Downlink SCMA Codebook Design with Low Error Rate by Maximizing Minimum Euclidean Distance of Superimposed Codewords
Journal
IEEE Transactions on Vehicular Technology
Journal Volume
71
Journal Issue
5
Pages
5231-5245
Date Issued
2022
Author(s)
Abstract
Sparse code multiple access (SCMA), as a codebook-based non-orthogonal multiple access (NOMA) technique, has received research attention in recent years. The codebook design problem for SCMA has also been studied to some extent since codebook choices are highly related to the system's error rate performance. In this paper, we approach the SCMA codebook design problem by formulating an optimization problem to maximize the minimum Euclidean distance (MED) of superimposed codewords under power constraints. While SCMA codebooks with a larger MED are expected to obtain a better BER performance, no optimal SCMA codebook in terms of MED maximization, to the authors' best knowledge, has been reported in the SCMA literature yet. In this paper, a new iterative algorithm based on alternating maximization with exact penalty is proposed for the MED maximization problem. The proposed algorithm, when supplied with appropriate initial points and parameters, achieves a set of codebooks of all users whose MED is larger than any previously reported results. A Lagrange dual problem is derived which provides an upper bound of MED of any set of codebooks. Even though there is still a nonzero gap between the achieved MED and the upper bound given by the dual problem, simulation results demonstrate clear advantages in error rate performances of the proposed set of codebooks over all existing ones not only in AWGN channels but also in some downlink scenarios that fit in 5G/NR applications, making it a good codebook candidate thereof. The proposed set of SCMA codebooks, however, are not shown to outperform existing ones in uplink channels or in the case where non-consecutive OFDMA subcarriers are used. The correctness and accuracy of error curves in the simulation results are further confirmed by the coincidences with the theoretical upper bounds of error rates derived for any given set of codebooks. © 1967-2012 IEEE.
Subjects
5G; alternating maximization; exact penalty; minimum Euclidean distance (MED); mMTC; non-orthogonal multiple access (NOMA); optimization; semidefinite relaxation (SDR); sparse code multiple access (SCMA)
SDGs
Other Subjects
5G mobile communication systems; Error analysis; Frequency division multiple access; Iterative methods; Optimization; Alternating maximization; Downlink; Euclidean distance; Exact penalty; Minimum euclidean distance; Minimum euclidean distances; MMTC; Multiple access; Non-orthogonal; Non-orthogonal multiple access; Optimisations; Semidefinite relaxation; Sparse code multiple access; Sparse codes; Upper Bound; Orthogonal frequency division multiplexing
Type
journal article