Minimum Cost Hierarchical Architecture For Correlated Data Aggregation In Sensor Networks With Mobile Access

We consider an energy-efficient data aggregation scheme for sensor networks with mobile access (SENMA), a design that opens the possibility of relieving low-power nodes from energy-consuming networking tasks. Specifically, we provide a novel distributed one-hop clustering algorithm to organize the network into a minimum cost hierarchical architecture (MCHA) for collecting all sensing data from the monitored area. For better load balancing and prolonged network lifetime, we exploit the joint effect of clustering and data correlation and provide a trade-off between communication cost and residual energy of sensor nodes. Our algorithm, which employs a message-passing algorithm for solving the min-sum optimization problem, is simple to implement and gives a near-optimal solution within a small number of iterations. Energy efficiency, scalability and robustness of the algorithm are also verified through extensive simulations.