Research Capabilities & Future Directions for the 5G RAN and Core Sub-

Systems @ Communications and Networking Department

Prof. Olav Tirkkonen olav.tirkkonen@aalto.fi RAN

Prof. Riku Jäntti riku.jantti@aalto.fi RAN

Dr. Kalle Ruttik kalle.ruttik@aalto.fi RAN implementation on SDR

Prof. Tarik Taleb tarik.taleb@aalto.fi Core

Prof. Raimo Kantola raimo.kantola@aalto.fi Core

Dr. Jose Costa-Requena jose.costa@aalto.fi Core implementation on SDN

+ partners

May 10th, 2016

• Multidisciplinary and open research platform for investigation and experimental evaluation of innovative ideas in networking and services of 5G.

• A common shared testbed for testing and validation of 5G network functions:• Network virtualization and cloud technologies

• Virtualized EPC with network slicing

• Novel RAN solutions & virtualization• Cloud-RAN (LTE)• Continuous Ultra Dense Network (C-UDN)• NB-IOT

• Services that require network responsiveness and end user experience.

• Cybersecurity, end to end security and trust in 5G

Take-5 testbed in Espoo, Finland – part of 5GTNF

User

Equipment

(UE)

Mobility

Management

Entity (MME)

Home Subscriber

Server (HSS)

SDN Control

S/P-GW

Control

Policy Control and

Charging Rules

Function (PCRF)

Data Plane

Control Plane

))))))))

S1-MME

SDN

SDN capable Backhaul

Virtualized EPC including S/P-GW control

User

Equipment

(UE)

))))))))

SDN

Station

(eNodeB)

Base

WiFI SDN AP

INTERNETIP Router

NAT

Nokia

VTT

Coriant

EXFO

Monitoring

AALTO

AALTO

Nokia

http://5gtnf.fi/http://take-5g.org/

• Software Defined Networking (OpenFlow) implementation on EPC

• Fully functioning eMME entity controlling real Nokia eNodeBs

• SDN based mobile backhaul

• Elasticity, scaling• Core network resources match demand

• Centralized management from cloud

• Slicing of the core network• Application specific slices

• Industrial Internet demo with ABB

Virtualized EPC

• Software defined radio implementation of radio access network (L1 & selected L2 features) • Written in C++, runs in Linux environment

(standalone or container)

• Virtual Hardware Enhancement Layer (VHEL) to handle timing and synchronization issues

• USRP as remote radio units

• Support for multiple air interfaces• TD-LTE Rel 8

• NB-IOT

• C-UDN

Aalto Radio access network framework (ARF)

• Network controlled D2D• TD-LTE with network controlled D2D on UL

resources• Interference cancellation: UL user and D2D

user can share a resource• METIS Demo in Mobile World Congress 2015

• Compressed sensing based MAC for massive MTC• TD-LTE UL resource blocks allocated to MTC

MAC• CDMA type of transmission of MTC messages• Join detection of active users and channel

estimation (Compressed sensing)• Interference cancellation

ARF METIS demos

J. Kerttula et al., "Spectrum sharing in D2D enabled HetNet," Dynamic Spectrum Access Networks (DySPAN), 2015 IEEE International Symposium on, Stockholm, 2015, pp. 267-268.doi: 10.1109/DySPAN.2015.7343911Y. Beyene et al., "Spectrum sharing for MTC devices in LTE," Dynamic Spectrum Access Networks (DySPAN), 2015 IEEE International Symposium on, Stockholm, 2015, pp. 269-270.doi: 10.1109/DySPAN.2015.7343912L. Zhou et al., "Creating secondary spectrum usage opportunity for D2D communication with interference cancellation," Dynamic Spectrum Access Networks (DySPAN), 2015 IEEE International Symposium on, Stockholm, 2015, pp. 273-274.

• WID in RP-152284:

“The objective is to specify a radio access for cellular internet of things, based to a great extent on a non-backward-compatible variant of E-UTRA, that addresses improved indoor coverage, support for massive number of low throughput devices, low delay sensitivity, ultra low device cost, low device power consumption and (optimized) network architecture.”

• 3 different modes of operation• Stand-alone: utilizing for example the spectrum currently being

used by GERAN systems as a replacement of one or more GSM carriers.

• Guard band: utilizing the unused resource blocks within a LTE carrier’s guard-band.

• In-band: utilizing resource blocks within a normal LTE carrier.

