Enhancing QoE Through Adaptive Bitrate Allocation in Collaborative MEC-Enabled Wireless Networks

Abstract

The recent exponential growth of HTTP video Adaptive Streaming (HAS) services is primarily driven by the increasing popularity of mobile devices, advancements in mobile networks, and wide availability of diverse video content online. Efficient resource allocation in telecommunication networks is becoming increasingly crucial for mobile operators. As networks become more complex, the demand for higher bitrates and increase in traffic continues. To meet the Quality of Experience (QoE) needs of HAS users, emerging wireless networks are incorporating technologies like Multi-access Edge Computing (MEC), Software-Defined Mobile Networks (SDMN), and Cloud Radio Access Networks (C-RAN). This paper studies optimal allocation strategies for radio, storage, and computing resources across a wireless network enabled by MEC, SDMN, and C-RAN to support high-quality adaptive video streams. Video streaming quality can deteriorate when users move between network nodes or when mobile network conditions worsen. A novel MEC Collaborative Cross-Layer Bitrate Allocation (MCCBA) algorithm is introduced to enhance QoE for HAS users by enabling collaboration between MEC servers and RAN components. By addressing a mixed-integer nonlinear programming problem that considers radio resources, MEC server resources, user QoE, system throughput, and RSRP measurements, MCCBA aims to maximize user QoE, improve the system utilization, and minimize discrepancies between throughput at the MAC layer and allocated bitrates for video frames at the application layer. Compared to a baseline scheme, MCCBA improves video quality by 15.97%, minimizes the deviation of the throughput in RAN MAC layer and user's application layer by 43.6% and reduces backhaul traffic by 58.77%.