In this paper, we study the two-way relay channel with energy harvesting nodes. In particular, we find transmission policies that maximize the sum-throughput for two-way relay channels when the relay does not employ a data buffer. The relay can perform decode-and-forward, compress-and-forward, compute-and-forward or amplify-and-forward relaying. Furthermore, we consider throughput improvement by dynamically choosing relaying strategies, resulting in hybrid relaying strategies. We show that an iterative generalized directional waterfilling algorithm solves the offline throughput maximization problem, with the achievable sum-rate from an individual or hybrid relaying scheme. In addition to the optimum offline policy, we obtain the optimum online policy via dynamic programming. We provide numerical results for each relaying scheme to support the analytic findings, pointing out to the advantage of adapting the instantaneous relaying strategy to the available harvested energy.Index Terms-Energy harvesting nodes, two-way relay channel, decode/compute/compress/amplify-and-forward, hybrid relaying strategies, throughput maximization.

No abstract

No abstract

Abstract-This work considers an energy harvesting transmitter that gathers a continuous flow of energy from intermittent sources, thus relaxing the modeling assumption of discrete amounts of harvested energy present in all previous work on energy harvesting communications. Tools from convex analysis are utilized to describe the optimal transmission policy as the boundary of a properly defined region based on the energy profile. The results are extended to include models where the transmitter has a finite capacity battery with various imperfections, as well as those that incorporate a processing cost (circuit power) at the transmitter whenever it is in operation.

Abstract-In this paper, a two-hop channel is considered with energy harvesting transmitter nodes. In particular, the offline throughput maximization problem is solved for a constant power relay, and a relay with one energy arrival, in both cases assuming a finite buffer is available at the relay for temporarily storing data received from the source. The focus is on assessing the impact of this data buffer at the relay on optimal transmission policies. The solution is found indirectly, by first assuming that the relay has an infinite size buffer, and then proving that an optimal policy exists that does not require any data buffer at the relay, thus solving the problem regardless of the data buffer size at the relay. Numerical results demonstrate that the proposed solution performs significantly better than naïve policies, and a constant relay rate limits the average throughput as the peak energy harvest rate for the source increases.