CMOS Image Sensor Pixel Architecture for Miniaturized Endoscopes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional very small area CMOS image sensors face challenges in miniaturization and cost-effectiveness, particularly in endoscopic applications where they need to be small enough to fit within narrow tubes while maintaining high image quality and dynamic range, and reducing the number of contact pads to minimize wire usage.
Innovation Solution
A CMOS image sensor with a 4T pixel architecture that directly transmits analog values to a host device, eliminating large column capacitors and using an on-chip clock with a novel output synchronization protocol to reduce contact pads, allowing for smaller chip size and lower production costs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If conventional CMOS image sensors use large column capacitors and on-chip A/D conversion circuitry, then image quality and dynamic range are maintained, but chip area increases significantly
Solution Approach 1:
The patent extracts the A/D conversion function and large column capacitors from the chip interior to external devices. The pixel array directly outputs analog values without internal A/D conversion, eliminating the need for large on-chip capacitors and complex conversion circuitry, thereby significantly reducing chip area while maintaining image quality through external processing.
Solution Approach 2:
The patent moves the A/D conversion operation from the spatial dimension (on-chip circuitry) to the temporal dimension (external processing after data transmission). By transmitting analog values directly outside the chip for conversion, the system achieves area reduction while preserving functionality through a different operational dimension.
2Loss of substance
If the number of contact pads is reduced to minimize wire usage, then wire quantity decreases, but signal transmission capability is limited
Solution Approach 1:
The patent makes the reduced set of contact pads multi-functional by having them handle both clock signal transmission and data output through time-division multiplexing. The same physical pads serve multiple purposes: transmitting clock signals during certain periods and analog pixel values during other periods, thereby minimizing wire quantity while maintaining full signal transmission capability.
Solution Approach 2:
The patent employs periodic action by using time-division multiplexing where contact pads alternately transmit clock signals and data signals in different time periods. This periodic switching allows the same physical infrastructure to carry multiple types of signals, reducing wire quantity without sacrificing transmission capability.
3Area of stationary object
If chip size is reduced to fit within narrow tubes, then endoscopic applicability increases, but manufacturing precision requirements increase
Solution Approach 1:
The patent extracts complex circuitry (A/D converters, large capacitors) from the chip to external devices, leaving only the pixel array and minimal output circuitry on-chip. This extraction simplifies the manufacturing process for small chips while maintaining high image quality, as the complex functions are performed externally where precision requirements can be better managed.
4Area of stationary object
If on-chip A/D conversion is eliminated, then chip area is reduced, but external processing complexity increases
Solution Approach 1:
The patent introduces an intermediary external device that performs A/D conversion and image processing functions. This external intermediary handles the complex processing tasks that were previously done on-chip, allowing the chip itself to remain simple and small while the complexity is managed externally where it can be better accommodated.
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
Enables the production of CMOS image sensors with dimensions of 0.7 mm or less, fitting into narrow tubes, while maintaining high image quality and dynamic range, and reducing production costs, suitable for endoscopic applications.
Implementation Method 1
Each pixel includes a sensing element (e.g., a photodiode) that is capable of converting a portion of an optical (or other radiant source) image into an electronic (e.g., voltage) signal
Data Source
AI summary
An endoscope system includes a host device and an endoscope including a very small area CMOS image sensor having only four pads (power, ground, digital in, analog out), and including an array of 4T pixels and associated control circuitry for performing correlated double sampling (CDS) to generate analog reset level and analog signal level values associated with light detected by photodiodes in each pixel. Instead of processing the analog values on-chip, the analog reset values and analog signal values are transmitted in separate sets one row at a time along with interleaved synchronization signals by way of a single analog contact pad to the host device of the endoscopic system, which uses the synchronization signals to reconstruct the sensor's internal clock in order to process the analog values. The endoscope housing thus requires only four wires and is made very small.