• 180 kHz UE RF bandwidth for both downlink and uplink

• OFDMA on the downlink

• SC-FDMA or Single tone in uplink

• An NB-IoT UE only needs to support half duplex operations

ARF NB-IoT

IMessage

Router

Interface

Message

time

SF time

S1

RRC

Scheduling

BuffersSubframe

Scheduler

MAC

User DB

Interface

TX A TX B RX A RX BPHY frame

creation

SF content

selection

Dispatcher

VHEL VHEL

Sample time

State machine

Data flow type

processingNB-IoT

Uplink

Transceiver

NB-IoT

Downlink

Transceiver

NB-IoT Protocol

Features Signal repetition for Indoor

coverage Support for C-RAN Flexible software-based

implementation. Standalone operation, for

example, in GSM bands.

NB-Scheduler

NB-MAC

NB-User DB

NB-IoT Downlink Transceiver

NB-PBCH

EnodeB

NB-PSS

NB-SSS

NB-PDCCH

NB-PDSCH

Buffers

Transmit workItem●Control (PSS, SSS, RS)

●System information (MIB, SIB)

●Transport channel (DCI + Data)

NB-IoT Downlink Transceiver

NB-PBCH

UE

NB-PSS

NB-SSS

NB-PDCCH

NB-PDSCH

Buffers

Transmit workItem●Cell acquisition

●Cell tracking

●NB-PDCCH search

●NB-PDSCH search

Dispatcher

VHEL VHEL

Sample time Dispatcher

VHEL VHEL

Sample time

Receive workItem●Cell info (cell ID, frame number)

●System info (MIB,SIB)

●DCI

●Transport block

5G Networks and High-Efficiency Device Positioning: Enabling Techniques, Demonstration and Verticals

• Continuous Ultra Dense Network (C-UDN)• Inter site distance 50 m (e.g. lampposts)• TDD MU-MIMO system

• Dynamic DL/UL selection• ≤ 200 ms frame size• Wide bandwidth ≥ 100 MHz

• Uplink beacons • Channel estimation (reciprocity)• Mobility management (antenna port selection)• Beamforming

• Extreme high localization accuracy at the network side based on the uplink beacons

• Data fusion over multiple measurement points• ToA/TDoA & DoA measurements bases on UL beacons

• Centimeter-scale localization, UE tracking and movement prediction

Clock model

+

UN movement

model

Time-of-arrival

(ToA) estimates

Direction-of-arrival

(DoA) estimates

Extended Kalman filter (EKF)

UN position

estimates

UN clock offset

estimates

Joint User Node Positioning and Clock Offset Estimation in 5G Ultra-Dense Networks: http://www.tut.fi/5G/positioning/media.html

C-UDN Frame structrue

C-UDN concept

Kari Leppänen, Huawei Kari.Leppanen@huawei.comRiku Jäntti, Aalto University riku.jantti@aalto.fiMikko Valkama, Tampere University of Technology mikko.e.valkama@tut.fi

5G Networks and High-Efficiency Device Positioning: Enabling Techniques, Demonstration and Verticals• Take-5 ARF C-UDN test-bed 4 Access Node (AN) sites

4 – 8 antennas per AN Linux server operates as BS

1 antenna in mobile

Constructed of USRP + PC Comm. stack in C++ TDD implemented by using USRP switch

between Tx - Rx chains (not circulator)

Transmission Frequency license 3.41 – 3.43 GHz Sampling rate 15.36 MHz 5 micro sec OFDM symbol length

easy to test also for different values

UN

AN1

AN2

AN3

AN4

8 antenna array

AN & UNC-UDN test-bed architecture

5G Networks and High-Efficiency Device Positioning: Enabling Techniques, Demonstration and Verticals• Testing environment for verticals

• Otaniemi – sub-urban / industrial area• Ruoholahti – urban area

• Verticals• TA7: 5G for Future MTC solutions:

• Device localization, energy efficiency through UE beacon based signaling

• TA20: Open Portfolio Target Action• Verticals benefitting from accurate localization

and low latency• V2X, Factories of the future, e-health,…

Otaniemi test-site (sub-urban / industrial)

Ruoholahti test-site (urban)

V2X

Facories of the future

Test sites

Aalto IndustrialInternet Campushttp://aiic.aalto.fi/en/

Nokia, Elisa, ABB, Konecranes,…

Vertical examples