A Unified Framework of New Multiple Access for 5G Systems Xin Su1 ,Bin Fan2, Jie Zeng1, Bei Liu2 1 Tsinghua National Laboratory for Information Science and Technology, Research Institute of Information Technology,Tsinghua University,Beijing,China 2 Broadband Wireless Access Laboratory, Chongqing University of Posts and Telecommunications Email:
[email protected]
Abstract. In order to meet the 5G requirements on spectral efficiency and the number of connections, new multiple access (NMA) is becoming an important technology in 5G. Different from orthogonal multiple access (OMA), NMA technology is tolerant of symbol collisions in orthogonal channels, thereby significantly increasing the number of served users. In this paper, a unified framework of NMA is proposed for the next generation radio access networks. The evaluation results of several NMA schemes, such as sparse code multiple access (SCMA), pattern division multiple access (PDMA) and interleaver-grid multiple access (IGMA), are given. And the results show that NMA can obtain significant block error rate (BLER) performance gain, compared with orthogonal frequency division multiple access (OFDMA). Keywords: 5G, New multiple access, Unified framework.
1
Introduction
From the perspective of an information theory, the wireless channel is a classical multiple access channel [1]. NMA 1 technology not only enhance the spectrum efficiency, but also approach multi-user channel capacity and support overloaded transmission. Furthermore, NMA enable reliable and low latency grant-free transmission and flexible service multiplexing [2]. At present, the industry has put forward a lot of candidate NMA technologies [3], such as SCMA [4], PDMA [5], IGMA [6], multi-user shared access (MUSA) [7], power domain non-orthogonal multiple access (NOMA) [8], resource spread multiple access (RSMA) and so on [9]. The main contribution of this paper is proposing a unified framework of NMA technology, including the principle, application analysis, and performance evaluation. 1
This work was supported by China’s 863 Project (No.2015AA01A709), the National S&T Major Project (No.2016ZX03001017), Science and Technology Program of Beijing (No.D161100001016002), and by Beijing Samsung Telecom R&D Center.
2 2.1
Unified Framework Principle of Framework
All proposed NMA technologies for UL transmission share the common features to suppress inter-user interference and providing overloading capability [10]. To support this operational requirement, this section we proposed a compatible multiple access uplink framework in the following as depicted in Fig.1. This framework could reuse related modules, improve resource utilization and reduce overhead. By this uplink unified framework, we can flexibly configure different multiple access schemes according to various 5G scenarios on the basis of minimizing the hardware functional module.
Fig.1. Unified framework
As depicted in Fig.1, the differences among various multiple access technologies lie in the different realization of interleaver, constellation optimization, factor graph and multiplexing domain. 2.2
Application of Framework
Based on the above framework, wide range of NMA technologies can be merged in this unified structure. SCMA is a code domain NMA technology based on multidimensional modulation design and sparse spreading. PDMA can achieve multiplexing and diversity gain by designing multi-user pattern matrix. MUSA is a NMA technology operating in the plural code domain. IGMA scheme could distinguish different users based on the different combinations of bit-level interleaver and grid mapping pattern. The element of the sparse matrix maybe different among various technologies, e.g. the element of the sparse matrix can only be “0” or “1” in the SCMA and PDMA , and it can also be “-1” in the MUSA. Considering the relatively independent design of interleaver and multiplexing domain modules with the constellation and factor graph and the probability of the hybrid multiple access scheme, we put more emphasize on the design of NMA technologies. For example, in the case of 3 users sharing 2 resources, the common sparse codebook set can be written as the following matrix:
[2,3]
C
e j1,1 j e 1,2
1 e 0
j2,1
0
1 e
j3,2
(1)
where is power scaling factor and is phase shifting factor. The optimal value of power scaling and phase shifting depend on the number of users and the shape of input constellation. The channel coding process in the unified framework can be either using simple repetition or directly using low coding rate forward error correction (FEC). For interleaver, the coded bits of each user is mapped to symbols. The bit/symbol of the other scheduled user not be considered.
