrix is partitioned as Q = [Q1 Q2]such that Q1 contains the first NT columns of Q. With alittle manipulation, it can be shown that the hypersphereconstraint becomes

R2≥ QH1 y|{z}x

−Rx

2

ð14Þ

It also can be written as

R2≥XM j

K¼1 yK−XM j

l¼K rK ;l xl� �2

ð15Þ

where yK, xK, are the kth element of y and x. Also rk, l isthe (k, l)th element of the upper triangular matrix R. Tosatisfy the condition in (14), a necessary but not suffi-cient condition is

R2≥ yN j−rN j ;N jxN j

� �2ð16Þ

This implies the following condition on xN j

yN j−R

rN j;N j

& ’≤xN j ≤

yN j þ RrN j;N j

$ %ð17Þ

Likewise, if we define yðMT−1jmT Þ ¼ yMT−1−rMT−1;MT sMT ,we get the following necessary condition on sMT−1

y N j−1jn jð Þ−RN j−1rN j−1; N j−1

& ’≤xM j ≤

y N j−1jn jð Þ þ RN j−1rN j−1;N j−1

$ %ð18Þ

The SDL algorithm proceeds similarly to obtain thelattice devices within the hypersphere.

5 Result analysisA simulation setup model [24] is established to verifythe proposed discovery scheme. A centralized modelwith the discovery area 500 × 500 m and 50 devices aredeployed randomly using Poisson point process (PPP) asshown in Fig. 4. Dotted lines are connected devices,qualified for direct D2D communication and relay com-munication. RRB selection is performed based on re-ceived power of discovery resources. These devices senddiscovery signal through a base station on given RRB todiscover the neighbor devices. RRBs are not chosen withthe lowest received power which causes to reduce theprobability of discovery of devices. If the required de-vice is to be found in its proximal using the condi-tion (8), then the base station will allow for D2Dcommunication. Using power conditions, some de-vices may make D2D LAN via relay devices underthe control link of the base station.Discovery ratio depends upon the searching power of

each device that lies in the cellular network area as

expressed in (6). Devices in the cell area make latticestructure, and each device from the lattice has a radiusR. By the SDL algorithm, the distance of each devicefrom the discoverer is calculated in lattice devices. QRfactorization helps to minimize the searching power asexpressed in (18) to find out the proximal devices. Per-formance of proposed QR factorized lattice structurebase scheme is shown in Fig. 5, and the results are com-pared with those of the Gaussian method. In the idealsituation, minimum discovery cycles discover maximumdevices. In our proposed model, discovery ratio meetsthe ideal. At five discovery signals, near about 1000 de-vices have been discovered.Further, we applied our proposed scheme on central-

ized and distributed system and compared the results.Centralized and distributed scheme for device discovery

Fig. 4 Randomly deployed devices in a cellular network in500 × 500 m area

Fig. 5 Performance of proposed QR factorized lattice structure basescheme. Legend: total result, QR decomposition, and Gaussianelimination and substitution


has been discussed in the Section 1. In the distributedplan, the main issue is power consumption and signalingoverhead. So, in this proposed model, these problemshave been overcome as shown in Fig. 6 and 7 respect-ively. Centralized approache is better in-term of powerconsumption and signaling overhead respectively. In thein-band cellular network device discovery, a centralizedapproach 0.1 mw power is consumed on discovery signalwhen 15 devices have been discovered.When the SDL algorithm with QR factorization is ap-

plied, the signaling overhead significantly decreases asshown in Fig. 8. Fast device discovery depends upon thenumber of transmitted discovery signal. If the discoverysignal is significant, then signaling overhead occurs. Inour proposed method in a centralized approach, signal-ing overhead decreases. Initially, five devices to discover

the proximal users need 10 signals, and 10 devices need20 signals; therefore, signaling overhead decrease in acentralized approach.

