Dynamic Spectrum Access in 5G Narayan B. Mandayam WINLAB, Rutgers - PowerPoint PPT Presentation

Dynamic Spectrum Access in 5G Narayan B. Mandayam WINLAB, Rutgers University narayan@winlab.rutgers.edu winlab.rutgers.edu/~narayan 1 WINLAB

What is 5G ?  Wide range of spectrum choices  Wide range of application choices 100s of MHz to 100 GHz, IoT, M2M, D2D Flexible BW, Licensed, Unlicensed V2V  Wide range of device choices  Wide range of QoS requirements Low power, Mid-to-high power Ultra low latency Low complexity, High complexity Very high data rate, Best effort  Wide range of networking choices  Wide range of networking paradigms Mesh, Capillary, Phantom, HetNets ICN, MF, NOM, User-centric  5G: Anything you want it to be!  5G: Academic’s dream ! 2 WINLAB

5G DSA: What’s out there ?  Three distinct approaches to DSA have been proposed  Agile/cognitive radio – autonomous sensing at radio devices to avoid interference  Spectrum Access System (SAS) – centralized Database to provide visibility of potentially interfering networks and/or global assignment  Distributed inter-network collaboration – peering protocols to support decentralized spectrum assignment algorithms 2. SPECTRUM SERVER 3. DECENTRALIZED 1. AGILE RADIO NETWORK COLLABORATION (Collocated Networks) Spectrum Server RF sensing Net B Internet Distributed RF sensing Net A Algorithm Net C Query/ AP/ BS Assignment A AP/ BS B WINLAB

5G DSA: Agile radio Cognitive radio networks require a large of amount of network (and channel)  state information to enable efficient  Discovery, Self-organization  Resource Management  Cooperation Techniques Cost of Cooperation? Scalability? PHY A PHY C PHY B Multi-mode radio PHY Ad-Hoc Discovery & Routing Capability Control (e.g. CSCC) Functionality can be quite challenging! 4 WINLAB

5G DSA: Spectrum Access System (SAS)  Primarily in 3.5 GHz spectrum SPECTRUM SERVER  Small Cells for Cellular  Coexistence with Navy Radar Design Principles and Architecture Internet  Registration with Spectrum Server/Database  Tiering and Prioritization of users Query/ Assignment  Protect Incumbents  Wide range of technical issues related to access  Licensed Shared Access  Generalized Authorized Access  Control and Network State Information  Radio and Network parameters exposed  Coordination across databases  Monitoring and Enforcement 5 WINLAB

5G DSA: Network Cooperation SAVANT: Spectrum Access Via Inter-Network Cooperation  Focus on decentralized architecture for sharing spectrum info  Parallels with BGP exchange of route information between peers  Architecture enables regional visibility for setting radio parameters  Further, networks may collaborate to carry out logically centralized optimization for max throughput subject to policy/technology constraints Local Adaptation to Cooperative Regional Observed Spectrum Use Optimization of Radio Net B Parameters Radio MAP Distributed Algorithm Information Net A Exhange Net C *Supported by NSF EARS grant CNS 1247764 WINLAB/Princeton Project WINLAB

SAVANT: Inter-Network Protocol Architecture involves two protocol interface levels between independent wireless domains: • Lower layer for sharing aggregate radio map using technology neutral parameters • Higher layer for negotiating spectrum use policy, radio resource management (RRM) algorithms, and controller delegation WINLAB

Elephant in Room: WiFi  Smart Phone growth is the U.S. from 2013 to 2015 is ~300%  Smartphone data consumption in 2015 ~10 GB/user/month  ~85% over WiFi and ~15% over Cellular  WiFi AP density in cities ~100-200 per sq km 25 San Francisco % of Enterprise/SP APs New York 20 Chicago Boston 15 10 5 0 01/2009 01/2010 01/2011 01/2012 01/2013 Date  Licensed Assisted Access (LAA) and other cooperative methods including aggregation/integration with WiFi 8 WINLAB

5G DSA: Technical Challenges  Noncontiguous Spectrum Transmission  TX power is no longer “King”! Control Plane Design  Scalability, Performance   Distributed/Hybrid Algorithms for Spectrum Coordination Stability, Convergence of Algorithms  9 WINLAB

Case for Noncontiguous Transmission - I C ? • Three available channels 3 1 2 • Node A transmits to node C via node B. 2 ? B • Node B relays node A’s data and transmits its X own data to node C. • Node X, an external and uncontrollable interferer, transmits in channel 2. A  If we use max-min rate objective and allocate channels, node B requires two channels; node A requires one channel  Scheduling options for Node A and Node B? 10

