An exponential distribution delay determination method based on light intensity and event threshold control

Abstract

This invention discloses a method for determining exponential distribution delay based on light intensity and event threshold control, comprising the following steps: Based on the fact that DVS image sensor events exhibit exponential distribution characteristics in the time domain when image information changes rapidly in a scene, an exponential distribution model is established that correlates the time constant with light intensity and a threshold; according to the exponential distribution model and the parameters calculated by the model, the exponential distribution delay of the DVS image sensor event to be analyzed in the time domain is determined, which is characterized by the event trigger rate, including the delay magnitude and its distribution in the spatial domain. This invention, based on the exponential distribution delay model, can quickly and conveniently process and calculate the exponential distribution delay of pixels, providing a basic explanation for the origin of motion artifacts in low-light conditions, and assisting researchers in developing artifact suppression methods based on delay mechanisms.

Description

Technical Field

[0001] This invention relates to the field of DVS image sensor design technology, and in particular to an exponential distribution delay determination method based on light intensity and event threshold control. Background Technology

[0002] Of the environmental information perceived by humans, visual information obtained through the human eye accounts for over 80%, and image sensors are electronic devices used to acquire image information. In the "frame"-based visual information processing mode, pixel array information is output for the entire frame regardless of whether the value of a pixel unit changes, resulting in a large amount of data redundancy.

[0003] The Dynamic Vision Sensor (DVS) abandons the concept of "frames," outputting only information about changing pixel units. When the local light intensity change exceeds a preset threshold, the DVS triggers an asynchronous event. This process begins with photon conversion in the photoreceptor, transforming incident light into a logarithmic voltage signal. This signal is then differentially amplified by a switched-capacitor amplifier. When the amplified voltage transient exceeds a threshold, a comparator triggers an ON or OFF event based on the direction of the voltage change. This event-driven encoding readout method effectively reduces data redundancy.

[0004] DVS pixels convert changes in light intensity in a dynamic scene into a difference in logarithmic voltage. This step is crucial for DVS to reflect changes in light intensity through events. However, the conversion process from photocurrent to logarithmic voltage is limited by the bandwidth of the photoreceptor. If the bandwidth of the photoreceptor is too small, the pixel latency will increase significantly. Latency is one of the key performance parameters of DVS pixels, affecting the response speed of the DVS sensor in high-speed scenes.

[0005] Therefore, it is crucial to analyze the distribution characteristics of DVS pixel events in the time domain and the influencing factors of large delays, which can provide a theoretical basis for motion artifact suppression technology. Summary of the Invention

[0006] The purpose of this invention is to overcome the shortcomings and defects of the prior art and to provide an exponential distribution delay determination method based on light intensity and event threshold control.

[0007] This invention is achieved through the following technical solution: A method for determining exponentially distributed delay based on light intensity and event threshold control, comprising the following steps: Based on the fact that DVS image sensor events exhibit an exponential distribution characteristic in the time domain when image information changes rapidly in a scene, an exponential distribution model is established that correlates the time constant with light intensity and threshold. Based on the exponential distribution model and the parameters calculated by the model, the exponential distribution delay of the DVS image sensor events to be analyzed in the time domain is determined, which is characterized by the event trigger rate, including the delay magnitude and its distribution in the spatial domain.

[0008] Preferably, the exponential distribution model includes the response to rapidly changing light intensity in low light conditions and the output voltage of the photoreceptor. V pr The model corresponds to the event trigger rate change, treating the photoreceptor as a single-pole system with the dominant pole as the key. p and photocurrent I ph The voltage of the photoreceptor is directly proportional to the light intensity; when the light intensity decreases, the output voltage of the photoreceptor increases. V pr The slower the response time and the greater the event delay, the more noticeable the motion artifacts become.

[0009] Preferably, the output voltage of the photosensor in low light conditions is [not specified]. V pr The model for the event trigger rate corresponding to the change includes: ; in, k n The subthreshold slope factor of the NMOS transistor. I ph Photocurrent, U T Thermoelectric voltage, C m The output voltage of the photodiode (PD) and the photosensor V pr Parasitic capacitance between nodes, It is the rate of change of light intensity. Indicates time Event triggering rate, It is a constant. K = U T ln / k n , Represents the decay time constant. It represents a logarithm.

