Raptor-Coded Scalable Video Transmission Against Packet Loss
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Scalable video coding over wireless networks is hindered by sensitivity to channel noise and error propagation due to unreliable wireless communication links, despite standard video coders offering high coding efficiency in scalable modes.
Innovation Solution
A distributed joint source-channel coding framework using Raptor codes for both video compression and packet loss protection, with a cross-layer design that performs joint video compression and protection, and an iterative soft-decision decoder for joint Raptor decoding, leveraging side information at the decoder.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If standard video coders (e.g., H.264) are used in scalable mode, then coding efficiency is improved, but sensitivity to packet loss and error propagation increases
Solution Approach 1:
The patent combines source coding and channel coding into a unified distributed joint source-channel coding framework. Instead of treating video compression and error protection as separate stages, the system integrates them using Raptor codes that simultaneously perform both functions, thereby maintaining high coding efficiency while inherently providing robustness against packet loss through the unified coding structure
Solution Approach 2:
Raptor codes serve multiple functions simultaneously: they perform distributed source coding (compression) and channel coding (error protection) in a single code structure. This multi-functionality allows the system to achieve both high coding efficiency and reliability without requiring separate coding schemes for each function
2Loss of energy
If scalable video coding is used over wireless networks, then bandwidth efficiency is improved, but sensitivity to channel noise increases due to error propagation
Solution Approach 1:
The patent exploits the statistical correlation between successive video frames, which normally causes error propagation, by designing Raptor codes that treat this correlation as a resource. The decoder uses previously decoded frames as side information to assist in decoding current frames, converting the potential harm of error propagation into a benefit through iterative soft-decision decoding that leverages temporal redundancy
3Device complexity
If separate design using Slepian-Wolf compression plus erasure protection is used, then implementation simplicity is improved, but video quality deteriorates compared to joint design
Solution Approach 1:
The patent merges Slepian-Wolf compression and erasure protection into a single Raptor code structure. The encoder generates Raptor code symbols that inherently contain both compression and error protection functionality, while the decoder uses iterative soft-decision algorithms to jointly perform both operations, achieving superior video quality compared to separate designs
4Reliability
If FGS coding plus erasure protection is used, then robustness to packet loss is improved, but coding efficiency deteriorates compared to joint Raptor design
Solution Approach 1:
Raptor codes provide universal functionality by simultaneously achieving the error robustness of FGS coding and the high efficiency of Slepian-Wolf compression. The unified code structure eliminates the need for separate FGS layering and erasure protection mechanisms, achieving both robustness and efficiency in a single integrated system
Data Source
AI summary
A video encoder system includes a base layer and an enhancement layer for encoding video data. The base layer encodes a reduced quality version of the video data to obtain base layer data. The enhancement layer encodes the video data using energy-concentrating transform operations, nested scalar quantization, and Raptor encoders. The base layer data and enhancement layer data are transmitted through a channel to a video decoder system. The decoder system decodes the base layer data to recover an estimate of the reduced quality video and decodes the enhancement layer data (using the reduced quality video as side information) to obtain blocks of coset indices. The decoder system then operates on the blocks of coset indices to generate estimates of the original video data.