Maximizing lifetime and handling reliability in wireless sensor networks

In this article, we face the problem of ensuring reliability of a wireless sensor network which is monitoring a given set of points of interest while maximizing its lifetime (i.e., the amount of time over which the monitoring activity can be performed). The two objectives are contrasting. Indeed, the traditional approach to achieve reliability involves providing redundant coverage, which, however, drastically reduces the network lifetime. We propose an alternative strategy where sensors adapt their sensing radii in response to failures to restore feasibility only when needed. We provide Column Generation exact algorithms for both the traditional approach and our variant, as well as a heuristic procedure for the coverage restoration phase. The advantages of our approach are shown by means of computational tests on a set of instances and failure simulations