Distributed Calibration of Wireless Multimedia Sensor Networks

Abstract

Wireless Multimedia Sensor Networks (WMSNs) are gaining popularity among researchers over the past few years. Knowledge of the geographic locations of the sensor nodes is very important in such sensor networks. Location calibration is a method that uses the connectivity information, the estimated distance information among the sensor nodes, as well as the vision images to find the location of the sensor nodes in WMSNs. We generate local maps for nodes in immediate vicinity and merge them together to get a global map. Local maps are prone to be incorrect mainly due to distribution symmetry in a grid based network and uncertainties in the classical multidimensional scaling. Performance measures to calculate and study the same have been developed and described in detail. In this research we propose a new algorithm which corrects the error to improve the location calibration in WMSN. The major contribution of our thesis lies in developing the theoretical framework for distributed camera calibration based on vision data, local inter-node distances and local topology. Moreover our work does not need any moving targets to perform distributed camera calibration. We further develop innovative algorithms for local map generation and map merging. We evaluated the proposed approach through computer simulations. This location calibration algorithm can thus be used to develop self-organized wireless multimedia sensor networks. We demonstrate the effectiveness of our proposed approach through computer simulation. We demonstrate the proposed approach for different base node and successfully build the global map. The percentage of accuracy obtained demonstrate that the 80% of the nodes can have a good local map, however the global map obtained can accurately localize all the nodes in the network. This brings a completion to the scope of this thesis, the framework developed further provides an opportunity to extend this algorithm to real time wireless multimedia sensor networks.