Case for Noncontiguous Transmission - II #3: Non-Contiguous OFDM #1: Contiguous OFDM #2: Multiple RF front ends (NC-OFDMA) Nulled C C C Subcarrier 3 1 3 1 1 2 B B B 2 2 2 X X 2 X 3 2 A A A • NC-OFDM accesses multiple Spectrum fragmentation • Transmission in link BC limited by number of radio fragmented spectrum chunks suffers interference in with single radio front end front ends channel 2 11 11

NC-OFDM Operation Non-Contiguous OFDM Nulled 0 X[2] = Subcarrier AP 3 1 X[1] x[1] X[1] X[3] Serial to x[2] Parallel Modulation IFFT D/A Parallel x[3] to Serial B X[3] 2 X 2 • Node B places zero in channel 2 and avoids interference • Node A, far from the interferer node X, uses channel 2. A • Both nodes use better channels. NC-OFDM accesses multiple fragmented spectrum chunks • Node B spans three channels, instead of two. with single radio front end • Sampling rate increases. 12

Resource Allocation in Noncontiguous Transmission Benefits:  Avoids interference, incumbent users  Uses better channels  Each front end can use multiple fragmented spectrum chunks Challenges:  Increases sampling rate  Increases ADC & DAC power  Increases amplifier power  Increases peak-to-average-power-ratio (PAPR)  Multiple RF Front Ends vs Single RF Front End ?  Centralized, Distributed and Hybrid algorithms for carrier and forwarder selection, power control ? 13

Spectrum Allocation under Interference and Spectrum Span Constraints Controller Available channels  How to allocate noncontiguous Radio nodes channels subject to ADC/DAC Interference nodes power constraints?

Maxmin Rate Allocation (Integer Linear Program) n 1 n 2 n 3 n 4 B n 5 C A n 6 n 7 n 8 L 1 n 1 -n 2 L 2 n 3 -n 4 L 3 n 5 -n 7 L 4 n 6 -n 8

Control Plane Design: Noncontiguous Transmission CDMA is back! Short PN-seq Long PN-seq Control Channel Data 16 WINLAB

Experimental Results from ORBIT testbed Result 1: Spectrum assignment while minimizing span of assigned subcarriers (reduces ADC/DAC power consumption) USRP ORBIT testbed Network Setup: • Multiple p2p secondary links operating in the presence of a primary transmission • 1 MHz BW, 64-subcarrier NC-OFDM with Reassigned subcarriers with minimal loss (< 10%)of throughput CDMA-based underlay (spreading sequence length 40-160) • Underlay to noise ratio ~ 0 dB, primary Result 2: Reliable timing and frequency recovery from transmission to noise ratio ~ 10 dB underlay control channel in the presence of primary transmissions Result 3: Control channel BER as a correct timing function of primary signal strength with instance underlay to noise ratio set to 0 dB; Control channel rate = 30 kbps peak indicating Primary Signal SNR BER timing instance detection 3 dB < 1e-3 peak detection 6 dB 6.3*1e-3 threshold 7.7 dB 2.6*1e-2 9.2 dB 9.2*1e-2

Network Coordination: LTE/WiFi Conventional LTE Conventional Wi-Fi Spectrum Exclusive licensed Shared unlicensed Operation OFDMA: channel hopping over CSMA/CA: Channel sensing technique time to exploit good channel before transmission to avoid condition packet collision Controller A single licensed carrier No common controller entity Advantage Packet efficient Cost effective, fair sharing 18 WINLAB

Formulating LTE/WiFi Cooperation as an Optimization problem Objective: Downlink power control optimization using Geometric Programming       Maximize sum-throughput across Wi-Fi and LTE       a b maximize 1 1 S w i i S l w i l j   i W j L      subject to ( 1 log ) , , S r i W Minimum SINR requirement for data rate transmission 2 , min w w i i      ( 1 log ) , j , S r L 2 , min l l j j        CCA threshold requirement at Wi-Fi , P G P G N , i W 0 k ik j ij C   b k j L M i    0 , , Range of Tx power P P i W L max i  Controllin g variables : , , P i W L Tx power i where : SINR at link i S i      Set of Wi-Fi APs in the CSMA range of AP a a 1 1 | | , : , a M M i i W i i i       b b Set of Wi-Fi APs in the interference range of AP 1 1 | | , : , b M M i i W i i i 19 WINLAB