Turbo-equalization for multilevel modulation: an efficient low-complexity scheme

Abstract

This paper addresses the problem of iterative equalization and decoding (turbo-equalization) when bit-interleaved multilevel modulations are considered. The goal is to propose an efficient low-complexity solution in this context, where optimal turbo-equalization is totally untractable. Therefore, we generalize reduced-complexity soft-in/soft-out (SISO) equalizers based on the minimum mean square error (MMSE) criterion (originally proposed in a BPSK context) to this particular scheme. Performances are illustrated through simulations. We show the influence of the mapping on the signal-to-noise ratio (SNR) threshold and on asymptotic performances, and link the obtained results to iterative demodulation and decoding on an additive white Gaussian noise (AWGN) channel.