Paper Title
Improved Machine Learning-Based Indoor Localization for Wireless Sensor Networks

Abstract
Indoor localization of sensor nodes is very important for implementing many wireless sensor networks (WSN) and Internet of Things (IoT) applications. Range free localization is a cost-effective technique, which makes it an appropriate solution for low cost indoor WSN. However, it suffers from lack of precision. In this paper, we introduce a machine learning based localization solution by combining the Artificial Neural Network (ANN) with the well known DV-Hop localization algorithm in order to improve its performances. Two methods are proposed: ANN-based, then combining ANN with Linear Least Squares (LLS). Through a simulation study, the performance of our two proposed methods are compared with the original DV- Hop algorithm, the experimental results show that our proposed localization methods always offer a better precision than DV-Hop and the ANN + LLS method outperforms the ANN in terms of localization accuracy. Keywords - WSN, Indoor Localization, DV-Hop, Machine Learning.