Abstract
Underlay cognitive networks operate simultaneously with primary networks satisfying stringent interference constraints, which reduces their transmission power and coverage area. To reach remote destinations, secondary sources use relaying techniques. Selecting the best relay among the available ones is a well known technique. Recently, selective cooperation is investigated in cognitive networks where the secondary nodes can adapt their transmission power to always satisfy the interference threshold. In this paper, we investigate a situation where the secondary nodes have a fixed transmission power and may violate the interference threshold. We present two relay selection schemes; the first one excludes the relays not satisfying the interference constraint and then picks up a relay from the remaining ones that can provide the maximum signal-to-noise ratio (SNR). The other scheme uses a quotient of the relay link SNR and the interference from the relay to the primary user and optimizes it to maximise the relay link SNR. We derive closed form expressions for outage probability, bit error rate, channel capacity and diversity of the system for both schemes by using tight approximations. We also study mutual effects of interference. Simulation results confirm the analytical results and reveal that the relay selection is feasible at low SNRs. Copyright (c) 2013 John Wiley & Sons, Ltd.