Path Loss Prediction in Smart Campus Environment: Machine Learning-based Approaches
Harsh Verdhan Singh, Shivam Gupta, Charchit Dhawan, Amrita Mishra
Abstract
This paper presents a novel application of various machine learning (ML)-based approaches towards prediction of path loss (PL) parameter for a smart campus environment. Measured data from [1] are used to train and evaluate the performance of popular ML techniques such as artificial neural network (ANN) and random forest (RF). Simulation results are presented to verify the PL prediction accuracy of the ML-based schemes. Further, a detailed comparison with the widely used empirical COST-231 Hata model demonstrates the superiority over conventional techniques thereby validating the suitability of employing ML for path loss prediction in challenging 5G wireless scenarios.
Topics & Concepts
Computer sciencePath lossArtificial neural networkRandom forestMachine learningArtificial intelligencePath (computing)WirelessPredictive modellingWireless networkData miningComputer networkTelecommunicationsMillimeter-Wave Propagation and ModelingPower Line Communications and NoiseAdvanced MIMO Systems Optimization