Variable-Length Numeric Coding With Base-2n Digit Prefixes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional techniques for representing numerical values with variable-length bit representations result in longer code lengths for small values due to the addition of bits indicating the number of significant figures, limiting high compression rates.
Innovation Solution
A coding method that converts numerical values into base-2n representation and adds an instantaneous code indicating the number of digits, allowing for shorter code lengths for small values and efficient compression by associating bits with the number of digits from the head, enabling identification of the number of digits from the instantaneous code.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If variable-length bit representation is used with bits indicating the number of significant figures, then numerical values can be represented with adaptive code lengths, but small numerical values result in longer code lengths reducing compression rate
Solution Approach 1:
The patent extracts the information about the number of significant figures from the code structure itself and uses it to determine the base-2n representation. By separating the significant figure count from the actual value encoding, the system can optimize the code length for each numerical value based on its magnitude, achieving shorter codes for small values while maintaining adaptability for larger values.
Solution Approach 2:
The patent changes the parameter of code length by dynamically selecting the base-2n representation according to the number of significant figures. This parameter change allows the system to use smaller bases (shorter codes) for small numerical values and larger bases (longer codes) for large numerical values, thereby optimizing the overall compression rate.
2Device complexity
If fixed base representation is used, then code structure is simplified, but code length cannot be optimized for different magnitude numerical values
Solution Approach 1:
The patent introduces dynamics into the code structure by making the base of representation variable rather than fixed. The base-2n changes dynamically based on the number of significant figures of the numerical value being encoded. This dynamic adaptation allows the system to maintain relatively simple code structures for small values while achieving optimized compression for mixed-magnitude datasets.
Data Source
AI summary
A non-transitory computer-readable recording medium having stored therein a coding program that causes a computer to execute a process. The process includes coding a numerical value to be coded, into a numeric code of base-2n representation; and generating code data that have been added with an instantaneous code indicating the number of digits of the base-2n representation of the numerical value to be coded, wherein “n” is a natural number equal to or greater than 1.