Dynamic Entropy Coding With Rank Swapping for Faster Compression
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional static entropy coding methods are inefficient when the image statistics do not match the assumed probability distribution, leading to higher compressed file sizes.
Innovation Solution
A dynamic entropy coding method that involves a two-stage mapping process, where symbols are first mapped to ranks based on their frequency, and then ranks are mapped to binary codes, using a synchronized encoder and decoder without data sharing to resolve constraint violations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If static mapping of symbols to binary codes is used, then the encoding process is simple and fast, but the compressed file size increases when image statistics do not match the assumed probability distribution
Solution Approach 1:
The patent implements dynamic entropy coding where the mapping from symbols to binary codes changes adaptively based on the actual frequency distribution of symbols in the image data. The encoder maintains a frequency table that is updated as symbols are processed, and the binary code assignment is recalculated to reflect the current symbol probabilities, thereby optimizing compression efficiency for the specific image statistics while maintaining reasonable processing speed
Solution Approach 2:
The system dynamically adjusts the probability distribution parameters used in entropy coding based on the actual symbol frequencies observed in the image. By changing the coding parameters (binary code assignments) to match the actual image statistics rather than using fixed assumptions, the method achieves better compression ratios without requiring complex computational processes
2Loss of information
If dynamic entropy coding techniques are used to change the binary code based on symbol frequency, then compression efficiency improves, but the computational intensity increases
Solution Approach 1:
The patent divides the dynamic coding process into discrete, manageable steps: counting symbol frequencies in a table, determining the current symbol's frequency, comparing with previous frequencies, and conditionally updating the binary code mapping. This segmentation of the dynamic process into simple, sequential operations reduces computational complexity while maintaining compression efficiency
Solution Approach 2:
The encoder performs adaptive updates to the binary code mapping based solely on the frequency information it already collects during normal encoding operations. The system uses its own operational data (symbol frequencies) to automatically adjust its coding strategy without requiring external intervention or complex computational resources, achieving dynamic optimization through self-service
Data Source
AI summary
A system and a method are disclosed for encoding data for transmission, including determining a rank of a first obtained symbol of the plurality of symbols, encoding, at an encoder, the rank of the first symbol, generating a new frequency entry for the first obtained symbol by incrementing an initial histogram frequency entry of the first obtained symbol, determining, based on the new frequency entry of the first obtained symbol, that the rank of the first obtained symbol of the plurality of symbols has a constraint violation with a rank of a first violating symbol in the first encoder LUT, swapping the rank of the first obtained symbol and the rank of the first violating symbol in the first encoder LUT so the constraint violation is resolved, and generating a compressed bit-stream by iteratively applying an encoding function to each symbol of the plurality of symbols.