Performance Evaluation of a Path Diversity-Based Video Streaming Approach over Best Effort Networks

Efficient delivery of video sequences over best effort networks is challenged by delay variations and packet losses which occur due to congestion at intermediate routers in addition to the stringent quality of service (QoS) requirements of video applications. In this paper, we propose a joint path diversity and priority-based queuing scheme for video streaming over best effort networks like the Internet. The proposed scheme exploits the differential sensitivity inherent in video frames as well as possible paths´ redundancies available in best effort networks. The performance of the proposed scheme is compared through simulations with a best effort streaming model as well as a priority-based model. The simulation results indicate that the proposed joint path diversity-based model outperforms the priority-based model in terms of the average end-to-end delay and losses which is reflected in the occupancy of the playback buffer.