Optimal cluster head selection schemes for hierarchical OFDMA based video sensor networks

In this paper we present a novel framework for cluster head (CH) selection and sub-carrier allocation towards intra-cluster communication in an Orthogonal Frequency Division Multiple Access (OFDMA) based Wireless Sensor Network (WSN). The proposed framework maximizes the overall performance of the WSN in terms of the throughput, video quality and network life. The OFDMA PHY optimization criterion for rate maximization is formulated as a bi-level optimization problem for CH node selection followed by optimal OFDMA sub-carrier allocation. It is then demonstrated that the above optimal resource allocation problem can be reduced to the solution of a weighted bipartite graph matching problem and is solved employing three different schemes, namely the popular Hungarian algorithm, game theory based multi-item auction and the Gale-Shapley stable matching scheme. Based on the above formulation, we also derive an optimal scheme to minimize the energy consumption in the WSN. Further, the performance of the proposed schemes is demonstrated through simulations in the context of scalable video transmission in a video sensor network. Results illustrate that these schemes significantly outperform suboptimal WSN cluster head selection schemes in terms of data rate and video quality.