Node and symbol power allocation in time-varying amplify-and-forward dual-hop relay channels

In this paper, we propose a simple power-allocation technique between the source and the relay node and between the data and pilot symbols in a dual-hop amplify-and-forward (AF) relay channel, which is subject to unknown but correlated channel variations. For this purpose, we prove tight lower and upper bounds for the channel-estimation-error variance and use it in an effective signal-to-noise ratio (SNR) to optimize the power-allocation coefficients in the system. The power-allocation solution depends on long-term channel statistics and system parameters and hence can be used in practice. We demonstrate the efficacy of the proposed power-allocation technique by comparing it with equal power allocation (EPA) between nodes and symbols and with well-known and recently derived capacity performance benchmarks. Numerical results indicate that, by proper power allocation, we are able to get closer to harvesting the inherent information rate of correlated dual-hop relay channels.