Optimum and sub-optimum detection of coded data disturbed by time-varying intersymbol interference (applicable to digital mobile radio receivers)

Abstract

e AbstractIn this contribution the MLSE approach is applied to the concatenation of equalizer and channel decoder. An equalization algorithm is presented which delivers the optimal soft-decision information for the channel decoder in case of ideal interleaving. From computer simulations for an application example a significant soft-decision gain can be obtained. Simplified variants with a complexity comparable to a conventional Viterbi equalizer are presented which perform almost as good as the optimum.

Comparative study of turbo decoding techniques: an overview

“…Similarly should give the logarithm of the probability of the received sequence given only that the trellis is in state at stage . However, the maximization in (25) used in the recursive calculation of the terms means that only the most likely path from state to the end of the trellis is considered, and not all paths.…”

“…In the forthcoming section, we first describe two related algorithms, the Max-Log-MAP [25], [26] and the Log-MAP [7], which are derived from the MAP algorithm, and then another, referred to as the SOVA [8], [9], [19], derived from the Viterbi algorithm.…”

“…Forward and backward recursions, both similar to the forward recursion used in the Viterbi algorithm, are invoked to calculate using (24) and employing (25). The branch metric is given by (26), where the constant term can be omitted.…”

“…However, its application in turbo codes renewed interest in the algorithm, and it was realized that its complexity can be dramatically reduced without affecting its performance. Initially the Max-Log-MAP algorithm was proposed by Koch and Baier [25] and Erfanian et al [26]. This technique simplified the MAP algorithm by transferring the recursions into the log domain and invoking an approximation to dramatically reduce the complexity.…”

Turbo Equalization: An Overview

“…This is true in particular for channels with large delay spread (memory) or given a large size of the signal alphabet, since the receiver complexity is of order per symbol. State-of-the-art systems for a variety of communication channels employ convolutional codes and ML equalizers together with an interleaver after the encoder and a deinterleaver before the decoder [8], [9]. Interleaving shuffles symbols within a given block of data to decorrelate error events introduced, or unresolved, by the equalizer between neighboring symbols.…”