Resource Allocation for Content Distribution Using Optimized Temporally‐Aware Adaptive Graph Convolutional Network in Fog Radio Access Networks

ABSTRACT

New multimedia applications are emerging at a rapid rate, propelling an exponential rise in mobile internet traffic and increasing the bar for next‐generation wireless networks in service quality. This indicates the advancement of wireless communication technology. Efficient allocation of diverse network resources is necessary for improving network quality of service together with end‐user experience leading to lower latency content transfer is an essential subject quickly addressed on the modern internet. In this paper, resource allocation for content distribution using optimized temporally‐aware adaptive graph convolutional network in fog radio access networks (RACD‐TAAGCN‐FRAN) is proposed. Here, the resource allocation for content distribution using temporally‐aware adaptive graph convolutional network (TAAGCN) to cooperative caching and routing. Finally, the fractional order water flow optimizer (FOWFO) algorithm is used for improving the TAAGCN, which accurately optimized resource allocation. Then, the proposed RACD‐TAAGCN‐FRAN is implemented, and the performance metrics like average network delay (AND), request arrival rate (RAR), relative cache size (RCS), network lifetime (NLT), computational time, and packet delivery ratio (PDR) are examined. The proposed RACD‐TAAGCN‐FRAN approach attains 14.14%, 18.41%, and 23.28% lower AND; 25.21%, 32.62%, and 20.22% higher RAR; and 19.91%, 31.62%, and 26.22% higher RCS when compared to the existing techniques: deep reinforcement learning depending on resource allocation of content distribution in FRAN (DRL‐RACD‐FRAN), a reinforcement learning‐dependent computing offloading with resource allocation system in FRAN (RL‐CURA‐FRAN), and deep reinforcement learning‐dependent joint optimization of computation offloading with resource allocation in FRAN (DRL‐JOCRA‐FRAN) methods, respectively.