Comparison of CART-based localization and SVMs-based localization in WSN

Abstract

Localization of sensor nodes is essential for wireless sensor network when it is applied to the special applications. We formed two models to estimate the location of sensor nodes, CART-based localization and SVMs-based Localization. During the training process, the received signal strength of the reference nodes is selected as the input of two models and the location information is regarded as the output of two models. During the localization process, the decision trees of CART and support vector machines are used to estimate the location of blindfolded nodes. We demonstrate the practicality and feasibility of the two models through simulations in the 100m×100m area.