![]() | Only 14 pages are availabe for public view |
Abstract Mobile networks face numerous challenges due to the rapid increase in data traffic and the need for high-quality service delivery. Good capacity and coverage for mobile networks are the main challenges, the integration of machine learning (ML) and artificial intelligence (AI) has emerged as a promising approach to address these challenges effectively. This thesis aims to analyze, visualize and study the nature of collected data that is in the form of Key Performance Indicators (KPIs) for cellular mobile networks in a specific region in Egypt, using Machine Learning (ML) algorithms that indicate service quality and accomplish resource utilization, and attempt to improve network performance and Quality of Services (QoS) satisfaction using offloading and traffic steering techniques. Four frameworks are proposed in this thesis with the help of ML models. The first framework is proposed to analyze 4G/LTE KPIs and identify the most affected areas while the second framework is designed to apply different traffic prediction models and choose the most suitable for this area. The third one is considered a third smarter integrated framework for best traffic prediction modeling technique-based Data structure using ML algorithms. Finally, adaptive multi-band optimal frequency carrier selection algorithm is proposed to choose the near optimal band for offloading using ML algorithms. The proposed algorithm can predict several users in the specific cell during a specific time. In this work, an enhanced proposed pyLTE (E-pyLTE) simulator is used to simulate the desired BSs and calculate the desired resources required to move. Furthermore, it helps in planning new coordinates for new BSs if there are insufficient resources in neighboring cells. |