4G-LTE-Network
Project 1 - 4G LTE Network IP traffic classification using machine learning
Real time internet traffic dataset has been developed by capturing information in the form of audio, video, text data using packet capturing tool and pre-processed for removing redundant features and modifying the dataset in order to be trained and tested. Five ML algorithms SVM, KNN, Random Forest, Decision Tree and Naïve Bayes are employed for IP traffic classification for dataset. The experimental analysis shows that Random Forest is an effective ML technique for near real time and online IP traffic classification with reduction in packet capture duration and reduction in number of features characterizing each application sample with Correlation based Feature Selection Algorithm.
Project 2 - 4G LTE Network IP traffic classification using machine learning
In this work, we used various Machine learning models i.e., Naive Bayes, KNN, SVM, Decision Tree, Random Forest to classify packets into their respective user activities. The dataset was created by capturing packets on a 4G connection using Wireshark. Then the captured packets were divided into various flows and 65 flow features were extracted for training the model. Random forest gave the best accuracy of around 87%.
Project 3 - 4G LTE Network IP traffic prediction using machine learning
The purpose of this work is to process 4G IP data, find basic patterns and build several training models that predict traffic on the LTE network.