Binary-Coded DAC Mismatch Shaping With Zero-Mean Error Encoding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Mismatch in individual elements of digital-to-analog converters (DACs) degrades the accuracy of both digital-to-analog and analog-to-digital conversions, particularly when used as feedback DACs in delta-sigma converters.
Innovation Solution
The Zero Mean Error Encoding (ZMEE) scheme applies a mismatch shaping technique directly to binary coded DACs, using negative feedback to bound the error from each bit to zero, reducing implementation complexity by eliminating the need for binary-to-thermometer decoding and element shuffling.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional binary-to-thermometer decoding and element shuffling are used for mismatch shaping, then mismatch errors can be shaped, but device complexity increases
Solution Approach 1:
The patent extracts and eliminates the binary-to-thermometer decoder and element shuffling circuits from the traditional mismatch shaping architecture. By directly applying mismatch shaping to binary-coded DAC elements, the invention removes the complex decoding and shuffling stages while preserving the essential mismatch error shaping function, thereby reducing device complexity without sacrificing accuracy
Solution Approach 2:
Instead of converting binary codes to thermometer codes and then shuffling elements to achieve mismatch shaping, the patent inverts the approach by directly shaping mismatch errors in the binary-coded domain. This reversal eliminates the need for code conversion and complex element shuffling, achieving the same accuracy improvement with significantly reduced circuit complexity
2Device complexity
If mismatch shaping is applied to binary coded DACs, then implementation complexity is reduced, but error bounding control becomes more challenging
Solution Approach 1:
The patent implements a feedback mechanism that monitors the usage of each DAC element and dynamically adjusts the coding scheme to bound the error from each bit to zero. This feedback control ensures that positive and negative usages of each element are equalized, providing reliable error bounding while maintaining the simplified binary-coded architecture without requiring complex decoding circuits
Data Source
AI summary
Described herein is a mismatch shaping technique applied in digital-to-analog converters (DACs) for high pass filtering mismatch related errors. The mismatch shaping scheme is based on a zero mean error encoding technique, which can be applied directly to binary coded signals, without the use for binary to thermometer decoding and element shuffling. In at least one example, an apparatus is provided which comprises a mismatch shaping circuitry to receive an N-bit binary input bits and to generate an (N+1)-bit digital output. In at least one example, the apparatus further comprises a digital-to-analog converter to receive the (N+1)-bit digital output and to generate an analog output, wherein the mismatch shaping circuitry is to encode the (N+1)-bit digital output to shape mismatch errors in the digital-to-analog converter.