Reducing Unnecessary Handovers Using Logistic Regression in 5G Networks

Abstract

The evolution of wireless communication has brought great benefits to society, such as multi-connectivity, increased connection speed, low latency, and elevated throughput. However, it has also raised concerns regarding security and the procedures of network swaps, more known as handovers. This paper proposes using machine learning in 5G mobile networks, employing a Logistic Regression algorithm to predict handovers. Considering selection criteria like Reference Signal Received Power, Signal-to-Interference-Plus-Noise Ratio, Received Signal Strength Indicator, and distance, our proposal shows that it can significantly reduce the consequences of frequent handovers and provide a better quality of service to the user, together with low computational complexity. The presented results prove the viability of this proposal.