Modeling Fading Channels With Binary Erasure Finite-State Markov Channels

TLDR: An enumerative technique to analytically evaluate the performance of a linear block code over the ternary Markov channel (TMC) and a new recurrence expression is developed for calculating the probability of codeword error for bounded distance error and erasure decoding.

Abstract: Finite-state Markov channels (FSMCs) are commonly used to characterize the memory of binary (binary-input, binary-output) discrete fading channels (DFCs). This paper proposes a new FSMC model, called the ternary Markov channel (TMC), to characterize ternary (binary-input, ternary-output) discrete channels with memory and soft-information. The TMC is used to model a DFC with the simplest soft-decision quantizer that labels low reliable received symbols as erasure symbols. This paper also proposes an enumerative technique to analytically evaluate the performance of a linear block code over the TMC. A new recurrence expression is developed for calculating the probability of codeword error for bounded distance error and erasure decoding.