Lossless Data Compression Using Delta Coding and Rice Encoding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing lossless data compression methods for small data sets, such as personalized hearing profiles, suffer from poor compression ratios and inefficient encoding, leading to prolonged transfer times and increased power consumption in portable electronic devices.
Innovation Solution
A lossless data compression method that includes delta-coding iterations to reduce variance and mean, using k-parameters for efficient encoding, and treating the last differential value separately to optimize encoding efficiency, combined with Rice encoding for improved compression.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If lossless compression algorithms are used to compress small data sets (less than 150 bytes), then data integrity is maintained, but compression ratio is poor (below 25%) due to excessive metadata overhead
Solution Approach 1:
The patent divides the data compression process into distinct segments: differential coding stage, run-length encoding stage, and variable-length encoding stage. Each stage processes specific patterns of data independently, allowing optimized handling of different data characteristics without uniform metadata overhead across the entire data set
Solution Approach 2:
The patent dynamically changes encoding parameters based on the actual data characteristics being processed. The variable-length encoding adapts to the frequency and patterns observed in the differential coded data, adjusting code assignments to minimize overhead while maintaining lossless compression for small data sets
2Productivity
If traditional compression algorithms are applied to hearing profile data, then some compression is achieved, but data loss occurs which is unacceptable for hearing profiles
Solution Approach 1:
The patent performs differential coding as a preliminary action before final encoding. This transforms the original hearing profile data into a form where redundant information is already reduced, allowing subsequent encoding stages to achieve better compression ratios while guaranteeing lossless reconstruction of the original data
3Productivity
If multiple iterations of delta-coding are performed to reduce variance and mean, then encoding efficiency is improved, but computational complexity increases
Solution Approach 1:
The patent implements a dynamic, multi-stage compression process where the complexity of each stage is activated based on the characteristics of the data at that point. The run-length encoding and variable-length encoding stages are dynamically applied based on patterns detected in the differential coded data, optimizing encoding efficiency without uniformly increasing computational complexity across all data types
Data Source
Figure 1~2b
Figure 3~5
Figure 6~8
AI summary
Provided is a lossless data compression method (1800) that comprises the receiving of a series (210) of data segments of a known length and a known characteristic and the pre- processing (1810) 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 (1820) of the series of differential data into at least two blocks of differential data and the encoding (1830) 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 (1840) of a data packet (220) 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.