Column-Wise Entropy Coding for Real-Time Fixed-Length Sensor Data
Find Innovative SolutionsGenerate 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
Engineering 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
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.
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.
2Loss of time
If data is transmitted without compression to maintain immediacy, then real-time transmission is achieved, but data transfer size increases
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.
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
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.
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.
Data Source
Figure 1
Figure 2(a)~2(c)
Figure 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.