Prediction of System Load in RF Signal Propagation Using Hyperparameter-Tuned Light Gradient Boosting Model

RF signal propagation refers to the transmission and behavior of radio frequency waves while propagating through different environments. Predicting system load while RF signal propagation is a very big challenge in modern communication networks since dynamic environmental factors like interference, attenuation, and propagation delays affect network performance. Accurate prediction is essential for efficient resource allocation, reduced latency, and optimized network throughput. In this work, we propose a new predictive model based on the LightGBM Regressor to solve this problem. Our model resorts to gradient boosting to iteratively learn from residuals and optimize predictions in terms of the Mean Squared Error loss function. For greater performance, we used an algorithmic grid search, optimizing the key hyperparameters: the number of estimators, learning rate, and maximum number of leaves. Experimental results show that the proposed model significantly improves the accuracy of prediction, with a substantial reduction in MSE compared to other machine learning methods which are DT, LR, K-NN, RF, AdaBoost, Gradient boosting.