Transmitting correlated sources using energy harvesting transmitters

In this paper, we consider the transmission of two correlated Gaussian sources over additive white noise channels (AWGN) using two energy harvesting nodes. Under the squired distortion measure, we minimize the weighted sum-distortion by optimizing the transmit power of each node. By formulating this problem as a standard convex optimization problem, we investigate the structure of the optimal power allocation. Our analysis demonstrates that the transmit power of each node is increasing with the water level associated with itself, which is a monotone increasing function of time, while decreasing with that associated with the other node. To solve the convex optimization problem efficiently, we develop an iterative search algorithm called the iterative generalized backward-water-filling algorithm. Our numerical experiments show that the proposed algorithm is highly efficient and can converge to the optimal solution within five iterations.