A Novel Range-Free Grid Localization Algorithm Based on Support Vector Machine in Wireless Multimedia Sensor Networks

摘要

Wireless multimedia sensor networks (WMSN) play an important role in many applications. The location of sensor nodes is the key to achieving sensor data effectively. In this paper, we use multi-category support vector machine (multi-category SVM) to divide the whole network into grids to resolve a localization problem in WMSN. By multi-category SVM localization model, we can predict the location of new nodes, and the result shows that this method can improve the accuracy of localization compared with the same kind.