Doubly compressed diffusion LMS over adaptive networks

Diffusion LMS is an efficient strategy for solving distributed optimization problems with cooperating agents. Nodes are interested in estimating the same parameter vector and exchange information with their neighbors to improve their local estimates. Successful implementation of such applications relies on a substantial amount of communication resources. In this paper, we introduce diffusion LMS strategies that offer significantly reduced communication load without compromising performance. We perform analyses in the mean and mean-square sense of these algorithms. Simulations results are provided to confirm the theoretical findings.