Adaptive M-QAM Systems with Diversity in Correlated Nakagami-m Fading and Shadowing

Adaptive M-QAM systems with maximal-ratio diversity (MRD) over composite arbitrarily correlated Nakagami-m fading and log-normal shadowing channels are analyzed. The slow adaptive modulation (SAM) technique is considered which adapts the constellation signaling to slow channel variations and achieves good performance in terms of spectral efficiency (SE) and bit error outage (BEO) despite its low feedback rate. Non-ideal channel estimation (CE) based on pilot symbols is assumed. Analytical expressions and tight bounds for the bit error probability are obtained which are used for deriving analytical expressions and tight bounds for the BEO and achieved SE (ASE). Performance evaluation results have shown that fading correlation and CE errors degrade the performance of SAM systems. The partition of resources for CE and information data requires a careful definition of the ASE and results have shown the tradeoff between CE quality and ASE.