Cellular Traffic Prediction using Federated Learning
Our project proposes a novel approach to collaboratively train a quality prediction model for cellular traffic using multiple edge clients. Our approach addresses the limitations of existing machine learning techniques by dividing clients into multiple clusters, using a small augmentation dataset, and locally storing raw data. Our proposed model achieved higher accuracy with limited data and processing resources, making it an ideal solution for practical applications in cellular networks. We used a federated learning framework to ensure privacy and security of data, with each client training its own model and sharing only model updates. Our experiments show that our approach is better than existing techniques in accuracy and efficiency, and can be used for traffic management, network optimization, and resource allocation.
Keywords: Cellular networks, Machine Learning, Wireless traffic prediction, Edge computing, Collaborative learning, Federated learning, Network optimization, Traffic management
Tools: Google Collab, Anaconda, Support Vector Regression (SVR), Long Short-Term Memory (LSTM), AutoRegressive Integrated Moving Average (ARIMA), Federated Learning, Flask
Department: Department of Electrical Engineering
Project Team Members
| Name |
Email |
|
Muhammad Hanzala Iqbal
|
hanzala2019@namal.edu.pk |
|
Muhammad Usama
|
musama2019@namal.edu.pk |
Project Poster
