An Efficient Resource Management Optimization Scheme for Internet of Vehicles in Edge Computing Environment

Area of Science:

• Computer Science
• Network Engineering
• Artificial Intelligence

Background:

• Vehicle networks face resource limitations and high user demand, leading to system delays and energy consumption.
• Edge computing offers a promising environment for optimizing Internet of Vehicles (IoV) resource management.
• Efficiently managing communication and computing resources is crucial for IoV performance.

Purpose of the Study:

• To propose an efficient resource management optimization scheme for the Internet of Vehicles in an edge computing environment.
• To comprehensively consider system delay and energy consumption under computing resource constraints.
• To minimize total system overhead in dynamic and time-varying vehicle networks.

Main Methods:

• Formulation of communication and computing costs in resource optimization.
• Definition of an optimization objective considering resource constraints, system delay, and energy consumption.
• Application of a distributed reinforcement learning algorithm to optimize resource management for IoV.

Main Results:

• The proposed scheme effectively reduces total system overhead in the Internet of Vehicles.
• Experimental results show a system cost of 3.502 with the proposed algorithm, compared to 4.732 and 4.251 for other methods at 40 MHz bandwidth.
• The distributed reinforcement learning approach proves effective in optimizing IoV resource management.

Conclusions:

• The proposed resource management optimization scheme enhances the efficiency of the Internet of Vehicles.
• Edge computing combined with distributed reinforcement learning provides a robust solution for IoV resource challenges.
• This approach significantly reduces overall system costs and improves network performance.