Distributed Dynamic Frequency Allocation In Wireless Networks Under Time-Varying User Activities

We propose a framework to analyze the performance of the distributed dynamic frequency allocation algorithm introduced in [1] in the presence of time-varying activity of clusters. In our network model, nodes are grouped into a number of clusters. Each cluster chooses its transmission frequency band based on its knowledge of the interference that it experiences. Furthermore, each cluster can be in active or inactive mode according to a two-state Markov model. We derive conditions on the convergence of the algorithm in mean and variance using stochastic analysis. Additional simulation studies verify the results of stochastic modeling and the performance of the algorithm.