Publication Date

Fall 2022

Degree Type

Master's Project

Degree Name

Master of Science (MS)


Computer Science

First Advisor

Navrati Saxena

Second Advisor

Genya Ishigaki

Third Advisor

Abhishek Roy


Machine Learning, Cell Selection, 5G, Vehicular Networks, Handover


This paper proposes a novel approach to handover optimization in fifth generation vehicular networks. A key principle in designing fifth generation vehicular network technology is continuous connectivity. This makes it important to ensure that there are no gaps in communication for mobile user equipment. Handovers can cause disruption in connectivity as the process involves switching from one base station to another. Issues in the handover process include poor load management for moving traffic resulting in low bandwidth or connectivity gaps, too many hops resulting in multiple unneccessary handovers, short dwell times and ineffective base station selection resulting in delays and other connectivity issues. Here, we propose an efficient handover model using trajectory prediction and optimized target base station identification.