Title :
Unequal error protection of SNR-scalable DPCM-coded video
Author_Institution :
Norwegian Univ. of Sci. & Technol., Trondheim, Norway
Abstract :
This paper presents a channel code allocation scheme developed for unequal error protection of video packets when transmitted over error-prone channels. The allocation procedure minimizes the expected distortion of one Group of Pictures, made by loss of a particular channel packet, which terminates the GOP´s decoding, and following error concealment. A source packet, generated by a quality-controlled block-based hybrid coding technique represents a rectangular region of macroblocks at a particular SNR layer in a certain video frame. The channel codes used are punctured parallel concatenated recursive convolutional codes. The work features the Viterbi Algorithm as allocation procedure. The distribution of channel codes shows that, while searching for the code allocation, which gives minimum distortion, the codec aims at the maximization of the average payload length of channel packets.
Keywords :
Viterbi decoding; block codes; channel allocation; channel coding; concatenated codes; convolutional codes; differential pulse code modulation; video codecs; video coding; SNR-scalable DPCM-coded video; Viterbi algorithm; channel code allocation scheme; channel packet payload length; codec; decoding; error concealment; error-prone channel; macroblock; picture group distortion; punctured parallel concatenated recursive convolutional code; quality-controlled block-based hybrid coding technique; source packet; unequal error protection; video frame; video packet; Codecs; Concatenated codes; Convolutional codes; Decoding; Error analysis; Error correction codes; Hybrid power systems; Payloads; Scalability; Viterbi algorithm;
Conference_Titel :
Data Compression Conference, 2004. Proceedings. DCC 2004
Print_ISBN :
0-7695-2082-0
DOI :
10.1109/DCC.2004.1281517