Relay subset selection in cognitive networks with imperfect CSI and individual power constraints

This paper considers the relay subset selection problem in an underlay cognitive network in which two secondary users communicate assisted by a set of N potential relays. More specifically, this paper deals with the joint problem of choosing the best subset of L secondary relays and their corresponding weights which maximize the Signal-to-Interference-plus-Noise ratio (SINR) at the secondary user receiver, subject to per-relay power constraints and interference power constraints at the primary user. This problem is a combinatorial problem with a high computational burden. Nevertheless, we propose a sub-optimal technique, based on a convex relaxation of the problem, which achieves a near-optimal performance with a reduced complexity. Contrary to other approaches in the literature, the secondary relays are not limited to cooperate at full power.