Real-time content-based video communications over heterogeneous networks.

摘要

The issues on real-time content-based video communications over heterogeneous networks are addressed.; First, a real-time low-complexity content-based adaptive error-resilient approach is presented for MPEG-2 video streams, encapsulated using real-time transport protocol (RTP) and delivered over heterogeneous networks. An algorithm, using spatial and temporal properties of MPEG-2 video, is derived without decoding the video for assigning weights to each packet based on the estimated perceptual error. These weights, which indicate the relative importance of RTP packets, together with the communication channel characteristics are used to how resources are allocated for providing improved error-resilience and for classifying data packets into various classes of service in order to enhance the quality of transmission. Parameters extracted from the RTP header are used to determine the weights, so that the proposed algorithm can be implemented in real-time and at any point of the network.; Second, a real-time adaptive rate control algorithm for video transmission is proposed. The algorithm aims to determine the optimal rate at which the video should be sent through the network in order to minimize the expected distortion while meeting the delay requirements of the streaming content. The packet loss rate is assumed to be a function of the transmission rate. The nodes of the network over which video packets are transmitted determine this function, possibly based on the congestion level of the network. Since this function is unknown an augmented state feedback regulator is introduced to control the packet loss rate by adjusting the transmission rate of the video. The expected distortion is calculated by dynamic programming at the source based on the desired packet-loss and transmission rate pair and then the optimal transmission rate is chosen among feasible transmission rates. The algorithm produces an adaptive adjustment of the video transmission rate or causes some part of the video to be dropped. The expected distortion, and thus the transmission rate of the video, depends on both the video content and the network policies.; Simulation results are presented for both algorithms to establish the significant improvement in performance.