Outage Based Power Allocation: Slepian-Wolf Relaying Viewpoint

Cloud-processing techniques in wireless network are emerging in recent years. Due to the resource energy restriction, power optimization schemes are demanded for designing future transmission strategies. However, it is not realistic to consider the techniques of the whole cloud network having massive amount of nodes. Instead, it is quite reasonable that we first decompose the whole network topology into several typical and simple structures, and identify the most suitable strategies, in general, for them. This work presents an optimal power allocation scheme for a simple Slepian-Wolf relay system over block Rayleigh fading channels. In the assumed decode-and-forward relay model, the information bit sequence obtained as the result of decoding at the relay may contains some errors, however they are highly correlated with the original information bit sequence sent from the source. It is well known that the exploitation of the correlation knowledge between the sequences from the source and relay nodes at the destination achieves significant performance improvement, according to the Slepian-Wolf theorem. In this paper, we obtain an approximated closed-form outage probability expression, based on our previous work for the outage analysis. It is shown that the power allocation for the proposed Slepian-Wolf relay system can be formulated as a convex optimization problem. Specifically, we aim to minimize the outage probability while keeping the total power fixed, and to minimize the total power under an given outage threshold. It is shown that the data obtained from the optimization is very close to the results of numerical calculation.