[0010] Preferably, the exponential distribution model includes the photoreceptor output voltage under strong light, based on the change in the poles of the photoreceptor caused by light intensity. V pr A model of the event trigger rate corresponding to the change.

[0011] Preferably, under strong light, when the poles of the photoreceptor change based on the light intensity, the output voltage of the photoreceptor...V pr The model for the event trigger rate corresponding to the change includes: ; in, This represents the transconductance of the common-source amplifier driving transistor of the photoreceptor.

[0012] Preferably, under strong light, when the poles of the photoreceptor change based on the light intensity, the change includes the poles of the photoreceptor. p 1. Continue to grow and eventually surpass the extreme point. p 2. Make the poles p 2 becomes the dominant pole, among which, p 1 and p 2 are the two poles of the photoreceptor. p 1 and photocurrent I ph Proportional p 2 and the bias current of the photoreceptor I bias Proportional.

[0013] Preferably, the exponential distribution model includes a distribution model in which the event rate under low light is related to the comparator bias current, based on the fact that the poles of the photoreceptor are proportional to the comparator bias current.

[0014] Preferably, the distribution model based on the relationship between event rate and comparator bias current under low light conditions includes: ; in, The DC gain of the comparator is represented by λ, which is the channel length modulation coefficient and is considered a constant. C L For comparator load capacitor, C 1, C 2 represents the two capacitors in the differential capacitor amplifier. This is the comparator bias current.

[0015] Preferably, the parameters used in the model calculation include at least illumination conditions, process parameters, capacitance parameters, and comparator parameters.

[0016] Preferably, the illumination conditions include the photocurrent corresponding to the initial light intensity. I ph Rate of change of light intensity The process parameters include the subthreshold slope factor of the NMOS transistor. k n Thermoelectricity U T The capacitance parameters include the parasitic capacitance of the PD. C PDPhotodiodes (PDs) and V pr Parasitic capacitance of nodes C m The comparator parameters include the comparator load capacitance. C L Channel length modulation coefficient λ and comparator bias current I event .

[0017] The technology of this invention is based on the fact that DVS events exhibit exponential distribution characteristics in the time domain when image information changes rapidly in a scene. It analyzes the sources and influencing factors of delay, establishes an exponential distribution model that correlates the time constant with light intensity and threshold, and proposes an exponential distribution delay model based on light intensity and event threshold control. Then, based on the exponential distribution delay model, the exponential distribution delay of pixels can be processed and calculated quickly and conveniently, providing a basic explanation for the origin of motion artifacts in low-light conditions and helping researchers develop artifact suppression methods based on delay mechanisms. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of the pixel array structure of an event-based DVS image sensor.

[0019] Figure 2 This is a schematic diagram of the DVS pixel unit structure.

[0020] Figure 3 This is a schematic diagram of the voltage and event response of DVS pixel units under different lighting conditions.

[0021] Figure 4 and Figure 5 The distribution statistics of ON and OFF events in the DVS pixel array in the time domain under different illumination intensities are shown, as well as the relationship between model parameters and illumination intensity.

[0022] Figure 6 It is a statistical graph showing the distribution of OFF events in the DVS pixel array in the time domain under different event thresholds, as well as the relationship between model parameters and the relative rate of change of thresholds. Detailed Implementation

[0023] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0024] Generally, due to bandwidth limitations, DVS pixel photosensitive sensors experience a delay between the actual change in light intensity and the actual triggering time of events when faced with rapidly changing scene light intensity information. Therefore, when the photosensitive sensor bandwidth is small, the reconstructed image from events exhibits severe motion artifacts when dealing with rapidly changing scene information. In low light, the photosensitive sensor bandwidth is controlled by the illumination intensity and is proportional to the photocurrent; therefore, the event delay is even greater in low light, resulting in more severe motion artifacts.

