Analytical evaluation of adaptive coding for Markov models of Nakagami fading

Packet radio systems that communicate over channels with slow fading experience changes in propagation loss from one packet to the next. In order to maintain high throughput, it is necessary for the packet radios to adapt their transmissions in response to changes in the channel. Practical adaptive coding protocols can respond to such changes by adjusting the code rate to maximize the expected throughput for each packet transmission. Performance evaluations for practical protocols typically require simulation of both the time-varying fading process and the adaptive protocol. We employ recently developed finite-state Markov models of Nakagami-m fading to obtain analytical evaluations of the throughput of an adaptive coding protocol. In our approach, neither the fading processes nor their Markov models are simulated.