3
Performance Evaluation
From all the new multiple access schemes observed, we can see that some schemes have common features based on OFDMA. In the 3GPP RAN WG1 Meeting #86bis, the company has submitted a variety of NMA technologies of the simulation analysis [11]. According to the simulation results, this section is comparing the non-orthogonal SCMA, PDMA and IGMA with the OFDMA multiple access technology. The main evaluation parameters of the uplink LLS are listed in Table 1. Table 1. Evaluation parameters Parameters Carrier Frequency Waveform System Bandwidth Transmission Bandwidth
Assumptions 2GH OFDM 10 MHz 4RB
Antenna configuration
1T2R
Propagation channel Channel coding Channel estimation
TDL-C 300ns, 3km/h LTE Turbo Ideal
Huawei has proved robust to SCMA codebook collision with random codebook allocation [12]. PDMA under different overloading factors enables robust grant-free transmission [13]. IGMA could be beneficial in combating inter-channel interference (ICI) in multi-cell case [14]. The BLER performance comparison between PDMA, OFDMA, IGMA and OFDMA are shown in Fig.2.
Fig. 2. Performance comparison
From the simulation comparison results, we observed the code domain spreading NMA has significant gain over OFDMA.
4
Conclusion
Facing the future huge data traffic, in order to meet the requirements of 5G system, the problem of increasing system capacity and improving spectral efficiency should been considered and solved. In this paper, we proposed an uplink unified framework of NMA and analyzed the feasibility of framework for 5G system. According to the link level simulation results comparison, NMA technologies such as SCMA, IGMA, PDMA, could get better performance gain than OFDMA. Future work of interest is to apply the NMA to software defined radio (SDR) [15], together with the approach of the convergence for three radio solutions: Wi-Fi, iBeacon and ePaper.
References 1. Whitepaper: Alternative Multiple access v1, Future(2015) 2. 3GPP TR 36.913 V0.2.1, Study on Scenarios and Requirements for Next Generation Access Technologies.(2015) 3. 3GPP TSG RAN WG1 Meeting #86 R1-167445: Classification of candidate UL nonorthogonal MA schemes Gothenburg, Sweden, China Telecom (2016) 4. LU Lei, CHEn Yan, GUO Wenting, YAnG Huilian, WU Yiqun, XInG Shuangshuang: Prototype for 5G New Air Interface Technology SCMA and Performance Evaluation.
Huawei Technologies, Co., Ltd., Shanghai, China(2015) 5. Shanzhi Chen, Senior Member, IEEE, Bin Ren, Qiubin Gao, Shaoli Kang, Shaohui Sun, and Kai Niu: Pattern Division Multiple Access (PDMA)-A Novel Non-orthogonal Multiple Access for 5G Radio Networks(2016) 6. 3GPP TSG-RAN WG1 Meeting #85 R1-163992: Non-orthogonal multiple access candidate for NR, Nanjing, China, Samsung(2016)
7. 3GPP TSG RAN WG1 Meeting #84bis R1-162226: Discussion on multiple access for new radio interface, ZTE Busan, Korea, (2016) 8. Peng Xu, Zhiguo Ding, Member, IEEE, Xuchu Dai and H. Vincent Poor, Fellow, IEEE: NOMA: An Information Theoretic Perspective. (2015) 9. 3GPP TSG-RAN WG1 #85 R1-164688: RSMA, Qualcomm Incorporated Nanjing, China (2016) 10. White Paper, “5G SIG”, FuTURE Forum(2015) 11. 3GPP TSG RAN WG1 #86bis Lisbon: Portugal, Summary of updated assumptions and results for NR MA, Agenda item: 8.1.1.2 (2016) 12. 3GPP TSG RAN WG1 Meeting #86 R1-166094: LLS Results for UL MA schemes, Huawei, HiSilicon Gothenburg, Sweden (2016) 13. 3GPP TSG RAN WG1 Meeting #86 R1-167870: LLS results of PDMA,CATT Gothenburg, Sweden(2016) 14. 3GPP TSG RAN WG1 Meeting #86 R1-166750: Link level performance evaluation for IGMA, Samsung Gothenburg, Sweden(2016) 15. Suciu, George, Marius Vochin, Cristian Diaconu, Victor Suciu, and Cristina Butca: Convergence of software defined radio: WiFi, ibeacon and epaper, In RoEduNet Conference: Networking in Education and Research, 2016 15th, pp. 1-5. IEEE(2016)
