Experimental Study on Deployment of Mobile Edge Computing to Improve Wireless Video Streaming Quality

Abstract

Due to the rapid increase in wireless network traffic, especially video traffic, innovative network architectures and algorithms need be developed to reduce congestion and improve the quality of service (QoS). Multi-access edge computing or mobile edge computing (MEC) is a new paradigm that integrates computing and storage capabilities at the edge of the wireless network. In this paper, we design and implement a wireless access network-aware video streaming system based on the MEC concept, called Edge-Controlled Adaptive Streaming (ECAS). ECAS employs in-network video bitrate adaptation to improve data delivery efficiency. In our design, a MEC function intercepts HTTP requests from the client, and a rate adaptation mechanism is employed to decide the best video representation for the client. Updated HTTP requests, with modified video rates based on the adaptation decision, are forwarded to the video server. We also design a resource allocation and streaming bitrate adaptation algorithm to achieve the overall optimization of multiple video streams, with fairness, subject to wireless transmission capacity constraint. This network-assisted adaptive streaming approach allows more accurate estimation of wireless network states and can enhance the QoS of multiple video streams. A prototype is implemented to prove the concept, and the experimental results demonstrate that the proposed scheme significantly improves video streaming performance compared to the de facto standard video streaming technique, Dynamic Adaptive Streaming over HTTP (DASH).