Image Sensor ADC Pedestal Sampling for Better Linearity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional CMOS image sensors suffer from non-linearity errors in analog-to-digital conversion due to the inherent architecture of ADCs, leading to reduced dynamic range and effective resolution of image charge processing.
Innovation Solution
Implementing Correlated Multi-Sampling (CMS) with randomized ADC pedestals within the row time to reduce non-linearity errors by oversampling and averaging errors, thereby improving ADC linearity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional ADC architecture is used for image sensor readout, then device complexity is reduced and ease of manufacture is improved, but non-linearity errors increase causing reduced dynamic range and effective resolution
Solution Approach 1:
The patent segments the readout process into multiple independent sampling operations. Instead of a single complex ADC conversion, the system performs multiple simpler ADC conversions with different pedestal values, then combines the results through correlation processing. This segmentation allows each individual ADC operation to remain simple while the overall system achieves improved linearity through the combination of multiple measurements.
2Measurement precision
If multiple sampling operations are performed to reduce non-linearity errors, then measurement precision is improved, but loss of time increases due to multiple conversion operations
Solution Approach 1:
The patent implements periodic action by performing multiple ADC sampling operations in a systematic sequence within each row readout period. The correlated multi-sampling process uses periodic pedestal values that are intentionally designed to cycle through different levels, allowing the system to collect multiple samples efficiently. This periodic structure enables time-efficient acquisition of the necessary data for error reduction without requiring excessive additional time.
Solution Approach 2:
The system performs preliminary action by pre-calculating and storing multiple pedestal values before the actual image data conversion. These pedestal values are prepared in advance and used systematically during the correlated multi-sampling process. By having the pedestal sequence ready beforehand, the system avoids computational delays during the critical readout phase, thereby reducing the overall time penalty associated with multiple sampling operations.
3Measurement precision
If non-linearity errors are reduced through correlated multi-sampling, then dynamic range is improved, but device complexity increases due to additional sampling and processing circuitry
Solution Approach 1:
The patent introduces an intermediary element in the form of the correlated processing stage that acts as a mediator between the multiple simple ADC operations and the final high-precision result. Rather than requiring a single complex high-precision ADC, the system uses multiple low-precision conversions combined through correlation processing. This intermediary processing step effectively synthesizes the improved dynamic range from simpler building blocks, reducing the complexity burden on individual circuit components.
Data Source
AI summary
A method of implementing Correlated Multi-Sampling (CMS) in an image sensor with improved analog-to-digital converter (ADC) linearity starts with an ADC circuitry included in a readout circuitry that generates a plurality of uncorrelated random numbers used as a plurality of ADC pedestals for sampling image data. A Successive Approximation Register (SAR) included in the ADC circuitry stores a different one of the ADC pedestals before each sampling of the image data. The ADC circuitry samples an image data from a row a plurality of times against plurality of ADC pedestals to obtain a plurality of sampled input data. The ADC circuitry converts each of the plurality of sampled input data from analog to digital, which includes performing a binary search using the SAR. Other embodiments are also described.