Adaptive video streaming over HTTP using stochastic bitrate prediction in 4G wireless networks

Video streaming over Hypertext Transfer Protocol (HTTP) used in multifarious applications creating a multimedia environment faces a challenge in 4G wireless network due to the fluctuating nature of internet traffic and variable capacity ofwireless channel over time. The existing Dynamic Adaptive Streaming over HTTP (DASH), though works well for stored video up to some extent, poses a complication in live transmission thereby depreciating the streaming quality due to high link bit rate fluctuation. In this paper, we have proposed an efficient ARIMA Based Bit Rate Adaptation (ABBA) model in the receiver/client side that estimate the link traffic based on the incoming packet bit rate to predict the subsequent future link capacity in order to notify the sender/server. Based on the response from the receiver the server adapt its outgoing stream as per forecasted link data rate, and hence eliminate the degradation of video due to channel throughput variations. The proposed ABBA algorithm was implemented on IP over 4G wireless network and the streaming quality was evaluated on several full reference metrics of video quality. The test result outperformed an existing buffer based approach and also a fuzzy based adaptation algorithm. For example, the ABBA algorithm exhibited an average increase of 22 % in PSNR and 9% in SSIM than a buffer based method.