Pixel Delta-Sigma Quantizer for Analog Residue Signal Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image sensor technologies face challenges in efficiently processing analog signals into digital signals within each pixel due to noise introduction and loss of analog residue, leading to reduced sensitivity and image quality, especially when dealing with weak signals or short integration periods.
Innovation Solution
A delta-sigma quantizer is implemented within each pixel, which integrates analog signals, converts them into digital output using a comparator and DAC, and captures analog residue to minimize noise and maintain signal integrity, allowing for high dynamic range and accurate A/D conversion by preserving and compressing errors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional A/D conversion is performed off-chip for CCD image sensors, then compatibility with CCD technology is maintained, but signal transmission distance increases causing noise and loss of analog residue
Solution Approach 1:
The A/D conversion system is segmented into multiple stages: a first quantizer at the pixel level performs initial conversion, and a second quantizer processes the analog residue. This segmentation allows partial on-chip conversion while maintaining CCD compatibility, reducing signal transmission distance and preserving signal integrity.
Solution Approach 2:
An analog residue signal acts as an intermediary between the first and second quantizers. This intermediary carries the remaining analog information after initial quantization, allowing further processing without complete digital conversion, thus maintaining signal integrity while managing system complexity.
2Productivity
If pixel-level A/D conversion is implemented in CMOS image sensors, then conversion speed increases, but circuit footprint and complexity increase
Solution Approach 1:
Instead of performing complete A/D conversion at the pixel level, the system performs partial conversion using a first quantizer that outputs both digital MSBs and analog residue. This partial action achieves speed benefits while leaving remaining conversion to be performed elsewhere, reducing the footprint requirement.
Solution Approach 2:
The system separates the A/D conversion process into different dimensions: the first quantizer handles the most significant bits in one dimension, while the analog residue processing handles the least significant bits in another dimension. This dimensional separation allows parallel processing and reduces the footprint of any single conversion unit.
3Measurement precision
If analog residue is discarded after quantization, then circuit complexity is reduced, but measurement precision and dynamic range decrease
Solution Approach 1:
The system discards the analog residue after it has served its purpose of carrying less significant bit information through the second quantizer. This selective discarding and recovering approach maintains precision by utilizing the residue during processing while simplifying the system by not requiring permanent storage or complex handling of the residue signal.
Data Source
Figure 1~2
Figure 3
Figure 4
AI summary
In one or more embodiments, an apparatus and method for processing an analog signal into a digital signal includes a quantizer that converts the analog signal, which can have any value within a given range of values, into a fixed set of discrete values. An analog residue, i.e. the quantization error caused by the difference between the analog value of the integrated analog signal and the closest corresponding discrete quantized value, is outputted. The analog residue can be further processed to increase the accuracy of the A/D conversion. Multiple quantizer stages can be provided to perform A/D conversion of the analog signal over multiple integration periods, e.g. in multi-shot and time-delay integration applications. The analog signal may represent an image signal.