Parallel Entropy Coding Bitstream Layout for Reduced Decoding Delay
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current data compression techniques, such as those used in video encoding, face challenges with parallelism due to inherent serial nature of entropy coding processes, leading to delays that are unacceptable in real-time applications like video conferencing.
Innovation Solution
The method involves parsing input sequences into subsequences associated with predefined probability estimates, parallel entropy coding these subsequences, and interleaving them to form a bitstream with phrases of known lengths, allowing for partial codeword completion and splitting across phrases, enabling reduced delay in outputting the encoded data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If serial entropy coding is used to ensure accurate decoding, then decoding accuracy is maintained, but processing speed deteriorates
Solution Approach 1:
The patent divides the input sequence into multiple subsequences, each associated with a specific probability estimate. These subsequences are encoded and decoded independently in parallel, allowing multiple decoding operations to proceed simultaneously without interfering with each other, thus maintaining accuracy while improving speed
Solution Approach 2:
The patent introduces a new dimension of parallelism by organizing codewords into phrases with known lengths and interleaving them. This structural transformation allows the decoder to process multiple phrases concurrently, converting a fundamentally serial process into a parallel one without sacrificing decoding accuracy
2Productivity
If parallel entropy coding is implemented to improve processing speed, then throughput is increased, but output delay increases
Solution Approach 1:
The patent performs preliminary actions by determining the lengths of phrases in advance during encoding and embedding this information in the bitstream. This allows the decoder to know how many bits to read for each phrase before decoding begins, eliminating the need to wait for complete sequences and reducing output delay
Solution Approach 2:
By segmenting the encoded data into phrases with predetermined lengths and interleaving them, the patent enables the decoder to process multiple segments simultaneously. This segmentation allows partial codeword completion across different phrases, reducing the time to produce output while maintaining high throughput
3Reliability
If entire sequences are encoded before output to ensure completeness, then data integrity is maintained, but processing delay increases
Solution Approach 1:
The patent performs preliminary encoding of subsequences into phrases with known lengths before final output. The interleaving structure and length information allow the decoder to reconstruct complete data incrementally, maintaining integrity without requiring complete sequence encoding beforehand
Solution Approach 2:
The patent introduces phrases as intermediary structures between the input sequence and final output. These phrases act as self-contained units that can be processed and transmitted independently, allowing incremental output while ensuring data integrity through the structured interleaving and length information
Data Source
AI summary
Methods and devices for encoding and decoding that involve sorting bins according to their respective estimated probabilities to form subsequences, each subsequence having an associated estimated probability. Subsequences are encoded to form codewords. Ordered sets of phrases of known length are then formed from the codewords. Each first of the phrases in a set contains at least part of one codeword. The first phrase has an associated estimated probability and the probability estimates associated with each of the other phrases in the set are determined based upon the probability estimate associated with the first phrase, which permits the phrases to be decoded in parallel.