Protocol analysis of meteor burst communication protocols for rural applications

A novel approach to MBC (meteor burst communication) protocol analysis is presented which differs from previous techniques in that the channel is not assumed to be binary symmetric and instead the channel errors are modeled by the Fritchman model. The large body of existing literature on protocol analysis applicable to continuous channels can generally be applied with few modifications. The protocols are compared on the basis of the expected number of bits transferred per burst. The results of applying this analysis approach to an ARQ (automatic repeat request), a Type-I hybrid, and a Type-II hybrid protocol are presented. It is shown that the hybrid protocols can provide a moderate increase in throughput, compared to the pure ARQ scheme, if the frame size and the error correction capability are chosen to suit the current channel conditions