[0025] To investigate the influencing factors of large event delays in DVS (Discrete Visual System), this invention starts with the distribution of events in the time domain under low light conditions. Analysis reveals that event delays exhibit an exponential distribution characteristic in the time domain. By studying the impact of illumination intensity and event thresholds (ON and OFF) on the time constant of this exponential distribution, and based on the characteristics of biomimetic vision imaging, a model is proposed that correlates the time constant with illumination intensity and thresholds—namely, an exponential distribution delay model controlled by illumination intensity and event thresholds. The accuracy and superiority of the model are demonstrated by implementing a correspondence between measured data and the model. These findings provide fundamental insights into the origin of motion artifacts in low-light conditions and establish a theoretical foundation for developing artifact suppression techniques and improving DVS performance in dark environments.

[0026] In an exemplary embodiment of this application, the exponential distribution delay determination method based on light intensity and event threshold control includes the following steps: Based on the fact that DVS image sensor events exhibit an exponential distribution characteristic in the time domain when image information changes rapidly in a scene, an exponential distribution model is established that correlates the time constant with light intensity and threshold. Based on the exponential distribution model and the parameters calculated by the model, the exponential distribution delay of the DVS image sensor events to be analyzed in the time domain is determined, which is characterized by the event trigger rate, including the delay magnitude and its distribution in the spatial domain.

[0027] The following analysis, combining event-based pixel array structures and pixel structures, focuses on the exponential distribution delay model based on light intensity and event threshold control of this invention, analyzing the principle of event delay generation, the time domain performance of the delay, and the influencing factors of large event delays.

[0028] Figure 1A schematic diagram of an event-based DVS pixel array structure is provided. This event-based DVS pixel array structure consists of pixel array 1, DVS pixel units 2, high-speed interface circuit 3, ESP pipeline interface 4, readout control circuit 5, row selector 6, and Y-direction address encoder 7. For example, pixel array 1 is a 1280*720 array. Pixel array 1 senses changes in light intensity and converts these changes into events, forming a sparsely encoded output. After a single pixel unit generates an event, it transmits the Y-direction address to the Y-direction address encoder 7 via a request signal, and then transmits the signal to the row selector 6. After the response signal returns to the row where the pixel unit is located, the pixel unit in that row will output event information. The information is read out as parallel data through high-speed interface circuit 3, and the parallel data is read out serially through Event Signal Processing (ESP) pipeline interface 4. The timing signals during the readout process are implemented by readout control circuit 5. However, when the light intensity is low, there is a significant delay between the event triggering time of the pixel array and the actual light intensity change time. This delay can cause motion artifacts during image reconstruction.

[0029] Figure 2 A schematic diagram of the event-based pixel structure is given. The pixel structure consists of photodiode 21, logarithmic diode 22, bias transistor 23, capacitor C124, transistor 25, transistor 26, transistor 28, transistor 29, transistor 30, transistor 31, capacitor C232, transistor 33, transistor 34, and the parasitic capacitance of the PD. C PD 35. Capacitor C m It consists of 36.

[0030] Among them, photodiode 21, logarithmic diode 22, bias transistor 23, transistor 33, and parasitic capacitance of PD. C PD Capacitor C m 36 forms a photosensitive sensor, and capacitor C124, transistor 25, transistor 31, capacitor C232, and transistor 33 form a differential capacitor amplifier. Transistors 26, 28, 29, and 30 form a comparator.

