Genetically optimized TD3 algorithm for efficient access control in the internet of vehicles

Abstract

The Internet of Vehicles (IoV) is currently experiencing significant development, which has involved the introduction of an efficient Access Control Mechanism (ACM). Reliable access control is evolving into mandatory in order to provide security and efficient transmission within the IoV environment as the volume of vehicles equipped with connectivity continues to expand and they become more incorporated into any number of applications. The primary objective of this research is to develop an ACM for the IoV system based on the use of a Genetically Optimized Twin-Delayed Delayed Deep Deterministic Policy Gradient (TD3) algorithm. The TD3 model modifies access policies to be in line with the current scenario using deep reinforcement learning (Deep RL) techniques. This allows vehicles to make access decisions that are intelligent about the environment in which they are performing. To prevent energy loss while the vehicle is in transit into the client system, the model also emphasizes access based on the vehicle's energy consumption (EC). Finally, with the support of the genetic algorithm (GA), the accuracy of the access control model can be improved by optimizing the high-level parameters in a manner in which they improves efficiency. In order to further enhance the model's environmental sustainability and reliability, the recommended model provides an approach that is both profound and efficient for access control in the constantly changing setting of the IoV.