Comparative study of learning-based localization algorithms for Wireless Sensor Networks: Support Vector regression, Neural Network and Naïve Bayes

In recent years, there has been a growing interest in localization for wireless sensor networks. Since the complex behavior of such network, various machine learning-based methods are proposed in order to improve localization goals. The objective of this paper is to compare three well known learning-based localization techniques using Received Signal Strength Indicator (RSSI): the Support Vector regression, Naïve Bayes and Artificial Neural Network. We take into consideration two performance keys: the localization error and the computation complexity.