Optimal energy allocation and storage control for distributed estimation with sensor collaboration

In wireless sensor networks with energy harvesting nodes, we study the problem of energy allocation and storage control for distributed estimation with sensor collaboration, where collaboration refers to the act of sharing measurements with neighboring sensors prior to transmission to a fusion center. To jointly design energy allocation and storage control polices, we formulate a nonconvex optimization problem in which the estimation distortion is minimized subject to energy harvesting and storage constraints. We show that the resulting optimization problem contains two special types of nonconvexities: cardinality function and difference of convex functions. By exploiting the problem structure, locally optimal solutions are found via an ℓ1 relaxation and a convex-concave procedure. Numerical experiments are provided to show the effectiveness of our approach.