Information Volume Evidential Markov Decision Process Based Hierarchical Meta Reinforcement Learning for Resource Allocation in Vehicular Networks
Year: 2024
Type: article
Abstract: Resource allocation in vehicular networks defines the strategic distribution of accessible communication resources like time slots, power, and bandwidth between infrastructures and vehicles to ensure the effective transmission of data.It handles a dynamic network condition and balances the requirements of different users while reducing interference and increasing network effectiveness.However, resource allocation faces challenges in effectively handling limited communication resources due to high energy consumption and network delays caused by frequent changes in network topology and varying user demands by the high mobility of vehicles.This research proposes the Information volume Evidential Markov Decision Process-based Hierarchical Meta Reinforcement Learning (IEMDP-HMRL) for resource allocation in vehicular networks.In traditional RL, the IEMDP-HM is incorporated to enhance resource allocation by enabling adaptive decision-making in dynamic environments.HMRL enables rapid adaptation to new tasks by leveraging learned policies from prior experiences which increases both efficiency and flexibility in resource management.In scenario 3, the IEMDP-HMRL achieves a lesser average task delay of 0.34 ms for the number of vehicles 10 compared to existing methods like Intelligent Distributed Resource Allocation and Task Scheduling (IRATS).
Language: English
Subfield: Electrical and Electronic Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: bronze
OpenAlex ID: W4405790853