Digital Gain Correction Using Dither to Eliminate Missing Codes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Digital gain error corrections in digital circuits often result in missing codes due to scaling, leading to increased differential nonlinearity (DNL) and patterned noise, which current methods address by increasing ADC resolution, complicating circuitry and requiring additional testing.
Innovation Solution
The method involves adding a random dither to digital codes before scaling and subtracting it after scaling, allowing for selective switching between dithering and non-dithering configurations and varying dither magnitudes to eliminate or reduce missing codes and improve DNL.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If digital scaling is applied to correct gain error, then gain accuracy is improved, but missing codes occur leading to increased DNL
Solution Approach 1:
A dither signal is added to the digital code before scaling operation. This preliminary action ensures that the scaled code transitions through all intermediate values, preventing missing codes and maintaining continuous code distribution after gain correction.
Solution Approach 2:
The dither signal acts as an intermediary element between the original digital code and the scaled output. By introducing this intermediate random signal, the system ensures smooth transitions during scaling without creating gaps in the code sequence, thereby maintaining DNL performance.
2Manufacturing precision
If ADC resolution is increased to eliminate missing codes, then DNL is improved, but circuit complexity and testing requirements increase
Solution Approach 1:
Instead of changing the physical resolution parameter of the ADC, the invention changes the statistical parameters of the code distribution by adding dither. This transforms the deterministic missing code pattern into a probabilistic distribution that fills all code values over time, improving DNL without increasing bit depth.
Solution Approach 2:
The invention creates a statistical copy of the ideal full-range code distribution through dithering, rather than physically implementing higher resolution hardware. The dithered output statistically replicates the code distribution of a higher-resolution system without requiring the actual additional hardware bits.
3Manufacturing precision
If dithering is applied before scaling, then missing codes are eliminated, but noise is introduced
Solution Approach 1:
The dither signal is applied at a level that is excessive for immediate noise concerns but necessary for code continuity. By using a small-amplitude dither that is just sufficient to bridge the scaling gaps, the invention achieves code completeness while minimizing the introduced noise, representing a partial action that balances both requirements.
Data Source
AI summary
A method for correcting digital gain error for a digital code includes receiving the digital code, generating a random number, adding a first dither to the digital code, in which a magnitude of the first dither is determined based on the random number, performing an operation on the digital code including the added dither with a factor to generate a scaled digital code, and subtracting a second dither corresponding to the first dither from the scaled digital code.