Performance Analysis and Power Allocation of AF Best Relay Selection in Generic Noise

In this paper, the authors provide a high Signal-to-Noise Ratio (SNR) error rate analysis of Amplify-and-Forward (AF) best relay selection in networks impaired by Rayleigh fading and generic, possibly non - Gaussian noise. The derived closed - form asymptotic error rate expressions are applicable to any type of noise with finite moments. Exploiting these analytical results, they formulate a power allocation problem with energy consumption fairness constraints for the source and the relays, which is solved using monomial fitting and a series of geometric programs.