NBS: A Network-Bandwidth-Aware Streaming Version Switcher for Mobile Streaming Applications under Fuzzy Logic Control

Abstract

Good quality of service for a video streaming application under time-varying network bandwidth capacity is often required in many real-world scenarios, e.g., streaming in a typical high-speed vehicular environment. However, developing the enabling algorithms and techniques have been proven to be very challenging. Adaptive bitrate streaming method can detect a user's network bandwidth availability in real time and then adjustthe quality of a video stream accordingly. Although this method brings in many advantages, it suffers from additional storage and encoding costs, and challenges with maintaining quality globally. We propose NBS (a Network-Bandwidth-aware streaming version Switcher) system for video streaming applications. Compared with adaptive bitrate streaming method, our method is a lightweight oneby switching among different versions (e.g., high quality, medium quality, low quality) that a video streaming server provides. Moreover, our method works in an appropriate way that can balance both the responsiveness and the stability. On one hand, responsiveness is necessary because adaptive actions should be taken in a real-time manner to sustain the live streaming when the network bandwidth changes. On the other hand, too sensitive actions will deteriorate the stability and affect the perceptual quality. NBS is developed based on a feedback fuzzy controller to support the dynamic and adaptive switching. Comparing with the traditional control approaches, the fuzzy logic enables NBS to enjoy much better responsiveness and stability. We have conducted comprehensive experiments to evaluate our approach and the experimental results show that the proposed approach can effectively up-scaling or down-scaling the quality of the stream according to the network bandwidth availability by a smooth and nearly unnoticeable switch without disrupting the continuous playback.