Image Sensor Sampler Array for True Correlated Double Sampling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image sensors using CMOS technology in TDI mode face limitations in low illumination due to pseudo correlated double sampling noise, which prevents true correlated double sampling and common integration time across all lines.
Innovation Solution
The implementation of an auxiliary matrix with N lines of P individual processing circuits for each row and column, allowing for true correlated double sampling and analog-to-digital conversion, with reset potential levels applied line by line and useful signals sampled and digitized for digital summation across all lines during a common integration time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If pseudo correlated double sampling is used in CMOS TDI sensors, then device complexity is reduced and manufacturing is easier, but measurement precision deteriorates due to kTC noise limiting detection in low light conditions
Solution Approach 1:
The sensor is divided into multiple independent integration lines (N lines), each with its own pixel array. This segmentation allows each line to operate independently with its own integration period, enabling true correlated double sampling while maintaining common integration time across all lines, thus resolving the contradiction between manufacturing ease and measurement precision
Solution Approach 2:
An auxiliary matrix of N lines of P individual processing circuits is introduced as an intermediary between the pixel array and the readout system. These processing circuits perform the correlated double sampling operation, eliminating kTC noise while allowing common integration time for all lines, thereby improving measurement precision without significantly increasing manufacturing complexity
2Measurement precision
If true correlated double sampling is implemented with common integration time for all lines, then measurement precision is improved by minimizing correlated switching noise, but device complexity increases due to the auxiliary matrix of processing circuits
Solution Approach 1:
Multiple functional elements (pixel, transfer transistor, reset transistor, follower transistor, and sampling circuit) are merged into a unified processing circuit structure. This merging reduces the overall device complexity by eliminating redundant components and simplifying the signal path, while still achieving true correlated double sampling with common integration time across all lines
3Measurement precision
If charge transfer is used in CCD sensors, then measurement precision is improved by natural point-by-point signal addition without read noise, but device complexity and manufacturing difficulty increase compared to CMOS technology
Solution Approach 1:
The patent replaces the mechanical charge transfer mechanism of CCD sensors with an electronic signal processing approach in CMOS technology. By using voltage signals and electronic correlation processing instead of physical charge transfer, the system achieves comparable measurement precision while significantly reducing device complexity and manufacturing difficulty
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 enables true correlated double sampling and minimizes correlated switching noise, allowing for improved signal-to-noise ratio and effective detection in low illumination conditions while maintaining a common integration time for all lines.
Implementation Method 1
Each pixel Pi,j comprises a photodiode PD
Data Source
Figure 1~2
Figure 3
Figure 4
AI summary
The invention relates to image sensor having N lines of P active photosensitive pixels in MOS technology (MT1). The sensor comprises digitizing circuits arranged in N lines of P processing circuits (MT2), each processing circuit of line row i and column row j including a respective sampler for carrying out correlated double sampling of the signals on a column conductor (CCj) of row j and corresponding to the observation of an image point during the same integration time for all the lines, and an analog/digital conversion means for providing digital values of the sampled analog signals. The sensor is particularly suitable for operating in TDI mode (image scrolling and integration).