A Probabilistic Spectrum Access Approach to Joint Sensing and Power Allocation in Multiband Cognitive Radio

In classical energy detection-based spectrum sensing, primary subbands are classified into either vacant or occupied according to the measured energy over a specific subband. Then, cognitive radio (CR) signal transmission is performed over subbands labeled as vacant. As an alternative methodology, here, we first exploit our recently-proposed probabilistic spectrum access approach, where a probability is assigned to the availability of each primary subband for opportunistic access of CR devices, by using a specific function of the measured energy over that subband. Second, based on the probabilistic spectrum access approach, we perform joint spectrum sensing and power allocation in a multiband CR scenario. The considered optimization problem is formulated with the aim of maximizing the average opportunistic CR data rate under interference and power budget constraints. The formulated optimization problem is converted into a convex problem and then, the optimal spectrum access function is derived. Finally, we provide several numerical results to confirm the superiority of the proposed spectrum sensing methodology over classical energy detection, in terms of average achievable CR data rates.