Joint source/channel coding to maximize perceptual video quality on packetnetworks

The QoS-guarantee in real-time communication for multimedia applications is significantly important. An architectural framework for multimedia network s based on substreams or flows are effectively exploited for combining sour ce and channel coding for multimedia data. In this paper, firstly, we desig ned a source coding scheme based on wavelet transform to exploit the human visual systems's spatial and temporal characteristics of video data. Second ly, we designed a classification scheme to partition video data into multip le substreams which have their own quality of service (QoS) requirements. F inally, we designed a management (reservation and scheduling) scheme for su bstreams to support better perceptual video quality such as the bound of en d-to-end jitter. We have shown that our joint source/channel coding scheme is better than two other schemes by simulation and a real video experiment. (C) 1999 Academic Press.