Adaptive Video Streaming Over Cognitive Radio Networks

Abstract

Several challenges face reliable video streaming over wireless networks due to the stringent requirements of high data rate, low error rate, and limited end-to-end delay. Cognitive radio (CR) networks offer a great advantage to unlicensed users (typically called secondary users) by allowing them to exploit the unused spectrum of licensed users (known as primary users) on an opportunistic basis. However, it is more challenging to deliver video services over dynamic CR channels that are available to secondary users not only intermittently but with all the challenges of wireless channels. In this research, several frameworks are proposed to stream different scalable videos from a base station to multiple secondary users over a CR network. The objective of this study is to ensure that end users will enjoy continuous video playback with acceptable perceptual quality. To achieve such a goal, a channel allocation algorithm is introduced to adaptively assign the available radio channels to secondary users while taking into considerations the quality of their assigned channels as well as their buffer occupancies. In addition, different streaming algorithms are devised to ensure the delivery of scalable video frames, with base and enhancement layers, within the delay constraints with priority given to the base-layer frames to guarantee the continuity of video playback. Moreover, adaptive modulation is used based on the importance of transmitted video information and the channel state information (CSI) as fed-back by secondary users. Extensive simulations are performed using SimEvents simulator in MATLAB, the results of which show that the proposed schemes that integrate the devised channel allocation and streaming algorithms with adaptive modulation and scalable source coding techniques react to the variations in the channel conditions and the dynamics of the secondary users’ playback buffers in an acceptable fashion. This in turn resulted in efficient usage of the available CR resources as demonstrated in the achieved peak signal-to-noise ratio (PSNR) of the reconstructed video streams with no interruptions in the playback process. It has also been shown that scalable videos outperform their single-layer counterparts in terms of the achieved video quality. Finally, it is shown that joint adaptive modulation and channel coding results in improved bandwidth utilization, continuous playback and enhanced perceptual video quality at the secondary users end.