On the Sparsity of Spreading Sequences for NOMA with Reliability Guarantee and Detection Complexity Limitation

• Published in: 2017 IEEE 85th Vehicular Technology Conference (VTC Spring) pp. 1 - 6
• Main Authors: Ting Qi, Wei Feng, Youzheng Wang
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.06.2017
• Subjects: Complexity theory; Multiuser detection; NOMA; Optimization; Reliability; Sparse matrices; Wireless networks
• DOI: 10.1109/VTCSpring.2017.8108689

Non-orthogonal multiple access (NOMA) is a promising technology for handling massive connectivity in future 5G wireless networks. Even though the spectral efficiency can be promoted, the transmission reliability is sometimes poor, due to the introduced multiple access interference (MAI), and the computational complexity for multiuser detection is usually high in NOMA. To solve this problem, we investigate the NOMA regime with sparse multiple-access sequences, so as to leverage sparse signal processing methods, e.g., the message passing algorithm (MPA) for low-complexity and highly- reliable multiuser detection. Particularly, we formulate an optimization problem to design the sparsity of spreading sequences, while maximizing the efficiency of NOMA subject to the allowable symbol error rate (SER) constraint as well as the affordable detection complexity constraint. To evaluate the detection performance of MPA, we give the large-system limit analysis and derive the density evolution method. The impact of variable and function sparsity on the performance is then analyzed. Based on the uncovered characteristics of the optimization problem, an efficient algorithm is proposed to derive the optimal sparsity achieving the desired trade-off between efficiency, reliability and complexity.