Scalable SCMA Jianglei Ma Sept. 24., 2017

Transcription

1 Scalable SCMA Jianglei Ma Sept. 24., 2017 Page 1

2 5G-NR Air-Interface embb SoftAI: Programmable Air-Interface Adaptive numerology Adaptive transmission duration Adaptive multiple access scheme Adaptive HARQ New Technologies for Future Releases embb f-ofdm mmtc URLLC 2 Page 2

3 From Orthogonal Multiple Access to Non-orthogonal Multiple Access Dual-layer non-orthogonality for SoftAI Inter-subband non-orthogonality: Subband based numerology optimization enabled by spectrum localized waveform from filtering/windowing Intra-subband non-orthogonality Non-orthogonal multiple access UE1 UE2 UE3 UEn UE1 Numerology-2 UE1 Numerology-1 Intra-band Non-Orthgonality -NOMA f-ofdm Inter-band Non-Orthgonality -f/w-ofdm time 3 Page 3

4 Multiple Access Scheme Evolution single dimensional multi-ue multiplexing User 2 User 3 User 4 User 2 User 3 time Two dimensional multi-ue multiplexing frequency Three dimensional multi-ue multiplexing (NoMA) Code frequency frequency User 3 time Code time User 3 Power domain and spatial domain separation can be further exploited 4 Page 4

5 NoMA Application Scenarios Scenarios Motivations embb Enhance spectrum efficiency Improve edge UE experience Enable seamless mobility Support more UEs mmtc Support massive connectivity Reduce signaling overhead Reduce UE energy consumption URLLC Reduce latency Improve reliability 5 Page 5

6 Non-orthogonal Multiple Access for UE Experience Enhancement Cloud Processor C-RAN Multi-TRP Cooperation for guaranteed data rate everywhere Transparent TRP HO for seamless mobility 6 Page 6

7 Non-orthogonal Multiple Access for Latency/Signaling Overhead/Energy Consumption Reduction SR/Grant Grant-free UE gnb UE gnb Scheduling request Scheduling grant Data Data SR Grant Data f Collision DL UL More delay and overhead for grant based TDD transmission t 7 Grant-free for sporadic small packet transmission Low latency Low signaling overhead Low energy consumption However potential UE collision Non-orthogonal Multiple Access Robust to UE collision Better SE Page 7

8 Grant-free for Different UE States NR Cell Active-State Grant Free mmbb URLLC UE-ID Echo State (Inactive-State) UE State Idle-State embb mmtc mmtc URLLC 8 Page 8

9 Non-orthogonal Multiple Access for Hyper Connectivity f UE1 UE2 UE3 UEn UE group-2 UE group-1 Asynchronous Group-2 f-ofdm t Group-1 9 Superposed transmission to support more concurrent users Enhanced connectivity Synchronous transmission is maintained within UE group without TA. Close proximity based local time synchronization Asynchronous transmission is allowed between UE groups. Time synchronization is not required between different groups. Sub-band filter is applied to eliminate int.er-group interference Page 9

10 Non-orthogonal Multiple Access Construction Non-orthogonal multiple access has been studied in Rel-14 NR study item phase. All proposed non-orthogonal MA schemes follow the following basic representation Bit level operations Symbol level operations FEC Bit-Level Interleaver/Scrambler Modulated Symbol Sequence Generator Symbol to Mapping Bit level Generate Non-zero Modulated Components Resource Mapping 10 Non-orthogonal multiple access components: Symbol level operation: Without spreading or with spreading (linear or non-liner) Resource mapping Sparse or on-sparse Page 10

11 Non-orthogonal Multiple Access - Symbol Level Operation Without spreading Power domain superposition Rely on advanced Rx to separate users With spreading Additional code domain separation Better inter-user interference suppression Linear spreading Symbol level repetition Spreading code optimization to allow more UE collision Non-linear spreading Symbol level non-repetitive spreading Additional code gain/diversity gain FEC b 0 b 1 b 32 Symbol Level Operation Linear spreading [ y ] = [ S ] [ x (b) ] Non-linear spreading: [ y ] = [ S(x) ] [ x (b) ] y 0 y 1 y 2 11 Page 11

12 Resource Mapping: Partial Collision With Sparse Spreading f f t t Full Collision UE1 UE2 Non-sparse Resource Mapping Partial Collision UE1 UE2 Sparse Resource Mapping 12 Sparse mapping: Partial collision reduces inter-ue collision Partial collision reduces multi-user detection complexity Page 12

13 SCMA (Sparse Coded Multiple Access) NoMA with Sparse Non-linear Spreading Example 1: SCMA 4-point codebook Example 2: SCMA 8-point codebook b 0 Bit stream-0 x 0 b 0 b 1 Bit stream-0 Bit stream-2 x 0 x 1 y b 1 b 2 Bit Stream-1 Bit Stream -2 x 1 x j 0 0 j y Non-zero component-1 Non-zero component-2 Non-zero component-1 Non-zero component Page 13

14 Scalable SCMA System requirements Spectrum Efficiency Coverage Connectivity SCMA Configuration SCMA Parameters SCMA Codebook Codebook Number of codewords of an SCMA codebook: M Spreading factor: K Max number of layers (or codebooks/signatures) : J Number of nonzero elements of each codeword: N SCMA codebook Adaptation to meet different system requirements Page 14

15 Flexible SCMA codebook Design for Different Application Scenarios Muted Active Active inactive Inactive Active M ed Muted Inactive Active Inactive Active For coverage enhancement Long codeword Non-sparse mapping For connectivity enhancement Short codeword Sparse mapping For PAPR reduction Short codeword Only one is active at each time Page 15

16 Benefits of Non-linear Spreading Parameter Value Scheme Receiver Transmission mode UL OFDMA MMSE & Turbo-ML Antenna configuration 1x2 uncorrelated MC-CDMA (linear non-sparse spreading) MMSE-SIC Channel model Channel estimation TDLC-1000 perfect LDS (linear sparse spreading) Turbo-MPA Resource allocation 12 RBs for NoMA, 1RB for OFDMA SCMA (non-linear sparse spreading) Turbo-MPA 16 Page 16

17 Conclusions Non-orthogonal multiple access can be applied to enhance the performance of different applications: embb mmtc URLL Scalable SCMA can be configured to meet different requirements of different services. 17 Page 17

18 Thank you Copyright 2015 Huawei Technologies Co., Ltd. All Rights Reserved. The information in this document may contain predictive statements including, without limitation, statements regarding the future financial and operating results, future product portfolio, new technology, etc. There are a number of factors that could cause actual results and developments to differ materially from those expressed or implied in the predictive statements. Therefore, such information is provided for reference purpose only and constitutes neither an offer nor an acceptance. Huawei may change the information at any time without notice. Page 18