Sensor Node Activation Policies using Partial or No Information

A rechargeable sensor network with redundant node deployment is considered. By redundancy we mean that at any point in time the system utility need not be a linear function of the number of nodes actually taking part in sensing/transmission job. Assuming that any given node can be in either of the three states a) active, b) passive and c) ready with some given dynamics of transitions between these states, we address the problem of optimally activating some of the existing ready nodes. Our results show optimality of threshold based activation policy for a broad class of utility function and state dynamics. We also provide a closed form expression for the threshold. A sub-optimal static parameter setting scheme is also proposed in which each node decides on activating itself randomly over time.