Combining strategies for the optimization of resource allocation in a wireless multiuser OFDM system

Wireless transmission systems are constrained by several parameters such as the available spectrum bandwidth, the mobile battery autonomy, the channel impairments, the transmission power, etc. In this paper, we investigate new strategies that aim at improving the allocation of resources in cognitive radio systems based on Orthogonal Frequency Division Multiplexing. We propose several techniques for the dynamic assignment of available subcarriers in such a way to maximize the mobile autonomy in uplink communications and minimize the required amount of bandwidth in downlink. In both transmission links, the system must guarantee a certain transmission data rate to each user. An iterative greedy approach, which assigns the users that have the weakest battery level or the smallest transmission rate the most favorable subcarriers, is introduced in order to maximize the overall system throughput. We show how an optimal solution for the combinatorial optimization problem can be determined by complex algorithms inspired by the field of statistical mechanics. Fortunately, our results show that for moderate values of the required data rates and the number of users, our greedy approach performs almost as well as the simulated annealing technique but with a much more affordable complexity.