[0031] The specific connection is as follows: one end of photodiode 21 is grounded, and the other end is connected to the parasitic capacitance of PD. C PD One end, the source of logarithmic transistor 22, the gate of transistor 34, and capacitor C m 36 at one end; parasitic capacitance C PD The other end is grounded, and the source of transistor 34 is grounded; the drain of logarithmic transistor 22 is connected to VDD, and its gate (output) is connected to VDD. Vpr Connect the drain of bias transistor 23 and capacitor C m The other end of transistor 36 is connected to the drain of transistor 34, one end of capacitor C124 in the differential capacitor amplifier, the other end of capacitor C124 is connected to the source of transistor 33, one end of capacitor C232, and the gate of transistor 25. The source of transistor 25 is connected to VDD, and its drain is connected to the drain of transistor 31, the other end of capacitor C232, and the drain of transistor 33. The gate of transistor 33 (reset transistor) is connected to the reset signal nrst. The drain of transistor 25 is connected to the gates of transistors 26 and 28. The sources of transistors 26 and 28 are connected to VDD. The drain of transistor 26 is connected to the drain of transistor 30. The drain of transistor 28 is connected to the drain of transistor 29. The sources of transistors 29, 30, and 31 are grounded, and their gates are connected to the voltage signal V. off V on V ref These are the OFF event voltage, ON event voltage, and reference voltage signal, respectively. The photodiode 21 converts the light intensity signal into a photocurrent. I ph The photocurrent is converted by the logarithmic diode in the logarithmic diode 22. I ph The voltage is converted to a logarithmic value and connected to a common-source stage composed of transistors 23 and 34 for amplification. The differential capacitor amplifier consists of a common-source stage composed of transistors 25 and 31, and capacitors 24 and 32. It samples and amplifies the input signal, i.e., the photovoltage, and its amplification factor is determined by capacitors 24 and 32. The comparator (threshold comparator) also uses a common-source stage, where the ON comparator is a common-source stage composed of transistors 26 and 30, and the OFF comparator is a common-source stage composed of transistors 28 and 29. Since the op-amps in the differential capacitor amplifier and the comparator are both common-source stages, the voltage of the ON and OFF events can be adjusted. V on and V off This controls the threshold values ​​for ON and OFF events. However, this adjustment is accompanied by comparator bias current. I on and I off Changes in [the event] can also cause differences in event delays.

[0032] Figure 3 The diagrams illustrating the voltage and event responses of DVS pixel units under different light intensities are presented. This is due to the photocurrent... I ph and the bias current in transistor 23 I bias The resulting bandwidth limitationV pr It cannot respond to changes in light intensity in real time. The specific model is shown in equations (1) and (2) below.

[0033] (1) (2) in, I ph Photocurrent, k n The subthreshold slope factor of the NMOS transistor. It is the relative rate of change of light intensity. U T Thermoelectric voltage, C m For photodiodes (PD) and V pr Parasitic capacitance between nodes, C PD Parasitic capacitance on the PD g m3 For the transconductance of transistor 34, p 1 and p 2 are the two poles of the photoreceptor. p 1 and photocurrent I ph Proportional p 2 and bias current I bias It is directly proportional. Therefore V pr The bandwidth-related settling time causes a delay between illumination switching and event triggering.

[0034] As the pole size decreases, the bandwidth decreases, and the delay becomes more pronounced. Under low-light conditions, the photoreceptor is simplified to a single-pole system, with the dominant pole... p and photocurrent I ph Proportional to the change in light intensity at this time V pr The impact of nodes is more pronounced. When illumination decreases, latency increases significantly. V pr The slower the response speed and the greater the event delay, the more noticeable the motion artifacts become. In low light conditions, facing rapidly changing light intensity... V pr The model for the event trigger rate corresponding to the change is shown in equation (3) below.

[0035] (3) This delay is constructed as an exponential distribution model, where the decay rate of the model parameter event rate in low light is 1 / tAs light intensity changes linearly K It is a constant.

[0036] When the light intensity is high, the pole p 1. Continue to grow and eventually surpass p 2, make p 2 becomes the dominant extreme. V pr The model for the event trigger rate corresponding to the change is shown in equation (4) below.

[0037] (4) At this point, although the event output rate still follows an exponential distribution, the decay rate depends only on the circuit architecture and is independent of the light intensity.

[0038] Figure 4 and Figure 5 The distribution statistics of ON and OFF events in the DVS pixel array in the time domain under different light intensities are presented, with the light intensity change rate being ±50%. The points in the figure represent actual measured data, and the curves represent model-fitted data. The subplot shows the decay rate 1 / ... of the event rate in the actual measured data. t The statistical relationship between amplitude and light intensity was determined through the model. It was found that under low-light conditions, the poles are directly proportional to the light intensity; therefore, the triggering event was modeled as being related to light intensity and photocurrent. I ph Related exponential distribution characteristics. The decay rate of the event rate 1 / t The amplitude changes synchronously with the light intensity, and both are linearly related to the light intensity.