6 ConclusionsD2D communication isolates the local cellular traffic tothe global cellular traffic. Accordingly, device discoveryis an underlying strategy in D2D communication. In thisresearch, device’s power-based discovery method isproposed in the centralized cellular network. This clearapproach for device discovery enhances the accomplish-ment by alleviating the discovery issues as fast discovery,minimum signaling overhead, and power consumption.Our proposed methodology is the QR factorized latticestructure of devices in a specific area. Discoverer devicecreates a hypersphere around it and sends a discoverysignal in the hypersphere using sphere decoder like analgorithm. Discoverer device visits all the lattice devicesthat lie in the hypersphere and finds the closest one toinitiate the D2D communication. Using the hypersphere,the designed discovery scheme saves the discovery timeand power for searching of unwanted devices. In future,this plan can be implemented on a distributed systemand can spare the time by exhaustive search.

AcknowledgementsThe authors would like to express their gratitude to the Ministry of HigherEducation (MOHE) in Malaysia and Universiti Teknologi Malaysia (UTM) forproviding the financial support for this research through the HICOE grant (R.J130000.7823.4J215). The grant is managed by Research Management Center(RMC) at UTM.

FundingThe Ministry of Higher Education (MOHE) in Malaysia and the UniversitiTeknologi Malaysia (UTM) are providing the financial support for thisresearch through the HICOE grant.

Fig. 6 Power consumption in j devices while power Pj scale is in nano.Legend: distributed approach and centralized approach

Fig. 7 Signaling overhead for distributed and centralized approaches.Legend: distributed approach and centralized approach

Fig. 8 Signaling overhead comparison of different schemes withproposed scheme. Legend: total result, QR decomposition, updateand back substitution, and centralized approach


Availability of data and materialsMostly, I got the writing material from different journals as presented in thereferences. A MATLAB tool has been used to simulate my concept.

Authors’ contributionsOH, an assistant professor at NUML, H-9, Islamabad, and a student of PhD inWireless Communication Center (WCC), Faculty of Electrical Engineering,UTM Johor Baharu, Malaysia under the supervision of associate professor RN,implemented the idea regarding device discovery. YZ, assistant professor atDepartment of Information and Communication Engineering, Basrah Univer-sity College of Science and Technology, Basrah, Iraq, helped to implementand write this paper. Overall, this paper is built under the approval andsupervision of RN. All authors read and approved the final manuscript.

Competing interestsThe authors declare that they have no competing interests.

Publisher’s NoteSpringer Nature remains neutral with regard to jurisdictional claims inpublished maps and institutional affiliations.

Author details1Department of Engineering, National University of Modern Language(NUML), H-9, Islamabad, Pakistan. 2Wireless Communication Centre (WCC),Faculty of Electrical Engineering, Universiti Teknologi Malaysia, 81310 JohorBahru, Malaysia. 3Department of Information and CommunicationEngineering, Basrah University College of Science and Technology Basrah,Iraq.

Received: 11 September 2017 Accepted: 18 March 2018

References1. M Agiwal, A Roy, N Saxena, "Next generation 5G wireless networks: A

comprehensive survey," IEEE Communications Surveys & Tutorials. 18(3),1617–1655 (2016)

2. O Hayat, R Ngah, Y Zahedi, "Cooperative Device-to-Device Discovery Modelfor Multiuser and OFDMA Network Base Neighbour Discovery in In-Band 5GCellular Networks," Wireless Personal Communications. 97(3) 4681–4695(2017)

3. M Dohler, T Nakamura, 5G mobile and wireless communicationstechnology. Cambridge University Press, University Printing House,Cambridge, (2016)

4. G Fodor, S Roger, N Rajatheva, SB Slimane, T Svensson, P Popovski, et al., Anoverview of device-to-device communications technology components inMETIS. IEEE Access 4, 3288–3299 (2016)

5. K Sharmila, V Mohan, C Ramesh, SP Munda, in Advances in Electrical, Electronics,Information, Communication and Bio-Informatics (AEEICB), 2016 2nd InternationalConference on. Proximity services based device-to-device framework design fordirect discovery (2016), pp. 499–502

6. D Tsolkas, N Passas, L Merakos, Device discovery in LTE networks: a radioaccess perspective. Comput. Netw. 106, 245–259 (2016)

