Sub-Word Variable-Length Coding for Real-Time Data Decompression
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data compression technologies face challenges in balancing compression rate and hardware cost, requiring complex hardware architectures that struggle with real-time decoding and large memory capacities, especially in edge devices, leading to inefficient data processing and potential errors in decoding.
Innovation Solution
A data compression method that decomposes original codewords into sub-words, calculates their frequencies, and applies variable-length coding to generate compression codes, reducing hardware complexity and memory requirements while ensuring accurate decoding.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of substance
If lossless compression is performed on data with long encoded data (13 bits), then compression rate is improved, but memory capacity requirement increases to 2^13 memory locations
Solution Approach 1:
The patent segments the 13-bit encoded data into multiple smaller chunks (e.g., 5-bit and 4-bit segments). This segmentation allows the memory system to handle smaller address spaces (2^5 + 2^4 = 32 + 16 = 48 locations) instead of requiring the full 2^13 = 8192 locations, thereby reducing memory capacity requirements while maintaining lossless compression capabilities.
2Loss of substance
If complex hardware architecture is used to achieve high compression rate, then compression performance is improved, but hardware cost and device complexity increase
Solution Approach 1:
The decoding process is segmented into multiple simple stages corresponding to the segmented encoded data. Each stage uses straightforward decoding logic for its specific segment, avoiding the need for complex unified decoding hardware. This segmentation approach achieves high compression rates while keeping individual hardware components simple and manageable.
Solution Approach 2:
The patent transforms the compression problem from a single-dimension approach (processing entire 13-bit codes) to a multi-dimensional approach by segmenting the code space. This dimensional change allows the system to use multiple smaller lookup tables or decoding paths instead of one large complex decoder, reducing hardware complexity while maintaining compression effectiveness.
3Loss of substance
If existing compression technology is used, then compression is achieved, but real-time decoding cannot be performed and circuit speed is limited
Solution Approach 1:
By segmenting the encoded data into smaller chunks, the patent enables parallel processing of multiple segments simultaneously. This segmentation allows the circuit to decode different segments in parallel, significantly increasing the overall decoding speed and enabling real-time processing capabilities that were not achievable with monolithic decoding approaches.
4Loss of substance
If existing compression technology is used, then compression is achieved, but it requires extremely large memory capacities which is not suitable for edge devices
Solution Approach 1:
The patent segments the memory address space into multiple smaller segments, allowing edge devices with limited memory capacity to implement the compression system. Instead of requiring a single large 2^13 memory array, the system uses multiple smaller memory structures (e.g., 2^5 and 2^4 locations), making it adaptable to resource-constrained edge computing environments while maintaining lossless compression functionality.
Data Source
AI summary
The disclosure describes a data compression method and a data decompression device corresponding thereto. Firstly, the first frequencies of occurrence of different original codewords are calculated, wherein each original codeword has at least two sub-words and each sub-word has at least one bit. Then, each original codeword is decomposed into sub-words. Based on the first frequencies, the second frequencies of occurrence of all different sub-words are calculated. Variable-length encoding is performed on all different sub-words in a compression manner to generate different compression codes corresponding to all different sub-words. Finally, based on the corresponding relationship between the compression codes and all different sub-words, the original codewords are converted into encoded codewords. Besides, before performing variable-length encoding on the sub-words, the sub-words of the original codewords may be recoded.