Optimal and suboptimal sensing sequences in multiuser cognitive radio networks

Designing proper spectrum decision schemes for cognitive radios (CRs) improves spectral usage efficiency as well as CR network throughput. One of the key effecting factors of the CR network throughput is the spectrum sensing sequences of the secondary users (SUs). In this paper, the SUs throughput maximization through appropriately selecting the sensing sequences of the SUs, called sensing matrix (SM), is investigated. More specifically, first the average throughput of the CR network is evaluated for a given SM in the case of error-free channel sensing. Then, an optimization problem is formulated with the aim to maximize the network throughput by finding the optimal SM. In order to avoid major challenges associated with the optimal solution, a novel sub optimal solution is proposed. Despite of having less computational complexity as well as lower consumed energy, the proposed solution performs quite well compared to the optimal solution (the optimal SM). Required considerations for the extension of the proposed approach to the case of non-perfect sensing are discussed, and then a set of illustrative numerical results is presented to validate its efficiencies.