MOTPS: Multi-Objective Optimization and Trajectory Prediction for Task Offloading in SDN-Based Vehicular Edge Networks

Abstract

This paper presents a novel approach for task offloading in Software Defined Networking (SDN)-based vehicular networks based on a multi-objective optimization algorithm which uses Non-dominated Sorting Genetic Algorithm II (NSGA-II) and Long Short-Term Memory (LSTM)-based vehicle trajectory prediction. The proposed solution addresses key challenges such as energy consumption, communication and computation delays, load balancing, task deadlines, and task division into sub-tasks. By leveraging SDN’s centralized control plane and multi-controller architecture, the framework efficiently manages resources in dynamic vehicular environments. Extensive simulations using real-world vehicular mobility datasets demonstrate that our SDN-enabled task offloading framework for NSGA-II based vehicular task offloading significantly improves task completion time, energy consumption, computation delay,load balancing, and overall resource management compared to existing solutions.