Column AD Converter Bit Allocation for Image Sensor SNR and Frame Rate
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Image sensors for security and vehicle cameras face challenges in improving signal-to-noise ratio and frame rate without increasing the number of bits in the analog-to-digital converter, as higher bit widths lead to complex circuits and cost increases, while oversampling conversion is slow and limits frame rate.
Innovation Solution
The image sensor employs a column analog-to-digital converter configuration with an oversampling AD converter and a recursive AD converter, dynamically allocating bits between them to maintain a total number of bits, allowing for improved signal-to-noise ratio and frame rate without unnecessary bit expansion.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the number of bits of the AD converter is increased to improve measurement precision, then the signal-to-noise ratio is improved, but the device complexity and manufacturing cost increase
Solution Approach 1:
The patent divides the AD conversion process into two separate converters: an oversampling AD converter that handles upper bits and a recursive AD converter that handles lower bits. This segmentation allows each converter to be optimized for its specific function, achieving high measurement precision without requiring a single complex high-bit converter, thus reducing overall device complexity
Solution Approach 2:
The patent combines the outputs of two different AD converters (oversampling and recursive types) to produce the final digital signal. By merging the strengths of both converter types, the system achieves high signal-to-noise ratio equivalent to a high-bit converter while maintaining the simplicity of using two lower-bit converters
2Measurement precision
If oversampling conversion is used to improve measurement precision, then the signal-to-noise ratio is improved, but the conversion speed decreases and frame rate is limited
Solution Approach 1:
The patent segments the bit processing into two paths: upper bits processed by oversampling converter (for noise reduction) and lower bits processed by recursive converter (for high speed). This segmentation allows the system to achieve both high signal-to-noise ratio and high frame rate simultaneously by optimizing each path for its specific requirement
Solution Approach 2:
Different parts of the digital signal (upper bits vs lower bits) are processed with different qualities and methods. Upper bits undergo slow but high-precision oversampling conversion, while lower bits undergo fast recursive conversion. This local quality differentiation resolves the contradiction between precision and speed
3Measurement precision
If the number of bits of the AD converter is increased to improve measurement precision, then the signal-to-noise ratio is improved, but the manufacturing cost increases
Solution Approach 1:
The patent segments the high-bit conversion task into two lower-bit converter circuits. By using two simpler converter circuits instead of one complex high-bit converter, the manufacturing cost is reduced while maintaining the equivalent measurement precision through the combination of both converters' outputs
Data Source
AI summary
Each of a plurality of analog-to-digital (AD) converters configuring a column AD converter used in an image sensor comprises: an oversampling AD converter that receives an output voltage of a pixel unit; a recursive AD converter that receives an analog residual signal of the oversampling AD converter; and a counter that adds a digital signal output from the oversampling AD converter and a digital signal output from the recursive AD converter. The controller dynamically allocates the number of bits of the oversampling AD converter and the number of bits of the recursive AD converter, while maintaining the total number of bits of the oversampling AD converter and the recursive AD converter.