[0039] Therefore, as illumination levels increase, event response speed gradually accelerates. Under these conditions, event triggering becomes more concentrated, and latency decreases. By fitting the measured data to the exponential distribution model, the coefficients of determination (COD) of all fitted curves exceed 95%.

[0040] However, under high illumination conditions, the relationship between the principal pole and illumination intensity weakens and tends to be constant. Although the measured data still conforms to the exponential distribution model, the decay rate of the event rate 1 / t The amplitude gradually deviates from the model that is positively correlated with light intensity and tends to be constant. That is, under different initial specular conditions, the temporal distribution of the event rate remains constant.

[0041] Figure 6 The distribution statistics of OFF events in the DVS pixel array in the time domain under different event thresholds are presented, with the illumination intensity varying rapidly from 0.02 lux to 0.01 lux. The points in the figure represent actual measured data, and the curves represent model-fitted data. The subplots show the decay time constant. t、 The statistical relationship between the reciprocal of amplitude and light intensity. Changes in the threshold value will cause a bias current in the ON comparator. I on and OFF comparator bias current I off The change in threshold and bias current I event ( I on and I off The general term for (the general term) shows an exponential relationship. The specific model is shown in equation (5) below.

[0042] (5) in, V ev It is the offset voltage of the differential capacitor amplifier. I ref It is the bias current of the differential capacitor amplifier. This is the subthreshold slope factor of the PMOS transistor. Both the ON and OFF comparators are common-source amplifiers, and both can be considered as single-pole systems. Since transistors 29 and 30 operate in the subthreshold region, the pole and comparator bias currents... I event Proportional. Under this condition, the event rate in low light is constructed as a function of the comparator bias current. I event Related exponential distribution model, decay time constant t It changes synchronously with the reciprocal of the amplitude. The specific model is shown in equation (6) below.

[0043] (6) in, The DC gain of the comparator is represented by λ, which is the channel length modulation coefficient and is considered a constant. C L This is the comparator load capacitor. Therefore, even excluding the influence of light intensity, the event trigger delay can still be constructed as an exponential distribution model related to the comparator bias current. The model parameters include a scale parameter and a decay time constant. t Both exhibit exponential changes with the relative rate of change of the threshold. Fitting the measured data to the exponential distribution model, the COD of all fitted curves exceeded 95%. As the OFF event threshold increases, t The event latency increases exponentially, leading to longer event delays. Conversely, as the ON event threshold increases, the event latency decreases. Therefore, under the same conditions, the OFF event always triggers more slowly than the ON event.

[0044] Therefore, through the exponential distribution delay model of this application, the specific magnitude and spatial distribution of this exponential distribution delay controlled by light intensity and event threshold can be clearly calculated.

[0045] The following is an example of calculating the delay of the exponential distribution.

[0046] Assuming the test is conducted under uniform lighting conditions, the specific conditions are as follows: (1) Illumination conditions: The initial light intensity is 0.04 lux, and the corresponding photocurrent is... I ph =11.04fA, rate of change of light intensity Both are 50%.

[0047] (2) Process parameters: Using a typical 110nm CMOS process, the subthreshold slope factor of the NMOS transistor is... k n =0.8, thermal voltage U T =26mV, (3) Capacitance parameters: parasitic capacitance of PD C PD =100fF, photodiode (PD) and V pr Parasitic capacitance between nodes C m =20fF.

[0048] (4) Comparator parameters: comparator load capacitance C L =1pF, channel length modulation coefficient λ=0.1V -1 Comparator bias current I event =100nA.

[0049] Under low light conditions, the photoreceptor can be simplified to a single-pole system, and the delay is mainly determined by the photocurrent. I ph The time constant of the event trigger rate decay is determined according to equation (3). t =2.36×10 6 us, which is basically consistent with the test results. At the same time, due to the presence of the comparator, the delay further increases. According to equation (3), in this case, after adding the comparator, the time constant of the event trigger rate decay of the increased delay is... t =100us. The calculation result is consistent with the test data, which is sufficient to prove the reliability of the distribution model in this application.

