Optimisation of convergence‐aware coded PDMA for 5G wireless systems

Abstract

Orthogonal frequency division multiplexing-based pattern division multiple access (PDMA) is considered as a promising multiple access scheme for future wireless communication systems due to its high spectral efficiency and powerful ability to support massive users. In this study, it is shown that two pattern matrices with unequal diversity gain in the PDMA system have better convergence performance compared with the signature matrices in a traditional low-density signature multiple access system by simulations and extrinsic information transfer (EXIT) chart analysis. For a low-density parity check (LDPC)-coded PDMA system, the EXIT characteristics between the front-end multiple-user detector (MUD) and the LDPC decoder may be mismatch, which usually leads to degraded performance when an iterative detection and decoding algorithm is used. To overcome this problem, a two-stage iterative optimisation algorithm, which is easy to implement is proposed to find an optimal (near optimal) degree distribution pair for the LDPC code. Both EXIT chart analysis and simulation results demonstrate that the proposed algorithm exhibits further performance gain compared with regular (3,6) LDPC-coded systems under both an additive white Gaussian noise channel and a Rayleigh fading channel.