Protocol optimization using discrete channel statistics

Abstract

This paper presents initial results on the application of Markov control process (MCP) theory to optimization of cost functions related to wireless communication channels subject to fading. Markov control process theory and Markov models of the fading channel are combined to solve the problem of designing a variable packet length protocol for the fading channel. The protocol selects the current packet length as a function of the recent fading history of the channel. This work makes use of previous empirical and analytical results in the construction of Markov descriptions of fast fading.