PROJECT TITLE :

A Dynamic Approach to Sensor Network Deployment for Mobile-Target Detection in Unstructured, Expanding Search Areas

ABSTRACT:

This paper proposes a completely unique strategy for the deployment of a static-sensor network based mostly on the employment of a target-motion probability model. The main target is on the $64000-time dynamic and optimal deployment of the network for detecting untrackable targets. The dynamic nature of the deployment refers to the on-line reconfigurability of the network as real-time info regarding the target becomes obtainable. The optimal locations of the network nodes, in flip, are determined primarily based on maximizing the chance of finding the target through the use of iso-cumulative-likelihood curves. The proposed strategy is adaptable to unstructured environments with natural terrain variation and also the presence of obstacles. Intensive simulations, a number of which are included in this paper, verified the advantage of our deployment strategy over alternative existing ways. Specifically, the proposed strategy will tangibly increase the success rate of target detection, while reducing the mean detection time, in comparison with uniform-coverage-primarily based approaches that do not contemplate probabilistic target-motion modeling. A comprehensive example is additionally included, herein, to illustrate the successful application of our proposed deployment strategy to a wilderness search and rescue scenario, where both static and mobile sensors are used inside a hybrid sensor-deployment strategy.