Quantized Ergodic Radio Resource Allocation in Cognitive Networks With Guaranteed Quality of Service for Primary Network

In this paper, for cognitive networks, ergodic resource allocation (ERA) in the downlink of an OFDMA-based secondary network is investigated, assuming the availability of quantized channel state information (CSI) between the secondary base station and secondary users. The main objective is to maximize the average sum rate of the secondary network subject to the average total transmission power constraint and the collision probability constraint of the primary users on each subcarrier. In the proposed scheme, there is no need for the instantaneous handshaking between secondary and primary networks to provide CSI between the secondary base station and primary receivers. In fact, the secondary base station only needs the knowledge of channel distribution information. Due to the probabilistic nature of the collision probability constraint in the proposed ERA problem, it cannot be solved by conventional methods such as dual decomposition. Therefore, we propose two novel suboptimal solutions called the iterative and analytical approaches. It is also demonstrated that the solutions obtained based on the proposed approaches are very close to the optimal solution. Numerical experiments are then carried out to demonstrate the performance of the proposed methods. Numerical results show that the performance of the analytical approach is inferior to that of the iterative approach, although it exhibits significantly less computational complexity.