7. Z Wu, VD Park, J Li, Enabling device to device broadcast for LTE cellular networks.IEEE Journal on Selected Areas in Communications 34, 58–70 (2016)

8. S-Y Lien, C-C Chien, F-M Tseng, T-C Ho, 3GPP device-to-device communicationsfor BeyonD 4G cellular networks. IEEE Commun. Mag. 54, 29–35 (2016)

9. Qualcomm Technologies.Inc, "LTE Direct Always-on Device-to-DeviceProximal Discovery," Qualcomm Technologies, 5775 Morehouse Drive SanDiego, CA 92121 U.S.A. (2014)

10. S Park, S Choi, in 2014 IEEE Global Communications Conference. ExpeditingD2D discovery by using temporary discovery resource (2014), pp. 4839–4844

11. L Zhang, J Luo, D Guo, Neighbor discovery for wireless networks viacompressed sensing. Perform. Eval. 70, 457–471 (2013)

12. K Doppler, CB Ribeiro, J Kneckt, inWireless Communication, Vehicular Technology,Information Theory and Aerospace & Electronic Systems Technology (Wireless VITAE),2011 2nd International Conference on. Advances in D2D communications: energyefficient service and device discovery radio (2011), pp. 1–6

13. F Baccelli, N Khude, R Laroia, J Li, T Richardson, S Shakkottai, et al., in 2012Fourth International Conference on Communication Systems and Networks(COMSNETS 2012). On the design of device-to-device autonomous discovery(2012), pp. 1–9

14. H Tang, Z Ding, BC Levy, Enabling D2D communications through neighbordiscovery in LTE cellular networks. IEEE Trans. Signal Process. 62, 5157–5170(2014)

15. S-L Chao, H-Y Lee, C-C Chou, H-Y Wei, Bio-inspired proximity discovery andsynchronization for D2D communications. IEEE Commun. Lett. 17, 2300–2303 (2013)

16. D Xenakis, L Merakos, M Kountouris, N Passas, C Verikoukis, Distancedistributions and proximity estimation given knowledge of the heterogeneousnetwork layout. IEEE Trans. Wirel. Commun. 14, 5498–5512 (2015)

17. CJ Katila, C Buratti, MD Abrignani, R Verdone, Neighbors-aware proportionalfair scheduling for future wireless networks with mixed MAC protocols.EURASIP J. Wirel. Commun. Netw. 2017, 93 (2017)

18. J Hong, S Park, S Choi, "Novel power control and collision resolutionschemes for device-to-device discovery," Peer-to-Peer networking andapplications, 9(5) 913–922 (2016)

19. Y Liu, C Liu, NJ Yuan, L Duan, Y Fu, H Xiong, et al., Intelligent bus routingwith heterogeneous human mobility patterns. Knowl. Inf. Syst. 50, 383–415(2017)

20. AH Sayed, A Tarighat, N Khajehnouri, Network-based wireless location:challenges faced in developing techniques for accurate wireless locationinformation. IEEE Signal Process. Mag. 22, 24–40 (2005)

21. M Klügel, W. Kellerer, "Dominant factors for device-to-device occurrenceprobabilities in cellular networks," Wireless Networks, 1–13 (2017)

22. H Sun, M Sheng, X Wang, Y Zhang, Y Shi, Distributed cooperative device-to-device transmissions underlaying cellular networks. Wirel. Netw 21, 1411–1423(2015)

23. O Hayat, M Qaisrani, M Akbar, in Emerging Technologies (ICET), 2010 6thInternational Conference on. Sphere decoding in the presence of channeluncertainty, 216–220 (2010)

24 N. Papalexidis, "Distributed Algorithms for Beamforming in Wireless SensorNetworks," DTIC Document, Naval Postgraduate School Monterey, CA, June2007


AbstractIntroductionResources for device discoveryD2D device discovery system modelPower scheme for discoveryResult analysisConclusionsAcknowledgementsFundingAvailability of data and materialsAuthors’ contributionsCompeting interestsPublisher’s NoteAuthor detailsReferences

device discovery for d2d communication in in-band cellular … · 2018. 4. 3. · device discovery...

Documents