Power Allocation Strategies for Fixed-Gain Half-Duplex Amplify-and-Forward Relaying in Nakagami-m Fading

TL;DR

This paper investigates power allocation strategies for fixed-gain amplify-and-forward relay networks in Nakagami-m fading, optimizing SNR and power consumption under various CSI conditions, with practical closed-form solutions and numerical validation.

Contribution

It provides new convex and concave formulations for power optimization in relay networks, deriving closed-form solutions for different CSI scenarios and analyzing the trade-offs involved.

Findings

Full CSI improves SNR maximization and power minimization.

Partial CSI offers near-optimal performance with reduced overhead.

Convex and concave problem formulations enable efficient solutions.

Abstract

In this paper, we study power allocation strategies for a fixed-gain amplify-and-forward relay network employing multiple relays. We consider two optimization problems for the relay network: 1) optimal power allocation to maximize the end-to-end signal-to-noise ratio (SNR) and 2) minimizing the total consumed power while maintaining the end-to-end SNR over a threshold value. We investigate these two problems for two relaying protocols of all-participate relaying and selective relaying and multiple cases of available channel state information (CSI) at the relays. We show that the SNR maximization problem is concave and the power minimization problem is convex for all protocols and CSI cases considered. We obtain closed-form expressions for the two problems in the case for full CSI and CSI of all the relay-destination links at the relays and solve the problems through convex programming when full CSI or CSI of the relay-destination links are not available at the relays. Numerical results show the benefit of having full CSI at the relays for both optimization problems. However, they also show that CSI overhead can be reduced by having only partial CSI at the relays with only a small degradation in performance.