Element-Wise Encoder Compression for Reoccurring Data Blocks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional data compression techniques achieve only a small compression ratio due to inefficient encoding of data blocks, where most elements remain unchanged, leading to suboptimal performance.
Innovation Solution
An improved encoder and decoder that identify substantial reoccurrences in data blocks, encode unchanged elements using a symbol indicating absence of change, and encode changed elements, allowing for quantized encoding and additional compression algorithms to enhance data compression efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If conventional encoders process entire data blocks, then the encoding process is simple, but the compression ratio is small
Solution Approach 1:
The patent divides data blocks into individual elements (bytes, words, integers, characters, pixel values, audio samples) and processes each element independently. This segmentation allows the encoder to identify unchanged elements at the element level rather than processing entire blocks, significantly improving compression ratio while maintaining manageable encoding complexity through systematic element-wise comparison with reference data.
2Productivity
If delta coding is used to encode differences, then compression is achieved, but entropy of data increases
Solution Approach 1:
The patent extracts only the changed elements from data blocks and encodes them, while unchanged elements are represented by references to previous data. This extraction approach avoids creating delta values for unchanged elements, thereby maintaining lower data entropy compared to traditional delta coding that processes all elements. The method achieves compression by transmitting only necessary change information.
3Productivity
If all elements in changed data blocks are coded, then completeness is ensured, but compression efficiency is reduced
Solution Approach 1:
The patent applies different encoding strategies to different elements within data blocks: unchanged elements are encoded with references to previous data (high compression), while changed elements are encoded with their actual values (data integrity). This local quality approach ensures that only necessary data is fully coded, maintaining compression efficiency while ensuring completeness of changed elements through element-level differentiation.
Data Source
AI summary
An encoder for encoding input data to generate corresponding encoded data is provided. The encoder identifies substantial reoccurrences of data blocks and/or data packets within at least a portion of the input data. The encoder then identifies, in respect of individual elements, where elements are unchanged and/or changed within the substantially reoccurring data blocks and/or data packets. Subsequently, the encoder encodes unchanged elements in the encoded data by employing at least one corresponding symbol, or at least one corresponding bit, for example a single bit, indicating an absence of change in the unchanged elements relative to corresponding elements in a reference data block and/or data packet. Moreover, the encoder encodes changed elements in the encoded data.