Performance of cross-layer optimal adaptive transmission techniques over diversity Nakagami-m fading channels

Adaptive modulation and antenna diversity are two important enabling techniques for future wireless network to meet the demand for high data rate transmission. We study a Markov decision process based cross-layer design of optimal adaptation policy over selection-combining and maximal-ratiocombining Nakagami-m fading channel for Markov modulated Poisson process traffic. Unlike most of the channel-dependent adaptation policies in the literature, proposed policy chooses modulation constellation dynamically depending on incoming traffic state and buffer state in addition to channel state. Because the channel-dependent policy is physical layer optimized, it does not guarantee upper-layer delay and overflow requirements. Furthermore, the throughput predicted in physical layer analysis without considering buffer and traffic is over-optimistic than the actual system-level throughput. Proposed cross-layer dynamic adaptation policy minimizes transmission power and also guarantees target bit error rate, delay and packet overflow rate requirements for specific application being considered.