Multiradix Data Coding for Near-Minimum Bit Usage
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Efficient data compression for sequences of symbols with a radix that is not a power of 2 poses challenges, as existing methods like Golomb Coding struggle to achieve theoretical minimum bit usage due to limitations in binary representation.
Innovation Solution
The implementation of a radix coder with buffering capabilities allows for efficient coding of symbols with a radix that is not a power of 2 by dynamically adjusting buffer lengths based on data statistics and using a multiradix approach to handle varying symbol ranges, enabling effective compression of data sequences with non-power of 2 radices.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If Golomb Coding is used for data compression, then data can be compressed, but it cannot achieve theoretical minimum bit usage when radix is not a power of 2
Solution Approach 1:
The patent changes the fundamental parameter of representation from binary (base 2) to arbitrary radix (base g). By representing symbols directly in their native radix g rather than converting to binary, the system achieves optimal compression efficiency without the bit waste inherent in binary-based Golomb coding when g is not a power of 2.
Solution Approach 2:
The patent replaces the mechanical binary conversion process with a direct radix-g arithmetic system. Instead of converting radix-g symbols to binary for processing and then converting back, the system performs coding operations directly in radix-g, eliminating the inefficient intermediate conversion steps.
2Ease of operation
If binary representation is used for coding symbols with radix not a power of 2, then data can be processed in binary systems, but additional bits are required beyond theoretical minimum
Solution Approach 1:
The patent transitions from the conventional binary dimension to an arbitrary radix dimension. By operating in radix-g arithmetic rather than binary, the system creates a new dimensional space for data representation that naturally accommodates symbols with radix g, eliminating the need for padding bits and achieving compact representation.
Solution Approach 2:
The patent creates a universal coding system that can handle any radix g efficiently. The radix coder is designed to work with symbols of any radix by using generalized arithmetic operations in base g, making it universally applicable rather than limited to binary or power-of-2 radices.
3Ease of manufacture
If fixed buffer size is used in coding, then implementation is simpler, but cannot adapt to varying data statistics for optimal compression
Solution Approach 1:
The patent implements dynamic buffer management where the buffer size adapts to the actual data statistics. Instead of using a fixed buffer size, the system adjusts the buffer dynamically based on the observed frequency distribution of symbols, allowing optimal compression efficiency for varying data patterns while maintaining implementation feasibility.
Solution Approach 2:
The system uses feedback from the coded data stream to adjust buffer management parameters. By monitoring the coding process and the statistics of the input data, the system dynamically adjusts buffer allocation to match the actual symbol distribution, improving compression efficiency without requiring complex predetermined configurations.
Data Source
AI summary
Embodiments of methods, systems and/or devices for data coding are disclosed. Briefly, in accordance with one embodiment, data is coded and transmitted via an input/output portion of a computing platform to one or more other portions of the computing platform. An interconnect is employed to facilitate transmitting the coded data.