Markov modeling of discrete fading channels using birth-death channels

Finite-state Markov channels (FSMCs) are widely employed to model the memory of time-correlated discrete fading channels (DFCs). This work proposes a special class of FSMCs, denoted birth-death channels (BDCs), to model the statistical behavior of the DFC. The number of states that generates accurate BDCs is estimated for a broad range of fading environments. Next, we evaluate the performance of error-correcting codes over the BDC. A statistic of interest for this analysis is the P(m,n), the probability the channel generates up to m errors in a block of length n. It also proposed a combinatorial method to evaluate the probability P(m,n) for the BDC. The study of the statistical description of this channels has applications in modeling real channels and in designing coded systems for channels with memory.