DTPRA: Dynamic Traffic and Priority-Aware Resource Allocation with Admission Control in Highly Dense Heterogeneous Wireless Networks

Abstract

This study addresses the pressing challenges of next-generation Heterogeneous Wireless Networks (HWNs), specifically the need for higher spectral efficiency, reduced latency, increased throughput, and faster data transmission. With the growth of wireless technology, HWNs are becoming increasingly dense, and achieving seamless connectivity, optimal resource allocation, and ideal admission control mechanisms is crucial for enhancing HWN performance and expanding overall coverage. The complex interference issues associated with urban mobility make resource allocation and admission control extremely challenging, thereby increasing network complexity. Recently, soft-computing techniques such as Machine Learning (ML) and Deep Reinforcement Learning (DRL) have been applied to resource allocation and admission control to improve throughput and enhance spectral efficiency, thereby meeting users' application demands. This paper presents a Dynamic Traffic Priority-aware Resource Allocation (DTPRA) strategy. DTPRA introduces a throughput-gain model that incorporates an interference-optimization model by integrating backoff-time optimization into the resource allocation algorithm. This model then optimizes resource allocation for call admission according to user priority requirements using the Optimized DRL (ODRL) model. The DTPRA-ODRL approach is effective in reducing resource access failure and delay, with higher throughput and delivery ratios, thereby improving resource access efficiency compared to current Efficient Resource Allocation Admission Control based on DRL (ERAAC-DRL) methods.