Multi-Stage Voting Circuit for ADC Bubble Error Correction

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional bubble error rejecters in analog-to-digital converters, particularly those using interpolation, fail to effectively correct bubble errors spanning multiple bits, leading to incomplete removal of errors in interpolating flash ADCs and folding-interpolating ADCs.

Innovation Solution

A cascade of voting sections is introduced, comprising front and rear voting sections that generate correction codes from thermometer codes, with each voter determining majority bit values from sets of consecutive or adjacent codes to correct bubble errors across multiple bits.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If a conventional single-stage bubble error rejecter is used, then the circuit complexity is low, but it fails to correct bubble errors spanning multiple bits in interpolating ADCs

Engineering Contradiction:

Improvebubble error correction capabilityVSAvoidvoting section structure

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The bubble error rejecter is divided into multiple voting sections (first voting section, second voting section, third voting section, etc.) that process thermometer codes in stages. Each voting section handles specific bit ranges, allowing multi-bit bubble errors to be corrected systematically while maintaining manageable circuit complexity at each stage.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent introduces a temporal dimension by processing thermometer codes through multiple sequential voting stages rather than a single parallel stage. This multi-stage approach transforms the error correction problem from a single-complex operation into multiple simpler sequential operations, enabling correction of multi-bit errors.

Inventive Principle:

Principle #17Another dimension (Dimensionality change)

2Measurement precision

If interpolation is used to improve input signal resolution, then the ADC resolution increases, but bubble errors spanning multiple bits occur more frequently

Engineering Contradiction:

Improveinput signal resolutionVSAvoidthermometer code accuracy

Core Design Contradiction:

Measurement precisionVSReliability

Solution Approach 1:

The first voting section performs preliminary error correction on thermometer codes before they are used by subsequent voting sections. This preliminary action removes single-bit and some multi-bit bubble errors early in the processing chain, preventing them from propagating and complicating later correction stages.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

Each voting section receives both original thermometer codes and correction codes from previous voting sections. This feedback mechanism allows each stage to build upon the corrections made by previous stages, systematically eliminating multi-bit bubble errors through iterative refinement.

Inventive Principle:

Principle #23Feedback

Data Source

PatentUS7327292B2Bubble error rejecter in data converter

Publication Date: 2008.02.05 SAMSUNG ELECTRONICS CO LTD

AI summary

A bubble error rejecter includes a cascade of front and rear voting sections for correcting bubble errors spanning multiple bits from interpolation. The front voting section generates first correction codes from first thermometer codes determined from preamplified signals. The rear voting section generates second correction codes from the first correction codes and second thermometer codes determined from interpolation of the preamplified signals.