TFET Differential Amplifier for Low-Noise Imaging

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional semiconductor imaging devices face challenges in reducing thermal noise from transistors, as the subthreshold coefficient of MOSFETs has a theoretical lower limit, making it difficult to suppress thermal noise below a certain noise level.

Innovation Solution

The use of Tunnel Field Effect Transistors (TFETs) for amplifying the difference between pixel and reference signals, along with a MOSFET supplying constant current, reduces thermal noise and allows for a more compact design by sharing sources and drains on a substrate, thereby reducing noise and power consumption.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Object-affected harmful factors

If MOSFETs are used for the differential pair in the comparator, then the device can operate with standard transistor characteristics, but thermal noise cannot be suppressed below a certain noise level due to the theoretical lower limit of the subthreshold coefficient

Engineering Contradiction:
Improvethermal noiseVSAvoidnoise suppression capability
Core Design Contradiction:
Object-affected harmful factorsVSReliability

Solution Approach 1:

The patent changes the fundamental transistor parameter (subthreshold coefficient) by replacing MOSFETs with TFETs. TFETs inherently possess a lower subthreshold coefficient due to their tunneling-based operation mechanism, enabling thermal noise suppression below the conventional MOSFET limit without requiring additional circuit modifications or parameter adjustments within the same device type.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The patent substitutes the conventional MOSFET transistor mechanism with a TFET transistor mechanism. This replacement fundamentally changes the underlying physical operation from field-effect modulation to band-to-band tunneling, achieving superior noise characteristics by employing a different transistor technology platform rather than optimizing the existing MOSFET structure.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

2Speed

If an ADC is arranged for each pixel to improve reading speed, then image data can be read faster, but the circuit area and complexity increase significantly

Engineering Contradiction:
Improvereading speedVSAvoidcircuit area
Core Design Contradiction:
SpeedVSArea of stationary object

Solution Approach 1:

The patent merges the ADC functionality into a shared resource that serves multiple pixels, rather than dedicating one ADC per pixel. By implementing a column-parallel ADC architecture where each column shares an ADC, the system achieves high-speed conversion while significantly reducing the total circuit area compared to a fully pixel-parallel ADC implementation.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The patent segments the ADC architecture into column-level units rather than pixel-level units. This segmentation strategy divides the image sensor into multiple columns, each with its own ADC, allowing parallel processing across columns while reducing the per-pixel circuit complexity and area requirements compared to having individual ADCs for every pixel.

Inventive Principle:
Principle #1Segmentation

3Object-affected harmful factors

If constant current is increased to reduce random noise, then noise performance improves, but power consumption increases

Engineering Contradiction:
Improverandom noiseVSAvoidpower consumption
Core Design Contradiction:
Object-affected harmful factorsVSUse of energy by moving object

Solution Approach 1:

The patent changes the transistor type from MOSFET to TFET, which fundamentally alters the noise characteristics. TFETs exhibit lower random noise due to their tunneling mechanism, achieving superior noise performance without requiring increased bias current. This parameter change in transistor physics allows noise reduction while maintaining or reducing power consumption levels.

Inventive Principle:
Principle #35Parameter changes

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

This approach effectively suppresses thermal noise and random noise components, reducing the need for increased constant current, which in turn lowers power consumption while maintaining image data quality.

Implementation Method 1

a pair of TFETs (Tunnel Field Effect Transistors) for amplifying the difference between the pixel signal and a predetermined reference signal

Methodology Applied
Scientific EffectTunnel field effect: Electromagnetic Induction

Implementation Method 2

a MOSFET (Metal Oxide Semiconductor Field Effect Transistor) that supplies a constant current to the pair of TFETs

Methodology Applied
Scientific EffectField effect transistor conduction: Conduction (electrical)

Implementation Method 3

a pixel circuit that generates a pixel signal by photoelectrically converting incident light

Methodology Applied
Scientific EffectPhotoelectric conversion: Photoelectric Effect

Data Source

PatentUS11350050B2Semiconductor integrated circuit and imaging device
Publication Date: 2022.05.31 SONY SEMICON SOLUTIONS CORP
  • US11350050B2 patent drawing
  • US11350050B2 patent drawing
  • US11350050B2 patent drawing

AI summary

In a solid-state imaging element provided with a differential pair of transistors, noise of a signal from the differential pair is reduced. The semiconductor integrated circuit includes a pixel circuit and a pair of TFETs (Tunnel Field Effect Transistors). In the semiconductor integrated circuit, the pixel circuit photoelectrically converts incident light to generate a pixel signal. Further, in the semiconductor integrated circuit, the pair of TFETs amplifies the difference between the pixel signal generated by the pixel circuit and a predetermined reference signal that changes with time, and outputs the amplified difference as a differential amplification signal.