ADC and DAC Calibration for Nonlinearity Compensation

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing electronic systems with AD and DA conversion units face challenges in reducing the area and power consumption of semiconductor chips due to non-linearity issues, particularly in foreground digital correction type A/D converters, where increasing resolution in the reference D/A conversion unit significantly increases chip area and power consumption.

Innovation Solution

An electronic system that includes A/D and D/A conversion units, AD and DA conversion compensation units, and a calibration unit, which sets operating characteristics during a calibration period to compensate for non-linearity in both AD and DA conversions, using a calibration unit to adjust the operating characteristics of the AD and DA conversion compensation units to emulate and cancel non-linearity, thereby reducing the need for high-resolution reference D/A conversion units.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If the resolution of the reference D/A conversion unit is increased to enhance precision during calibration, then the measurement precision of the A/D conversion is improved, but the area occupied by the semiconductor chip and power consumption increase significantly

Engineering Contradiction:

Improveprecision of analog output signalVSAvoidarea occupied by reference D/A conversion unit

Core Design Contradiction:

Measurement precisionVSArea of stationary object

Solution Approach 1:

The patent changes the resolution parameter of the reference D/A conversion unit from high (e.g., 16-bit) to low (e.g., 1-bit or 1.5-bit) by introducing dither signals and using statistical processing. This parameter change allows low-resolution hardware to achieve high-precision calibration through multiple measurements and averaging, thereby reducing chip area while maintaining measurement precision.

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

Instead of using a single high-resolution reference D/A conversion unit, the patent employs multiple low-resolution reference D/A conversion units that collectively provide the equivalent precision of a high-resolution unit through ensemble averaging. This copying approach with multiple low-cost components replaces one expensive high-resolution component.

Inventive Principle:

Principle #26Copying

2Measurement precision

Engineering Contradiction:

Improveprecision of analog output signalVSAvoidpower consumption of reference D/A conversion unit

Core Design Contradiction:

Measurement precisionVSUse of energy by stationary object

Solution Approach 1:

The patent changes the resolution parameter of the reference D/A conversion unit from high to low, which directly reduces power consumption. Low-resolution D/A converters consume significantly less power than high-resolution ones, and the patent compensates for the reduced resolution through dither signals and statistical processing methods.

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The patent employs periodic dither signals added to the reference D/A conversion output during calibration. These periodic signals enable the system to extract precise transfer characteristic information through correlation processing or averaging, allowing low-power low-resolution hardware to achieve high-precision calibration.

Inventive Principle:

Principle #19Periodic action

3Measurement precision

If a background digital correction type A/D converter is used, then the measurement precision is improved through high-resolution reference A/D conversion, but the area occupied by the semiconductor chip increases

Engineering Contradiction:

Improvehigh-resolution A/D conversionVSAvoidtotal area occupied by semiconductor chip

Core Design Contradiction:

Measurement precisionVSArea of stationary object

Solution Approach 1:

The patent inverts the conventional background correction approach by performing calibration in the foreground (before normal operation) rather than in the background (during operation). This foreground calibration using low-resolution reference D/A conversion with dither signals eliminates the need for a large high-resolution reference A/D converter, thereby reducing total chip area while maintaining precision.

Inventive Principle:

Principle #13The other way round (Inversion)

Solution Approach 2:

The patent performs A/D converter calibration in advance during a dedicated foreground calibration period before normal operation begins. This preliminary calibration establishes correction data that is stored and applied during normal operation, eliminating the need for continuous high-resolution reference conversion and reducing overall system area requirements.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS9054723B2Electronic system and operating method thereof

Publication Date: 2015.06.09 RENESAS ELECTRONICS CORP

AI summary

To compensate for non-linearity of an AD conversion unit and non-linearity of a DA conversion unit in an electronic system including the DA conversion unit and the AD conversion unit, an electronic system includes an A/D conversion unit, a D/A conversion unit, an AD conversion compensation unit, a DA conversion compensation unit, and a calibration unit. During a calibration operation period, the calibration unit sets an operating characteristic of the AD conversion compensation unit and an operating characteristic of the DA conversion compensation unit. The operating characteristic of the AD conversion compensation unit set during the calibration operation period compensates for non-linearity of AD conversion of the A/D conversion unit. The operating characteristic of the DA conversion compensation unit set during the calibration operation period compensates for non-linearity of DA conversion of the D/A conversion unit.