Machine Learning-Driven Wireless Optimization: Improving Network Performance and ISP Selection
摘要
Wireless communication remains crucial as the deployment of 5G networks expands and 4G continues to be essential. The integration of emerging technologies into corporate processes has intensified competition and created new opportunities for service providers. Businesses are shifting focus from acquiring to retaining customers due to the growing dependence on mobile devices across individual, corporate, and governmental sectors. Ensuring proper resource allocation, network design, and Quality of Service (QoS) is vital for implementing advanced wireless technologies globally. Among QoS metrics, network throughput is critical for user satisfaction. However, validating the authenticity and reliability of throughput measurements at vendor endpoints is equally important. Despite contributions from multiple telecommunications providers, QoS remains inconsistent in some regions. This study evaluates the performance of 2G, 3G, 4G, and 5G technologies, identifying which providers deliver the best service in specific areas. The research uses machine learning to predict the most suitable Internet Service Provider (ISP) based on parameters like download/upload speeds and latency. This research is significant for analyzing and identifying the best network provider that supports optimal download and upload speeds and latency in specific areas. Furthermore, it can help network providers analyze and target required regions based on user requirements and use cases, ultimately delivering the best network experience for everyone.