Dual Conversion Gain Image Sensor Noise Cancellation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Dual conversion gain (DCG) image sensors suffer from image quality degradation due to power supply noise, which affects signal quantization and is exacerbated by parasitic capacitance and noise variations in different frequency ranges, impacting both low and high conversion gain modes.

Innovation Solution

A dual conversion gain image sensor with a power supply noise cancellation circuit that mimics the transfer functions of both low and high conversion gain modes, using independent paths to decouple noise cancellation mechanisms and produce ramp voltage signals, thereby mitigating the effects of power supply noise.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Device complexity

If a DCG image sensor uses a shared power supply noise cancellation path for both low and high conversion gain modes, then the device complexity is reduced, but the noise cancellation effectiveness deteriorates due to conflicting transfer function requirements

Engineering Contradiction:
Improvepower supply noise cancellation circuit complexityVSAvoidnoise cancellation effectiveness
Core Design Contradiction:
Device complexityVSReliability

Solution Approach 1:

The power supply noise cancellation circuit is segmented into two independent paths: a low conversion gain path and a high conversion gain path. Each path is dedicated to canceling noise in its respective mode, allowing each path to be optimized for its specific transfer function requirements without interference from the other mode's requirements.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The circuit dynamically switches between the low conversion gain path and the high conversion gain path based on the operating mode. The low gain select switch and high gain select switch enable the circuit to adapt its configuration to match the required transfer function for the current conversion gain mode, ensuring optimal noise cancellation performance in both modes.

Inventive Principle:
Principle #15Dynamics

2Device complexity

If a single power supply noise cancellation path is used for both conversion gain modes, then the circuit design is simplified, but the transfer function matching accuracy deteriorates

Engineering Contradiction:
Improvecircuit design complexityVSAvoidtransfer function matching accuracy
Core Design Contradiction:
Device complexityVSMeasurement precision

Solution Approach 1:

The noise cancellation circuit is divided into separate low gain and high gain paths, each with its own transfer function optimization. This segmentation allows each path to be independently tuned to accurately match the specific transfer function characteristics of its corresponding conversion gain mode, achieving high matching accuracy for both modes simultaneously.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Each path (low gain path and high gain path) is designed with local quality optimized for its specific function. The low gain path components are optimized for low conversion gain mode characteristics, while the high gain path components are optimized for high conversion gain mode characteristics, ensuring precise transfer function matching in each local operating condition.

Inventive Principle:
Principle #3Local quality

3Device complexity

If power supply noise cancellation is not implemented, then the device complexity is minimized, but image quality deteriorates due to noise affecting signal quantization

Engineering Contradiction:
Improveoverall device complexityVSAvoidpower supply noise impact on image quality
Core Design Contradiction:
Device complexityVSObject-affected harmful factors

Solution Approach 1:

The power supply noise is extracted and canceled separately from the signal path using dedicated noise cancellation paths. By taking out the noise cancellation function into independent low gain and high gain paths, the main signal path remains clean while the noise is actively suppressed, improving image quality without significantly increasing overall device complexity.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The noise cancellation paths act as intermediaries between the power supply and the signal processing chain. These intermediate paths capture and cancel the power supply noise before it can affect the signal quantization, serving as a protective buffer that improves image quality while maintaining relatively simple overall device architecture.

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentUS11418734B1Dual conversion gain image sensor
Publication Date: 2022.08.16 OMNIVISION TECH (SHANGHAI) CO LTD
  • US11418734B1 patent drawing
  • US11418734B1 patent drawing
  • US11418734B1 patent drawing

AI summary

The present invention provides a dual conversion gain image sensor comprising: a pixel circuit, through which pixel power supply voltage noise is transferred to a bit line; a power supply noise cancellation circuit with an input to which the pixel power supply voltage is applied, the power supply noise cancellation circuit mimicly producing a first transfer function with the aid of a low conversion gain path, the power supply noise cancellation circuit mimicly producing a second transfer function with the aid of a high conversion gain path; and a comparator. According to the present invention, the low and high conversion gain paths are two independent power supply noise cancellation paths that result in different transfer functions capable of tracking the variation of the pixel power supply voltage in low and high conversion gain modes.