Improving mobile video streaming with link aware scheduling and client caches

The rapid growth in multimedia traffic is straining mobile networks necessitating the need for efficient content delivery mechanisms. In this paper we show that significant gains are possible, especially for streaming non-live, pre-recorded content (e.g. Video on Demand, Youtube clips) by opportunistically exploiting the variations in the wireless link quality due to "slow fading" and by leveraging the storage on the end-users device. The proposed mechanisms and the resulting gains are complementary to the existing multi-user diversity scheduling mechanisms for fast time-varying channels (e.g. proportional fair scheduling) that are deployed at the Base Stations. Our contributions include analytic quantifications of the gains from slowly time-varying shadow fading channels as well as the design of a scheme to exploit these gains for better network efficiency and higher content capacity. Our scheme works by dynamically and selectively adapting network resource allocation to contending sessions (e.g. in a cell) so that more content is sent to sessions at times when they have better link quality. Any excess content is buffered in the large device storage thus ensuring uninterrupted playback at times of deep fades. Moreover sufficient rate guarantees are provided even over short time intervals to keep the device buffers from underflowing. Our scheme can be easily implemented at the application layer using end to end monitoring. Our simulations show that the proposed scheme results in substantial gains (60 􀀀 100% more users) for mobile video content delivery.