Received signal strength prediction model for wireless underground sensor networks using machine learning algorithms

Abstract

Analysis of the Received Signal Strength (RSS) from the Wireless Underground Sensor Network (WUSN) is essential before the deployment of the sensor network. It is difficult to predict received signal strength from an underground sensor network using a theoretical model due to variations in environmental conditions. Prediction of RSS using the machine learning approach considers varying environmental conditions. In this work, we develop the RSS prediction model using machine learning concepts based on linear regression, Support Vector Machine (SVM) and random forest algorithms. The RMSE and MAE of the machine learning approach based on the SVM algorithm were found to be the least among the three techniques. It is further observed that the RSS prediction using machine learning methods produces lower RMSE than the theoretical model. Accurate estimates of RSS from a machine learning based prediction model help in deciding the node positioning in order to achieve effective communication within a WUSN.