Sigma-Delta Quantizer Correction for Metastability Errors

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Sigma-Delta modulators face stability issues at high sampling frequencies due to parasitic poles and additional delays, leading to metastability errors that degrade performance, especially for small input signals, which affect the signal-to-noise ratio and dynamic range.

Innovation Solution

A data processor with a metastability compensation module that includes a full-scale quantizer, digital-to-analogue converters, and an analogue combining circuit to generate a compensated output signal by correcting metastability errors outside the feedback loop, ensuring the quantized output is independent of the error and maintaining stability.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If the sampling frequency is increased to improve productivity, then the processing speed increases, but metastability errors increase due to parasitic poles and additional delays

Engineering Contradiction:
Improvesampling frequencyVSAvoidmetastability error
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The system is divided into two independent parts: the feedback loop for stability and the correction module for metastability compensation. This segmentation allows the feedback loop to operate at high sampling frequencies while the correction module handles metastability errors separately, resolving the contradiction between high productivity and reliability.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The metastability correction function is extracted from the feedback loop and implemented as a separate correction module. This extraction removes the harmful interaction between feedback delays and metastability, allowing the feedback loop to focus on stability while the correction module addresses metastability errors independently.

Inventive Principle:
Principle #2Taking out (Extraction)

2Productivity

If the sampling frequency is increased to improve productivity, then processing speed improves, but the signal-to-noise ratio deteriorates due to metastability errors

Engineering Contradiction:
Improvesampling frequencyVSAvoidsignal-to-noise ratio
Core Design Contradiction:
ProductivityVSMeasurement precision

Solution Approach 1:

By segmenting the system into feedback loop and correction module, the correction module can specifically target metastability-induced noise without affecting the feedback loop's signal processing. This allows high sampling frequencies to be maintained while improving signal-to-noise ratio through dedicated error correction.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The correction module uses feedback from the quantized output to generate correction signals that compensate for metastability errors. This feedback mechanism in the correction module allows the system to maintain high sampling frequencies while improving measurement precision by actively canceling metastability-induced noise.

Inventive Principle:
Principle #23Feedback

3Measurement precision

If a correction module is added to reduce metastability errors, then the signal-to-noise ratio improves, but the device complexity increases

Engineering Contradiction:
Improvesignal-to-noise ratioVSAvoidcircuit complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The correction module is extracted as a separate entity from the feedback loop, allowing it to be optimized independently. This extraction enables the use of specialized correction techniques without complicating the feedback loop design, thus improving signal-to-noise ratio while managing device complexity through modular architecture.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The correction module serves multiple functions: it compensates for metastability errors, improves signal-to-noise ratio, and maintains compatibility with existing feedback loop designs. This multi-functionality justifies the added complexity by providing comprehensive error correction without requiring complete system redesign.

Inventive Principle:
Principle #6Universality (Multi-functionality)

4Reliability

If the correction module processes signals outside the feedback loop, then metastability errors are corrected without affecting feedback stability, but the correction delay increases

Engineering Contradiction:
Improvefeedback stabilityVSAvoidcorrection delay
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

By segmenting the correction function from the feedback loop, the correction module can operate independently with its own timing. This segmentation allows the feedback loop to maintain its stability-critical timing while the correction module processes signals with minimal delay outside the feedback path.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The correction module performs preliminary correction on the quantized output before it is used for final signal processing. This preliminary action approach allows metastability errors to be corrected in advance without interfering with the feedback loop's real-time stability requirements, thus reducing the impact of correction delay.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP3340475B1Metastability compensation
Publication Date: 2021.04.28 NXP BV
  • EP3340475B1 patent drawingFigure 1~2
  • EP3340475B1 patent drawingFigure 3~4a
  • EP3340475B1 patent drawingFigure 4b~4c

AI summary

The disclosure relates to a data processor comprising: a data processing module comprising: an input for receiving an input signal; an output for providing a quantized output signal; a combining unit configured to combine a feedback signal from the output with the input signal; and a quantizer configured to provide the quantized output signal based on the combined signal, wherein the quantized output signal comprises a metastability error, and a correction module configured to: receive the quantized output signal; generate a full-scale digital signal based on the quantized output signal; determine the a metastability error in the full-scale digital quantized output signal; and provide a compensated output signal based on the quantized output signal and the determined metastability error.