Cognitive MIMO Relay Networks With Generalized Selection Combining

We propose transmit antenna selection with receive generalized selection combining in dual-hop cognitive decode-and-forward relay networks with spectrum sharing for reliability enhancement and interference relaxation. In this network, a single antenna, which maximizes the receive signal-to-noise ratio (SNR) is selected at the secondary transmitter, and a subset of receive antennas with the highest SNRs is combined at the secondary receiver. To demonstrate the advantages of our proposed framework, we derive new exact closed-form expressions for the outage probability and the symbol error rate of the secondary network in Rayleigh fading. We also derive easy-to-evaluate asymptotic expressions in the high-SNR regime to gain practical insights. Several important design insights are reached. Under the proportional interference power constraint, the full diversity gain is achieved and is entirely determined by the total number of antennas available in the secondary network. This result is independent of the number of receive antennas combined and the number of primary users. The positive impact of the number of receive antennas combined and the negative impact of the number of primary users on the secondary network are showcased in the SNR gain. Under the fixed interference power constraint, error floors are displayed, and the diversity gain is lost.