Sensitivity Analysis of Static Sensor Deployment and Reconfiguration Algorithms

This study examines the effects of location inaccuracies on two static sensor deployment and reconfiguration algorithms, Coverage-Preserving Node Scheduling Scheme (CPNSS) and Optimal Geographical Density Control (OGDC) algorithms, developed by Tian et al. 2002 and Zhang and Hou (2003) respectively. For the purposes of examining the extent to which the deployment and reconfiguration algorithms are capable of reducing coverage holes and node redundancy, a simulator environment was set up, using a custom-designed simulation tool. By integrating the environment with that of a GIS application, real-world distance and scaling can be applied, allowing the assessment of the algorithms to be performed in a virtual world mimicking that of a real-world deployment. The simulation results suggest the OGDC algorithm has the potential of reducing the number of active nodes in the network whilst maintaining the same degree of coverage to a better extent than that of the CPNSS algorithm. However, this advantage depended on the starting node for the algorithm and could not be guaranteed if an arbitrary starting node was selected. OGDC was also found to depend more critically on the accuracy of location data. A single node deactivating in the network unintentionally results in a loss of coverage and increase in the overall coverage hole percentage. The CPNSS algorithm was found to be immune to location inaccuracies within the limits of the various configurations tested. From this study it was observed that the CPNSS algorithm is superior to the OGDC.