Optimal active sensing in heterogeneous cognitive radio networks

In this paper, we consider a wideband cognitive radio network with limited available frame energy and treat a fundamental energy allocation problem: how available energy should be optimally allocated for sensing, probing, and data transmission to maximize the achievable average opportunistic spectrum access (OSA) throughput. By casting this problem into the multi-armed bandit framework under probably approximately correct learning, we put forth a proactive strategy for determining the optimal sensing cardinality and probing cardinality that maximize the average throughput of the secondary user. Numerical results show that our framework gives the the optimal diversity-energy tradeoff for the average OSA throughput.