Switchless Active Pixel Sensor Architecture for CMOS Image Sensors
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional voltage mode and current mode active pixel sensors face limitations in resolution due to high dark currents, temporal noise, and fixed pattern noise, primarily caused by the number and placement of transistors and interconnections per pixel, which affect image quality and sensitivity.
Innovation Solution
The solution involves a current/voltage mode active pixel sensor architecture with a pixel array comprising a photodiode, a reset transistor, and a readout transistor, where the addressing switch is moved outside the pixel, reducing the transistor count to 2 or 1.5 per pixel, enhancing linearity between output current and photodiode voltage, and using a biasing circuit with current conveyor and correlated double sampling to minimize fixed pattern noise.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a three transistor (3T) active pixel sensor topology is used, then the device complexity is reduced, but the image quality deteriorates due to high dark currents and high temporal noise
Solution Approach 1:
The patent extracts the address switch transistor from the pixel structure, moving it to the periphery. This removal eliminates the source of spatial variations and fixed pattern noise while maintaining pixel functionality through peripheral switching control
Solution Approach 2:
The patent combines current mode photodetection with voltage mode readout in a hybrid architecture, utilizing the advantages of both modes: current mode provides linear response and low noise, while voltage mode enables standard CMOS integration and signal processing
2Reliability
If a four transistor (4T) active pixel sensor topology is used, then the image quality is improved through lower dark current and correlated double sampling, but the device complexity increases leading to larger pixel size and lower resolution
Solution Approach 1:
The address switch transistor is extracted from the pixel and placed in the periphery, reducing the pixel transistor count from 4 to 2 while maintaining the ability to perform correlated double sampling and control pixel operation
Solution Approach 2:
The peripheral address switches serve multiple pixels simultaneously, providing addressing functionality for the entire pixel array rather than requiring dedicated switches within each pixel. This multi-functionality reduces overall device complexity while maintaining image quality
3Device complexity
If the address switch transistor is moved outside the pixel, then the device complexity is reduced and linearity is enhanced, but the ease of operation may be affected by peripheral switching control
Solution Approach 1:
The patent transitions from intra-pixel switching to inter-pixel switching by moving the address switch to the periphery. This dimensional change in control architecture allows single switches to address multiple pixels through row and column bus structures
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 results in higher resolution, reduced pixel size, lower fixed pattern noise, and improved signal-to-noise ratio, enabling higher frame rates and more efficient imaging with increased sensitivity and linearity.
Implementation Method 1
each pixel in the pixel array comprises a photodiode
Data Source
AI summary
A voltage and current mode active pixel sensor for high resolution imaging is presented. The photo pixel is composed of a photodiode and two transistors: reset and transconductance amplifier transistor. The switch transistor is moved outside the pixel, allowing for lower pixel pitch and increased linearity of the output photocurrent. The reset and amplifier (readout) transistors may also be shared among adjacent pixels by the introduction of transfer switches between the photodiodes and the source of the reset transistor and the gate of the readout transistor. The switch transistor outside the pixels provides biasing voltages or currents to the readout transistors to selectively turn them on when readout of the corresponding photodiode is desired and turns the readout transistor off when the corresponding photodiode is not to be read out. The increased linearity of the image sensor has greatly reduced spatial variations across the image after correlated double sampling and the column fix pattern noise is greatly improved.