Performance Analysis of Cooperative Systems with Unreliable Backhauls and Selection Combining

In this paper, a cooperative wireless system with unreliable wireless backhaul connections is investigated. To increase the throughput and maximize the receiver signal-to-noise ratio (SNR), a selection combining protocol is employed. Each transmitter is connected to the control unit via independent but unreliable wireless backhaul connections. Simultaneously taking into account the reliability of each backhaul and different fading conditions of Nakagami-m fading channels, the statistical properties of the effective-SNR at the receiver are investigated. Closed-form expressions are derived for several performance metrics including the outage probability, average spectral efficiency, and average symbol error rate. The effects of backhaul reliability on these performance metrics are also investigated. The scaling relationship between the convergence behavior of these performance metrics and the conventional diversity gain is also analytically investigated in the asymptotic high SNR regime. Monte-Carlo simulations are conducted to verify the derived impact of backhaul reliability on the performance.