Tabu search for dynamic spectrum allocation in cellular networks

In this paper, we present and analyse a Tabu Search algorithm for Dynamic Spectrum Allocation in cellular networks. We study a case where an operator is providing packet services to the end users. The objective of the cellular operator is to maximise its reward while taking into account the trade-off between the spectrum cost and the revenues obtained from end users. These revenues are modelled here as an increasing function of the achieved throughput. The cost is proportional to the bandwidth of the spectrum leased to the regulator or some spectrum broker. Results show that the algorithm allows the operator to increase its reward by taking advantage of the spatial and temporal heterogeneities of the traffic in the network, rather than assuming homogeneous traffic for its radio resource allocation. Our Tabu Search-based Dynamic Spectrum Allocation algorithm is efficient in terms of the required memory space and convergence speed. Results show that the algorithm is fast enough to suit a dynamic context.