Deploying long-lived and cost-effective hybrid sensor networks

In this paper, we study the problem of network deployment in hybrid sensor networks, consisting of both resource-rich and resourceimpoverished sensor devices. The resource-rich devices, called microservers, are more expensive but have significantly greater bandwidth and energy capabilities compared to the low-cost, low-powered sensors. Such hybrid sensor networks have the potential to support the higher bandwidth communications of broadband sensor networking applications, as well as the fine-grained sensing that is made possible by smaller sensor devices. We propose several techniques to investigate some fundamental questions on hybrid sensor network deployment — for a given number of microservers, what is the maximum lifetime of a sensor network and the optimal microserver placement? What benefit can additional microservers add to the network, and how financially cost-effective is it to introduce these microservers? For our investigation, we propose a cost model for energy usage in hybrid sensor networks, which is then formulated into an integer linear optimization problem and solved optimally. The integer linear problem solution does not scale with network size thus we introduce an approximation algorithm using tabu-search technique. Our studies show that network life time can be extended by more than 60% by adding an extra microserver to the network; the network life time of optimized microservers’ placement can be more than 500% longer than the worst case life time. We