Prediction of LOS NLOS using Machine Learning

The project “Prediction of LOS NLOS Using Machine Learning” focuses on classifying wireless communication channel conditions as Line-of-Sight (LOS) or Non-Line-of-Sight (NLOS) using machine learning algorithms. A synthetic dataset is generated with multiple features, including time-domain, frequency-domain, phase-based, derived/statistical, and mobility parameters, which influence signal propagation. This dataset is used to train and evaluate three different machine learning models: Support Vector Machine (SVM), Random Forest, and XGBoost. These models aim to predict the signal condition as either LOS or NLOS based on input data. The project implements a web-based application using the Flask framework, where users can upload data files, receive classifications, and compare model performance. The accuracy of the models is evaluated using metrics such as precision, recall, F1-score, and accuracy. This system provides an automated solution to optimize communication networks by predicting signal conditions, which helps in better planning and performance improvement. The project offers a user-friendly interface and accurate predictions, helping engineers make informed decisions about network setup.

Keywords:

Machine Learning, LOS, NLOS, Support Vector Machine, Random Forest, XGBoost, Classification, Wireless Communication, Signal Propagation, Flask