CMOS Image Sensor Pixel RC Circuit for Direct Time-of-Flight Noise Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
CMOS image sensors for direct time-of-flight measurement face challenges in achieving high dynamic range and low noise levels, particularly in detecting weak signals and distinguishing close pulses, due to limitations in pixel structure and operation, including saturation and thermal noise issues.
Innovation Solution
Implementing a nonlinear resistor within the pixel combined with a low pass filter operation and high frequency integration function in the RC circuit, which separates signal and noise frequencies, allowing for efficient detection of pulse locations through filtering and improving the signal-to-noise ratio.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional CMOS pixel operation with separate integration and readout phases is used, then the pixel structure is simple and easy to manufacture, but the sensor cannot achieve direct time-of-flight measurement and has poor ability to detect weak signals and distinguish close pulses
Solution Approach 1:
The patent implements dynamic phase shifting of the reset signal to enable direct time-of-flight measurement. By dynamically adjusting the phase of the reset signal relative to the integration signal, the pixel can measure time-of-flight directly while maintaining a relatively simple CMOS structure. This dynamic approach allows the pixel to operate in a mode that provides both measurement precision and structural simplicity.
Solution Approach 2:
The patent changes the operational parameters of the CMOS pixel by introducing phase-shifted reset signals and modifying the timing sequences. By changing the phase relationship between reset and integration signals, the pixel achieves direct time-of-flight measurement capability without requiring complete structural redesign, thus balancing measurement precision with device complexity.
2Object-affected harmful factors
If high capacitance values are used at the sense node, then thermal noise is reduced, but the ability to distinguish close pulses and detect weak signals deteriorates due to saturation and reduced dynamic range
Solution Approach 1:
The patent employs periodic reset signaling with phase shifting to manage the sense node capacitance dynamically. By using periodic reset signals with varying phases, the system can clear accumulated charge at optimal intervals, preventing saturation while maintaining low thermal noise. This periodic action allows the pixel to handle both weak signals and close pulse discrimination effectively.
Solution Approach 2:
The patent ensures continuous integration of photocharges while maintaining dynamic reset control. The continuous operation mode, combined with phase-shifted reset signals, allows the pixel to continuously accumulate charges from weak signals without saturation, while the dynamic reset ensures thermal noise remains low. This continuity enables both weak signal detection and close pulse distinction.
3Measurement precision
If the pixel operates in direct time-of-flight mode with continuous readout, then pulse location detection is improved, but thermal noise and readout noise increase due to continuous electronic activity
Solution Approach 1:
The patent applies preliminary anti-action by using correlated double sampling (CDS) to cancel thermal noise before readout. The CDS technique subtracts a reference signal (taken at a different phase) from the actual signal, thereby eliminating thermal noise components that would otherwise interfere with pulse location detection. This preliminary noise cancellation enables high precision detection without being overwhelmed by readout noise.
Solution Approach 2:
The patent implements feedback through the phase-shifted reset mechanism that uses information about the integration phase to optimally time the reset operation. This feedback approach ensures that reset operations occur at phases that minimize noise impact while maximizing signal capture, thereby improving pulse location precision without proportionally increasing readout noise.
4Measurement precision
If multiple transistors are added to each pixel for direct TOF measurement functionality, then measurement capability is improved, but fill factor and quantum efficiency decrease due to increased pixel area occupation
Solution Approach 1:
The patent makes existing CMOS pixel components multi-functional to achieve direct time-of-flight measurement without adding significant area. The reset transistor, integration capacitor, and readout circuitry are made to serve dual purposes: conventional image sensing and direct TOF measurement. This universality allows the pixel to perform both functions using the same physical structures, thereby improving measurement capability without increasing pixel area occupation.
Solution Approach 2:
The patent uses dynamic signal processing and phase-shifting techniques to enable direct TOF measurement using existing pixel infrastructure. By dynamically controlling the phase and timing of reset and integration signals, the pixel achieves TOF measurement functionality without requiring additional dedicated hardware components, thus maintaining high fill factor and quantum efficiency while improving measurement precision.
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
This approach enables effective detection of weak signals and reduces thermal noise, enhancing the signal-to-noise ratio and allowing for accurate reconstruction of pulse shapes, even with low capacitance values, thereby improving the overall performance of CMOS pixels for direct time-of-flight measurements.
Implementation Method 1
the nonlinear resistor combining with a capacitance at the sense node form a RC circuit that has a low pass filtering function and high frequency integration function
Implementation Method 2
the nonlinear resistor combining with a capacitance at the sense node form a RC circuit that has a low pass filtering function and high frequency integration function
Implementation Method 3
a photodetector operating as a current source
Implementation Method 4
the transfer transistor is the ON state all along the measurement phase, the transfer transistor then operating as a decoupling element between a photodetector node and the sense node
Implementation Method 5
filtering means configured to apply one of a band pass or high pass filter, before or after analog to digital conversion, having the effect of increasing the signal to noise ratio
Implementation Method 6
filtering means configured to apply one of a band pass or high pass filter, before or after analog to digital conversion, having the effect of increasing the signal to noise ratio
Data Source
AI summary
A direct TOF optic sensor is based on CMOS pixels, wherein a pixel structure comprises a photodetector PhD, a non linear resistance R and a transfer MOS transistor in series, and delivers an output signal at a sensing node SN between the resistor and the transfer transistors. The photogenerated current is continuously drained into the nonlinear resistance and converted to a voltage signal by the RC circuit formed by the nonlinear resistance and a capacitance at the sense node SN. The voltage signal is continuously transmitted to a readout circuitry 300 having a fast analog to digital converter. The RC circuit within the pixel structure has a low pass filtering function and a high frequency integrating function, so that noise, in particular thermal noise due to the nonlinear resistance is mainly shifted in a low frequency range, separate from a high frequency range of the main signal component corresponding to a pulse light signal received at the photodetector. The main signal component is recovered by means of one of a band pass or high pass filter F implemented in the readout circuitry, that increases the signal to noise ratio in the high frequency range.