NOMA Codebook Optimization by Batch Gradient Descent

The non-orthogonal multiple access (NOMA) has the potential to improve the spectrum efficiency and the user connectivity compared to the orthogonal schemes. The codebook design is crucial for the the performance of the NOMA system. In this paper, we comprehensively investigate the NOMA codebook design involving the characteristics from multiple signal domains. The minimizing of the pairwise error probability is considered as the target of the optimization. The neural network framework is explored for the optimization, and the mapping functions on the edges are considered as weights. The method of batch gradient descent is applied for optimizing the weights and correspondingly the codebook. The simulation results reveal that with the optimized codebook the error performance is significantly improved compared to the schemes in the literature.

[1]  Ming Xiao,et al.  Joint Multiuser Detection of Multidimensional Constellations Over Fading Channels , 2017, IEEE Transactions on Communications.

[2]  Kai Niu,et al.  MP-WFRFT and constellation scrambling based physical layer security system , 2016 .

[3]  Shaoguo Wen,et al.  Non-Orthogonal Multiuser Transmission through Constellation Rearrangement , 2019, 2019 IEEE Wireless Communications and Networking Conference (WCNC).

[4]  Yan Chen,et al.  Capacity analysis for non-orthogonal overloading transmissions under constellation constraints , 2015, 2015 International Conference on Wireless Communications & Signal Processing (WCSP).

[5]  F. Alberge,et al.  Constellation design with deep learning for downlink non-orthogonal multiple access , 2018, 2018 IEEE 29th Annual International Symposium on Personal, Indoor and Mobile Radio Communications (PIMRC).

[6]  Guan Gui,et al.  Deep Learning for an Effective Nonorthogonal Multiple Access Scheme , 2018, IEEE Transactions on Vehicular Technology.

[7]  Fumiyuki Adachi,et al.  Deep-Learning-Based Millimeter-Wave Massive MIMO for Hybrid Precoding , 2019, IEEE Transactions on Vehicular Technology.

[8]  Sha Xuejun,et al.  Pattern matrix design of PDMA for 5G UL applications , 2016, China Communications.

[9]  B. Sundar Rajan,et al.  On Two-User Gaussian Multiple Access Channels With Finite Input Constellations , 2011, IEEE Transactions on Information Theory.

[10]  Xin Su,et al.  Multiuser detection algorithm for PDMA uplink system based on SIC and MMSE , 2016, 2016 IEEE/CIC International Conference on Communications in China (ICCC).

[11]  Shaoguo Wen,et al.  Optimization of the Factor Graph for the Multiuser Superposition Transmission , 2018, 2018 IEEE Information Theory Workshop (ITW).

[12]  Hong Wen,et al.  The Rayleigh Fading Channel Prediction via Deep Learning , 2018, Wirel. Commun. Mob. Comput..

[13]  Pingzhi Fan,et al.  The Application of Machine Learning in mmWave-NOMA Systems , 2018, 2018 IEEE 87th Vehicular Technology Conference (VTC Spring).

[14]  Ming Xiao,et al.  Error Performance of Sparse Code Multiple Access Networks with Joint ML Detection , 2016, 2016 IEEE 83rd Vehicular Technology Conference (VTC Spring).

[15]  Cong Shen,et al.  An Iterative BP-CNN Architecture for Channel Decoding , 2017, IEEE Journal of Selected Topics in Signal Processing.

[16]  Gaojie Chen,et al.  A Deep Learning-Based Approach to Power Minimization in Multi-Carrier NOMA With SWIPT , 2019, IEEE Access.

[17]  Zhongwei Si,et al.  Optimization of Power Allocation for Multi-User Superposition Transmission Systems , 2018, 2018 IEEE International Conference on Communications Workshops (ICC Workshops).

[18]  Nam-I Kim,et al.  Deep Learning-Aided SCMA , 2018, IEEE Communications Letters.

[19]  Zhongwei Si,et al.  Improved spatially-coupled multiuser transmission via constellation rotation , 2017, 2017 IEEE 28th Annual International Symposium on Personal, Indoor, and Mobile Radio Communications (PIMRC).