Al-Based Path Selection for Optimizing Voice Call Routing Between 5G and Wired Networks

Abstract

This research looks at the use of artificial intelligence (AI) to optimize voice call routing across 5G and wired networks. Traditional routing systems often fail to adapt to changing network circumstances, resulting in poor speech quality, high latency, and wasteful resource use. To solve these issues, the paper offers an Al-based route selection model that uses machine learning methods, notably reinforcement learning, to dynamically identify optimum routing pathways based on real-time Quality of Service (QoS) measures like as latency, jitter, and packet loss.

An experimental research approach was used, with simulation tools (such as NS-3 and OMNET++) used to simulate hybrid network settings and assess the proposed Al-based routing framework. The findings showed that the Al model greatly enhanced voice call quality, decreased latency, eliminated packet loss, and outperformed standard routing systems under simulated settings. The paper ends with practical suggestions for deploying Al-based routing in real-world telecom infrastructures, emphasizing the model's scalability, flexibility, and prediction accuracy.