|
This thesis presents some main problems found in audio transmission over Internet, and proposes an approach to overcome side effects of packet loss and transmission delay. The new approach is realized by adopting the concept of robust packets. Comparing with the traditional approach, in which a packet contains a single audio frame, a robust packet contains some audio frames. That is , the □th, robust packet can contain the □th, □ + 1th,…, □ + □th audio frames. If the □th packet is lost, the □th frame contained in the □ - 1th or more previous packets can be used. In this way, more robust audio transmission can be achieved in Internet. We presnt different types of robust packets, the corresponding transmission models, performance, and applications in the thesis.
|