Mobile Edge Computing Task Offloading Strategy Based on Parking Cooperation in the Internet of Vehicles

Area of Science:

• Computer Science
• Electrical Engineering
• Vehicular Networks

Background:

• Onboard vehicle computing is insufficient for growing demands.
• Mobile Edge Computing (MEC) offers enhanced capabilities for vehicles.
• Roadside parked vehicles present an opportunity for resource utilization.

Purpose of the Study:

• To propose a novel MEC task offloading strategy leveraging roadside parked vehicles.
• To establish resource sharing among vehicles, roadside units (RSUs), and cloud servers.
• To optimize collaborative task offloading for reduced delay and energy consumption.

Main Methods:

• A cooperative task offloading model was developed.
• The problem was framed as a multi-constraint optimization problem.
• A hybrid genetic algorithm (HHGA) with a mountain-climbing operator was employed to solve the problem.

Main Results:

• The HHGA algorithm demonstrated significant improvements in reducing delay and energy consumption compared to traditional methods.
• For 25 tasks, delay improved by 24.1% and energy consumption by 11.9%.
• System overhead was reduced by 7.9% for 1.0 MB tasks.

Conclusions:

• The proposed roadside parking cooperative MEC strategy effectively reduces task offloading costs.
• This approach enhances computational capabilities for vehicles by utilizing idle resources.
• HHGA provides an efficient solution for complex, multi-constraint offloading problems.