Approximate MAP Decoding on Tail-Biting Trellises

TL;DR

This paper introduces two approximate MAP decoding algorithms for tail-biting trellises that select specific nodes, demonstrating their effectiveness through simulations on the Golay code and a convolutional code over an AWGN channel.

Contribution

The paper presents novel approximate decoding algorithms that reduce complexity by selecting node subsets in tail-biting trellises, with validated performance through simulations.

Findings

Effective decoding on the Golay code and convolutional code

Reduced computational complexity compared to exact methods

Successful application on AWGN channel simulations

Abstract

We propose two approximate algorithms for MAP decoding on tail-biting trellises. The algorithms work on a subset of nodes of the tail-biting trellis, judiciously selected. We report the results of simulations on an AWGN channel using the approximate algorithms on tail-biting trellises for the $(24,12)$ Extended Golay Code and a rate 1/2 convolutional code with memory 6.