Adaptive Latency and Throughput Aware Multi-Access Scheduler for 5G Advanced and Beyond

Fifth Generation (5G) advanced and beyond communication systems aim to support a large volume of applications with diverse Quality of Service (QoS) requirements. With limited spectrum resources, 5G and Wireless Local Area Network (WLAN) multi-access scheduling is one of the proposed solutions that efficiently utilizes licensed and unlicensed frequency bands. While such heterogeneous systems help serve many users, they also present significant challenges in packet scheduling due to the varying serving capacities of the licensed vs unlicensed bands, measured in terms of latency and through-put. The existing scheduling algorithms typically try to reduce resource costs, latency, or congestion without any specific focus on the unique QoS requirements of different 5G applications. In this paper, we propose an Adaptive Latency and Throughput Aware Multi Access Scheduler (ALTAMAS) that makes the scheduling decisions based on application QoS requirements and time-varying channel characteristics. ALTAMAS utilizes the Goal programming-based Multi-Objective Optimization (MOOP) method to optimize the scheduling decision. Extensive simulations conducted to verify the veracity of the scheduler show that ALTAMAS can meet the QoS requirements of all the applications (measured in terms of throughput and latency) up to link utilization of 90 %.