Optimal power allocation for relay-based cooperative communication systems with energy harvesting

In this paper, the problem of throughput maximization in cooperative communication systems is considered, in which the nodes use energy harvesters instead of conventional energy sources. The 3-node Gaussian relay channel with decode-and-forward relaying scheme is investigated, where the source and relay harvest energy from the environment. In this work, optimal power levels for 2 cases are obtained. First, a deterministic model is considered, where the noncausal (offline) knowledge of the harvested energy and channel states is available. The problem is shown to be a convex optimization problem, where the solutions to which are derived using convex optimization techniques. Then, the online case is investigated, in which only the casual knowledge of the harvested energy and channel states are used. In the online setting, the problem is formulated as a Markov decision process. The performance of the system in both offline and online cases, along with a few suboptimal schemes introduced in the literature, is evaluated using computer simulations.