Integrating ADC Non-Uniform Quantization for Accuracy-Power Tradeoffs

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional analog-to-digital converters (ADCs) with non-uniform quantizers face issues such as high power consumption and low accuracy due to their design.

Innovation Solution

The proposed ADC architecture employs a sample-and-hold circuit and a non-uniform quantizer with resistors of differing resistances and a current steering DAC, allowing for multiple measurements within a sample period using currents of different magnitudes to improve accuracy, and optionally includes a predictive algorithm to adjust the reference voltage for higher resolution.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If non-uniform quantizer with resistors of differing resistances is used, then measurement precision is improved, but power consumption increases

Engineering Contradiction:
ImproveaccuracyVSAvoidpower consumption
Core Design Contradiction:
Measurement precisionVSUse of energy by moving object

Solution Approach 1:

The patent implements dynamic reference voltage adjustment based on signal characteristics. The quantizer adapts its reference voltages in real-time according to the input signal's voltage swing and statistical properties, allowing the system to optimize between precision and power consumption based on actual operating conditions rather than using fixed high-precision references throughout.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent changes the parameters of the quantizer dynamically by adjusting reference voltages based on signal statistics. The system monitors input signal characteristics and modifies the quantization levels accordingly, transforming the quantizer from a static to a adaptive structure that can operate at different precision levels to minimize power consumption while maintaining required accuracy.

Inventive Principle:
Principle #35Parameter changes

2Measurement precision

If non-uniform quantizer with resistors of differing resistances is used, then measurement precision is improved, but device complexity increases

Engineering Contradiction:
ImproveaccuracyVSAvoiddevice complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent segments the quantization process into multiple stages with different precision requirements. By dividing the signal range into segments and applying different quantization strategies to each segment, the system achieves high overall precision without requiring all components to operate at maximum precision simultaneously, thereby reducing device complexity.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent performs preliminary analysis of the input signal characteristics before full quantization. By pre-processing the signal to determine its statistical properties and voltage swing range, the system can configure the quantizer optimally in advance, simplifying the main quantization process while maintaining high precision.

Inventive Principle:
Principle #10Preliminary action

3Measurement precision

If multiple measurements within a sample period are performed, then measurement precision is improved, but productivity decreases

Engineering Contradiction:
ImproveaccuracyVSAvoidconversion speed
Core Design Contradiction:
Measurement precisionVSProductivity

Solution Approach 1:

The patent implements periodic multiple measurements within each sample period, taking several measurements at different phases or with different reference voltages and then combining them. This periodic sampling approach improves precision by averaging out noise and errors while maintaining a structured timing that minimizes impact on overall conversion speed.

Inventive Principle:
Principle #19Periodic action

Solution Approach 2:

The patent ensures continuous useful action by overlapping measurement phases and maintaining the conversion pipeline throughout the sample period. Rather than stopping for multiple measurements, the system performs them in an integrated manner where measurements can proceed in parallel or with minimal sequential delay, keeping the conversion process continuously active.

Inventive Principle:
Principle #20Continuity of useful action

Data Source

PatentEP2751927B1Method and apparatus for performing data conversion with non-uniform quantization
Publication Date: 2021.02.24 TEXAS INSTRUMENTS INC
  • EP2751927B1 patent drawingFigure 1
  • EP2751927B1 patent drawingFigure 2~3
  • EP2751927B1 patent drawingFigure 4

AI summary

An analog-to-digital converter (ADC 200-A) functions as an integrating ADC. Sample-and-hold (S/H) circuit (202) samples input signal x(t) with clock signal (SCLK). A digital-to-analog converter (DAC 212) applies a current set by control signal (DCNTL) to change voltage at the output node of the S/H circuit from the sample voltage. Comparator (204) compares this voltage to a reference voltage (REF) and outputs a pulse (part of output signal COUT) to controller (206) when the reference voltage is reached. As the voltage is changed from the sample voltage, counter (208) increments a count value to measure the time. Output circuit (210) generates a digital output signal (y[n]) from the time interval and applied current from DAC (212).