Fusion-Aware Routing in Wireless Sensor Networks

Abstract

In Wireless Sensor Networks, nodes collaborate to perform a single task or a set of related tasks. Data generated by sensors is very often correlated. Data/decision fusion has been proposed to avoid redundant data occupying network bandwidth and consuming energy. In military setting, this suppression is required also to defeat cryptanalysis efforts by adversaries. However, in most existing routing algorithms, fusion is incidental. We argue that a coupling between the two can result in early fusion thereby suppression of redundant transmission and better energy conservation. This study proposes a CDMA-based cross-layer design framework that enables early fusion in an energy-efficient, deterministic and scalable manner with upper bound on delivery time. The algorithm also allows time-series measurement history to be maintained, as required by many fusion techniques.