Bounds on the ML Decoding Error Probability of RS-Coded Modulation over AWGN Channels

Abstract: This paper is concerned with bounds on the maximum-likelihood (ML) decoding error probability of Reed-Solomon (RS) codes over additive white Gaussian noise (AWGN) channels. To resolve the difficulty caused by the dependence of the Euclidean distance spectrum on the way of signal mapping, we propose to use random mapping, resulting in an ensemble of RS-coded modulation (RS-CM) systems. For this ensemble of RS-CM systems, analytic bounds are derived, which can be evaluated from the known (symbol-level) Hamming distance spectrum. Also presented in this paper are simulation-based bounds, which are applicable to any specific RS-CM system and can be evaluated by the aid of a list decoding (in the Euclidean space) algorithm. The simulation-based bounds do not need distance spectrum and are numerically tight for short RS codes in the regime where the word error rate (WER) is not too low. Numerical comparison results are relevant in at least three aspects. First, in the short code length regime, RS-CM using BPSK modulation with random mapping has a better performance than binary random linear codes. Second, RS-CM with random mapping (time varying) can have a better performance than with specific mapping. Third, numerical results show that the recently proposed Chase-type decoding algorithm is essentially the ML decoding algorithm for short RS codes.