Column-Wise Entropy Coding for Real-Time Fixed-Length Sensor Data

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing data compression techniques are inefficient for fixed-length data in sensor networks, leading to suboptimal compression efficiency and increased data transfer volumes, especially when immediacy is required, as they either require significant accumulation for compression or transmit data without compression, resulting in larger data transfer sizes.

Innovation Solution

A data compression coding method that treats fixed-length bit strings from sensors as virtual tabular data and compresses them in a column direction using entropy coding schemes like Huffman and arithmetic codes, allowing for adaptive and accumulative coding methods to assign shorter codes to high-probability symbols and longer codes to low-probability symbols, thereby achieving effective compression without relying on correlations between columns.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Quantity of substance

If existing data compression techniques are used for fixed-length sensor data, then data transfer volume is reduced, but compression efficiency is suboptimal and requires significant data accumulation

Engineering Contradiction:
Improvedata transfer volumeVSAvoidcompression efficiency
Core Design Contradiction:
Quantity of substanceVSProductivity

Solution Approach 1:

The patent segments the fixed-length data records into multiple fields based on their data types and characteristics. Each field is then processed independently through appropriate coding schemes (ASCII, BCD, or binary coding), allowing compression to occur at the field level rather than requiring accumulation of entire records. This segmentation enables immediate compression of individual fields without waiting for data accumulation.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent applies different compression strategies to different fields based on their local characteristics. Numeric fields use BCD or binary coding for efficient compression, while alphanumeric fields use ASCII coding. This localized approach optimizes compression efficiency for each field type independently, achieving high compression ratios without requiring significant data accumulation.

Inventive Principle:
Principle #3Local quality

2Loss of time

If data is transmitted without compression to maintain immediacy, then real-time transmission is achieved, but data transfer size increases

Engineering Contradiction:
Improvetransmission delayVSAvoiddata transfer size
Core Design Contradiction:
Loss of timeVSQuantity of substance

Solution Approach 1:

The patent performs preliminary compression of data fields as they are generated, before transmission is required. By applying coding schemes to individual fields in real-time without waiting for data accumulation, the system achieves both immediate transmission capability and reduced data transfer sizes. The compression action is taken in advance at the field level, eliminating the need to choose between compression and immediacy.

Inventive Principle:
Principle #10Preliminary action

3Quantity of substance

If conventional compression methods are used, then some compression is achieved, but they are not optimized for fixed-length data structures with multiple data types

Engineering Contradiction:
Improvecompression ratioVSAvoiddata type handling
Core Design Contradiction:
Quantity of substanceVSAdaptability or versatility

Solution Approach 1:

The patent applies different compression strategies to different fields based on their local characteristics. Numeric fields use BCD or binary coding for efficient compression, while alphanumeric fields use ASCII coding. This localized approach optimizes compression efficiency for each field type independently, achieving high compression ratios without requiring significant data accumulation.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The patent changes the representation parameters of data fields based on their type. Numeric values are converted from decimal to BCD or binary representation, which uses fewer bits. Alphanumeric values are converted to ASCII coding. These parameter changes are applied selectively to different fields, optimizing the overall compression ratio for the fixed-length data structure.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP3490153B1Data compression coding method, device therefor, and program therefor
Publication Date: 2023.11.01 KOUSOKUYA
  • EP3490153B1 patent drawingFigure 1
  • EP3490153B1 patent drawingFigure 2(a)~2(c)
  • EP3490153B1 patent drawingFigure 3

AI summary

A compression coding method, apparatus, and program suitable for continuously coding pieces of fixed length data are provided. The compression coding method includes: dividing, into columns each with a predetermined bit width, records consisting of a fixed-length bit string that includes one or more fields, pieces of data of the same type being described in the same field among fields determined in advance; and determining, for each column, a probability of occurrence of a bit value in the column at the same position among a plurality of records, and coding the plurality of records on the basis of the probability of occurrence using an entropy coding method.