Optimal power allocation for delay-constrained multiple-channel cognitive radio systems

Power allocation for the multiple-channel cognitive systems is studied in this paper. Our objective is to maximize the throughput of the cognitive link while providing delay guarantee for the secondary user. By using the effective capacity theory, the problem is modeled as effective capacity maximization problem with average power constraint, which turns out to be a constrained convex optimization problem. Through Lagrangian dual theory, the problem is divided into a group of sub-problems, where each sub-problem aims at maximizing the effective capacity under a certain channel sensing result. Then a gradient descent method based algorithm is proposed which solves the problem effectively. Simulation results show that the proposed power allocation scheme can effectively improve the effective capacity for the cognitive link and provide specific delay guarantee in terms of delay violation probability.