Reliable rate-adaptive video transmission over cognitive cellular networks using multiple description scalable coding

The promise of high-bandwidth and low-latency wireless applications has placed unprecedented stress on the limited spectrum for future 5G networks. Thus, dynamic spectrum sharing has emerged as a promising solution to increase the capacity and the coverage of future cellular networks. In this paper, we tackle the issue of video delivery over cognitive cellular networks using a combination of the legacy licensed bands as well as the spectrum holes left unused in TV bands. To actively react to fluctuating bandwidth, the original video undergoes a multiple description scalable coding (MDSC), whereby the video sequence is fragmented into odd and even sub-streams, then the decomposed video is hierarchically encoded using H.264/SVC to produce two independent descriptions that refine each other. As long as the available licensed radio resources are not sufficient to deliver all the scalable video layers, the cognitive base station exploits the available portion of TV spectrum using a hybrid interweave and underlay approach to extend the capacity of the legacy infrastructure. Numerical simulations show that the proposed transmission scheme provides enough capacity, up to 1.62 Mbps increase, to handle the rising demands in terms of video quality in a timely manner without disrupting the primary link and without connection release. Moreover, the proposed MDSC framework offers an increment of 20% (24%, respectively) in the PSNR (MS-SSIM, respectively) of the received video.