作者: Dookyoon Han , Jinyoung Han , Youngbin Im , Myungchul Kwak , Ted Taekyoung Kwon
关键词: Computer science 、 Computer network 、 Video streaming 、 Mobile computing 、 Video quality 、 Measurement study 、 Contextual information 、 Download 、 Real-time computing 、 Entropy (information theory) 、 Wireless
摘要: Wireless/mobile video streaming has become increasingly popular, which makes wireless link bandwidth scarce. To provide services to mobile users, it is crucial adapt the condition and traffic fluctuation. We investigate factors in natural environments user contexts affect available bandwidth. this end, we conduct a measurement study contains 38 repeated trips along same 5~km circular road campus of Seoul National University April May 2013. measure download throughput from two different networks (3G 4G LTE) with varying location, time, humidity, speed. Our results reveal that humidity location are more important 3G network, while speed, ones LTE network predict then propose an adaptive framework, MASERATI, where information used demonstrate MASERATI significantly improves QoE users terms playout success rate, quality, stability, comparison DASH.