Differential Rice Encoding for Small Lossless Data Blocks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing lossless data compression methods are inefficient for small data sets, such as hearing profiles, resulting in poor compression ratios and increased transfer times, especially when used in low-speed control channels for high-speed data channels in portable electronic devices.
Innovation Solution
A lossless data compression method that preprocesses data into differential values, divides them into blocks, and uses Rice encoding with k-parameters to optimize encoding, allowing for efficient compression and decompression, even with limited resources, by minimizing overhead and metadata.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If lossless compression algorithms are used on small datasets (less than 150 bytes), then data integrity is maintained, but compression ratio is poor (below 25%) due to excessive overhead metadata
Solution Approach 1:
The patent segments the data into fixed-size blocks (e.g., 16 bytes each) and processes them independently with Rice encoding. This segmentation allows the algorithm to achieve better compression ratios on small datasets by reducing the relative overhead of metadata compared to processing the entire small dataset as a single unit.
Solution Approach 2:
The patent changes the encoding parameter by using Rice encoding with variable k-values adapted to the statistical properties of each block. This parameter adaptation allows optimal compression for different data patterns while maintaining lossless reconstruction, improving overall compression ratio without sacrificing data integrity.
2Productivity
If traditional compression algorithms are applied to hearing profiles, then some compression is achieved, but data loss occurs which is unacceptable for hearing profile transmission
Solution Approach 1:
The patent replaces traditional mechanical compression approaches (like dictionary-based or transform coding) with a statistical arithmetic coding approach (Rice encoding). This substitution maintains exact data reconstruction by using probabilistic modeling of differential values, achieving lossless compression suitable for hearing profiles while improving compression efficiency.
3Measurement precision
If more overhead metadata is included in compressed data packets, then decoding accuracy is improved, but transmission time and power consumption increase on low-speed control channels
Solution Approach 1:
The patent performs preliminary actions by pre-calculating and embedding only the essential k-parameter needed for Rice decoding, rather than including complete statistical models or multiple parameters. This preliminary optimization reduces metadata overhead while maintaining sufficient decoding accuracy for the application.
Solution Approach 2:
The patent applies partial action by including only the minimum necessary metadata (k-parameter) for accurate decoding, rather than excessive metadata that would improve decoding further but increase transmission overhead. This partial approach optimizes the trade-off between decoding accuracy and transmission efficiency.
Data Source
AI summary
A lossless data compression method that comprises the receiving of a series of data segments of a known length and a known characteristic and the pre-processing of the series of data segments to generate a series of differential data, which comprises a first value, one or more differential values and a last differential value. The method further comprises the dividing of the series of differential data into at least two blocks of differential data and the encoding of at least one block of differential data using Rice encoding, thereby providing at least one block of encoded differential data. In addition, the method comprises the generating of a data packet that comprises said at least one blocks of encoded differential data and any blocks of differential data not having been encoded. In addition, a lossless data decompression method, a portable electronic device for data compression, a portable electronic device for data decompression, a computer program product for data compression and a computer program product for data decompression is provided.