Sensing-throughput optimisation for multichannel cooperative spectrum sensing with imperfect reporting channels

The problem of multiband cooperative spectrum sensing with imperfect reporting channels is considered. It is assumed that the secondary users (SUs) make use of an energy detection scheme to sense a number of frequency subbands (assigned to primary users) and report their sensing results to a fusion centre through noisy channels. Exact expressions for the opportunistic throughput, false alarm and missed detection probabilities are derived in terms of the overall sensing-plus-reporting time, sensing subintervals, reporting subintervals and decision thresholds for each SU in each subband. Then, an optimisation problem is formulated that aims to maximise the opportunistic throughput of the cognitive radio network so as to maintain the interference incident on the primary network below a given level. The constraints of the optimisation problem are shown to be non-convex and the conditions under which these constraints become convex are examined. It is also shown that the objective function of the problem is non-convex and two suboptimal techniques are introduced to effectively solve the problem using the algorithms developed for solving convex optimisation problems.