Parallel Entropy Coding via Context Reordering for Faster Compression
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current entropy coding methods, such as those used in H.264/AVC, are inherently serial and computationally demanding, posing challenges for high-speed processing, especially in devices with limited processing power, and struggle with efficient compression of binary data due to their reliance on serial processing and adaptive context modeling.
Innovation Solution
A parallel processing architecture for entropy coding and decoding is introduced, where the input sequence is reordered into groups with unique contexts, allowing each bit to be processed independently and assigned to a subsequence based on its probability, enabling parallel entropy coding and decoding, and incorporating a reordering schedule to ensure that each bit has a distinct context from others, facilitating parallel context modeling.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If serial entropy coding methods are used, then compression efficiency is maintained, but processing speed decreases and computational demand increases
Solution Approach 1:
The input sequence is segmented into multiple subsequences based on context groups, allowing parallel processing of independent context groups. Each context group contains bits with unique contexts that can be processed independently, enabling segmentation of the serial coding process into parallel operations while maintaining compression efficiency.
Solution Approach 2:
The patent transitions from serial processing (one-dimensional time sequence) to parallel processing by introducing a new dimension of context groups. Bits are reorganized from a single sequence into multiple independent context groups that can be processed simultaneously, effectively adding a parallel processing dimension to the entropy coding operation.
2Speed
If parallel processing is implemented, then processing speed increases, but device complexity increases
Solution Approach 1:
The encoder is segmented into multiple context modelers, each handling a specific context group. This segmentation distributes the computational load across independent units that can operate in parallel, reducing the complexity burden on any single processing unit while maintaining overall processing speed through concurrent operations.
Solution Approach 2:
The input sequence is pre-processed to identify and separate bits into distinct context groups before parallel encoding begins. This preliminary reordering action organizes the data structure to enable parallel processing without requiring complex synchronization mechanisms during the actual encoding phase, thereby reducing device complexity.
3Loss of information
If context modeling is performed adaptively, then compression ratio improves, but processing time increases
Solution Approach 1:
Adaptive context modeling is segmented into independent context modelers that operate in parallel on different context groups. Each context modeler maintains its own probability estimates and adapts independently, allowing the system to achieve high compression ratios through accurate context modeling while reducing processing time through parallel adaptation across multiple context groups simultaneously.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method for parallel context modeling through reordering the bits of an input sequence to form groups of bits in accordance with a context model-specific reordering schedule. The reordering schedule is developed such that the groups of bits are formed to satisfy two conditions: first, that the context for each of the bits in a group of bits is different from the context of each of the other bits in that group, and the context of each of the bits in that group is determined independently from each of the other bits in that group. The parallel context modeling may be used in encoding or decoding operations.