Delta-Sigma ADC Gain Control for Low-Light Image Precision
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional CMOS image sensors using ΔΣ ADCs face issues with applying gain during AD conversion, leading to reduced bit precision and increased noise, particularly in low-luminance areas, due to the need for digital gain post-conversion, which amplifies noise.
Innovation Solution
Analog-to-digital converters employing ΔΣ modulation with a control unit that adjusts the feedback amount based on the applied gain, using a smaller feedback amount for higher gains and controlling the feedback loop to stabilize the conversion process, thereby applying gain during AD conversion without reducing bit precision or amplifying noise.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Power
If digital gain is applied after AD conversion in a ΔΣ ADC, then the signal level is increased, but bit precision is reduced and noise is amplified
Solution Approach 1:
The patent applies gain during the AD conversion process itself rather than after conversion. The gain is applied to the analog input signal before the ΔΣ modulation, allowing the amplified signal to be converted to digital form while maintaining full bit precision. This preliminary action prevents the precision loss that would occur if digital gain were applied afterward.
Solution Approach 2:
The patent implements a feedback mechanism where the output of the ΔΣ ADC is fed back to the input through a feedback path. This feedback allows the system to maintain stability while applying gain during conversion, enabling the system to achieve both signal amplification and noise reduction by controlling the feedback amount based on the applied gain.
2Power
If digital gain is applied after AD conversion in a ΔΣ ADC, then the signal is amplified, but noise becomes more noticeable particularly in low-luminance areas
Solution Approach 1:
By applying gain during the analog-to-digital conversion process rather than after, the patent amplifies the signal before quantization occurs. This allows the amplified signal to utilize the full dynamic range of the ADC, thereby improving signal-to-noise ratio and making noise less noticeable, particularly in low-luminance areas where noise would otherwise be more apparent.
Solution Approach 2:
The feedback mechanism controls the amount of gain applied during conversion based on the specific signal conditions. By adjusting the feedback amount according to the applied gain, the system can amplify signals while suppressing noise, particularly in low-luminance conditions where noise would otherwise be amplified along with the signal.
3Stability of the object's composition
If a large feedback amount is used in ΔΣ ADC, then conversion stability is improved, but gain application capability is reduced
Solution Approach 1:
The patent makes the feedback amount dynamic rather than fixed. The feedback amount is adjusted based on the gain that needs to be applied during conversion. This dynamic adjustment allows the system to maintain stability when needed while also being able to apply gain when required, resolving the contradiction between stability and gain capability.
Solution Approach 2:
The patent changes the feedback parameter (feedback amount) based on the operating conditions and required gain. By varying the feedback amount as a controllable parameter, the system can optimize both stability and gain application capability depending on the specific conversion requirements, rather than being constrained by a fixed feedback value.
Data Source
AI summary
An analog-to-digital converter comprises: an analog-to-digital conversion unit that performs analog-to-digital conversion using ΔΣ modulation on an image signal output from pixels; a judgement unit that judges a gain to be applied to the image signal; and a control unit that controls a feedback amount used in the analog-to-digital conversion of the image signal by the analog-to-digital conversion unit based on the gain. The control unit controls the feedback amount smaller in a case where the gain is a first gain than in a case where the gain is a second gain smaller than the first gain.