Efficient Resource Allocation for Ultra Reliable Low Latency Communication Delay Minimization in Fifth Generation Networks

Abstract

In Modern world, Fifth Generation (5G) technology is ubiquitous, so it is necessary to meet all of its service requirements. Hence, resource allocation for every service is very important. This research addresses the problem of resource allocation for both Enhanced Mobile Broadband (eMBB) and Ultra Reliable Low Latency Communication (URLLC) users. The work comprises both static and dynamic resource allocation for eMBB and URLLC users. This research aims to minimize latency for URLLC users by taking into account the time and energy constraints of both eMBB and URLLC in static resource allocation, formulating these constraints as a convex optimisation problem. The results show that the static resource allocation strategy performs better than the fixed bandwidth and CPU cycle schemes. However, the use of static resource allocation becomes inefficient as the number of users increases. We propose a dynamic resource allocation strategy to address this issue, which uses online conformal prediction to schedule URLLC traffic on top of eMBB. The dynamic resource allocation strategy outperforms all previous resource allocation methods, ensuring 67% eMBB efficiency and 1 millisecond latency for URLLC users.