Content Aware Resource Allocation for Video Service Provisioning in Wireless Networks.

Video service has been a killer application over wireless networks. Many cross-layer optimization techniques have been proposed to improve the quality of video services in wireless networks. However, most of them did not consider video content type information in resource allocation, which greatly affects the quality of users’ watching experience. In this paper, we take video type information into consideration for resource allocation at base stations. Accordingly, for given transmission power at base station, we build an optimal model to achieve maximal achievable total Mean Opinion Score (MOS) by allocating appropriate powers and video rates for different users watching different types of videos. Numerical results show that our model can achieve much higher MOS compared with existing scheme that does not consider such video type information.