Quantcast

Cache-Aware QoE-Traffic Optimization in Mobile Edge Assisted Adaptive Video Streaming

Research paper by Abbas Mehrabi, Matti Siekkinen, Antti Ylä-Jääski

Indexed on: 19 May '18Published on: 19 May '18Published in: arXiv - Computer Science - Networking and Internet Architecture



Abstract

Multi-access edge computing (MEC) enables placing video content at the edge of the network aiming to improve the quality of experience (QoE) of the mobile clients. Video content caching at edge servers also reduces traffic in the backhaul of the mobile network, hence reducing operational costs for mobile network operators (MNOs). However, minimizing the rate of cache misses and maximizing the average video quality may sometimes be at odds with each other, particularly when the cache size is constrained. Our objective in this article is two fold: First, we explore the impact of fixed video content caching on the optimal QoE of mobile clients in a setup where servers at mobile network edge handle bitrate selection. Second, we want to investigate the effect of cache replacement on QoE-traffic trade-off. An integer nonlinear programming (INLP) optimization model is formulated for the problem of jointly maximizing the QoE, the fairness as well as minimizing overall data traffic on the origin video server. Due to its NP-Hardness, we then present a low complexity greedy-based algorithm with minimum need for parameter tuning which can be easily deployed. We show through simulations that the joint optimization indeed enables striking a desired trade-off between traffic reduction and QoE. The results also reveal that with fixed cached contents, the impact of caching on the QoE is proportional to the desired operational point of MNO. Furthermore, the effect of cache replacement on QoE is less noticeable compared to its effect on backhaul traffic when cache size is constrained.