Assessing Performance Tradeoffs in Undersea Distributed Sensor Networks

We consider the problem of determining the correct set of sensors to employ in the design of large area undersea surveillance sensor networks. As sensor technologies evolve, such networks are becoming increasingly practical. In turn, optimal selection of the number and type of sensors to deploy becomes an increasingly nontrivial process. Choices of field level detection and false alarm performance, as well as cost, all enter into this tradeoff decision space. In particular, the multiobjective nature of the problem leads to families of "optimal" solutions that each correspond to different tradeoffs between these often conflicting objectives. In this paper, we address these tradeoffs using a simple model of multi-sensor search performance and show the tradeoffs as Pareto efficient sets of solutions that satisfy system constraints. We also provide a means to determine the specific characteristics of the systems that lead to different design choices and explain how these designs perform as comprehensive search systems