Evaluation of Dynamic Channel and Power Assignment for Cognitive Networks

In this paper, we develop a unifying optimization formulation to describe the Dynamic Channel and Power Assignment (DCPA) problem and an evaluation method for comparing DCPA algorithms. DCPA refers to the allocation of transmit power and frequency channels to links in a cognitive network so as to maximize the total number of feasible links while minimizing the aggregate transmit power. We apply our evaluation method to five representative DPCA algorithms proposed in the literature. This comparison illustrates the tradeoffs between control modes (centralized versus distributed) and channel/power assignment techniques. We estimate the complexity of each algorithm. Through simulations, we evaluate the effectiveness of the algorithms in achieving feasible link allocations in the network, and their power efficiency. Our results indicate that, when few channels are available, the effectiveness of all algorithms is comparable and thus the one with smallest complexity should be selected. The Least Interfering Channel and Iterative Power Assignment algorithm does not require cross-link gain information, has the overall lowest run time, and achieves the highest feasibility ratio of all the distributed algorithms; however, this comes at a cost of higher average power per link.