[0050] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. It will be apparent to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and that the present invention can be implemented in other specific forms without departing from the spirit or basic features of the present invention. Therefore, the embodiments should be regarded as exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, it is intended that all variations falling within the meaning and scope of the equivalents of the claims be included within the invention.

[0051] Furthermore, it should be understood that although this specification describes embodiments, not every embodiment contains only one independent technical solution. This narrative style is merely for clarity. Those skilled in the art should consider the specification as a whole, and the technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.

Claims

1. A method for determining exponentially distributed delay based on light intensity and event threshold control, characterized in that, Includes the following steps: Based on the fact that DVS image sensor events exhibit an exponential distribution characteristic in the time domain when image information changes rapidly in a scene, an exponential distribution model is established that correlates the time constant with light intensity and threshold. Based on the exponential distribution model and the parameters calculated by the model, the exponential distribution delay of the DVS image sensor events to be analyzed in the time domain is determined, which is characterized by the event trigger rate, including the delay magnitude and its distribution in the spatial domain.

2. The method for determining exponentially distributed delay based on light intensity and event threshold control according to claim 1, characterized in that, The exponential distribution model includes the response to rapidly changing light intensity in low light conditions and the output voltage of the photoreceptor. V pr The model corresponds to the event trigger rate change, treating the photoreceptor as a single-pole system with the dominant pole as the key. p and photocurrent I ph The voltage of the photoreceptor is directly proportional to the light intensity; when the light intensity decreases, the output voltage of the photoreceptor increases. V pr The slower the response time and the greater the event delay, the more noticeable the motion artifacts become.

3. The method for determining exponentially distributed delay based on light intensity and event threshold control according to claim 2, characterized in that, The output voltage of the photosensor in response to rapidly changing light intensity under dim light conditions. V pr The model for the event trigger rate corresponding to the change includes: ； in, κ n The subthreshold slope factor of the NMOS transistor. I ph Photocurrent, U T Thermoelectric voltage, C m The output voltage of the photodiode (PD) and the photosensor V pr Parasitic capacitance between nodes, It is the rate of change of light intensity. Indicates time Event triggering rate, It is a constant. K = U T ln / κ n , Represents the decay time constant. It represents a logarithm.

4. The method for determining exponentially distributed delay based on light intensity and event threshold control according to claim 3, characterized in that, The exponential distribution model includes the changes in the poles of the photoreceptor based on the light intensity under strong light. V pr A model of the event trigger rate corresponding to the change.

5. The method for determining exponentially distributed delay based on light intensity and event threshold control according to claim 4, characterized in that, Under strong light, when the light intensity causes a change in the poles of the photoreceptor, the output voltage of the photoreceptor changes. V pr The model for the event trigger rate corresponding to the change includes: ； in, This represents the transconductance of the common-source amplifier driving transistor of the photoreceptor.

6. The method for determining exponentially distributed delay based on light intensity and event threshold control according to claim 4, characterized in that, Under strong light, when the poles of the photoreceptor change based on the light intensity, the poles of the photoreceptor are included. p 1. Continue to grow and eventually surpass the extreme point. p 2. Make the poles p 2 becomes the dominant pole, among which, p 1 and p 2 are the two poles of the photoreceptor. p 1 and photocurrent I ph Proportional p 2 and the bias current of the photoreceptor I bias Proportional.

7. The method for determining exponentially distributed delay based on light intensity and event threshold control according to claim 6, characterized in that, The exponential distribution model includes a distribution model in which the event rate under low light is related to the comparator bias current, based on the fact that the poles of the photoreceptor are proportional to the comparator bias current.

8. The method for determining exponentially distributed delay based on light intensity and event threshold control according to claim 7, characterized in that, The distribution model based on the relationship between event rate and comparator bias current under low light conditions includes: ； in, The DC gain of the comparator is represented by λ, which is the channel length modulation coefficient and is considered a constant. C L For comparator load capacitor, C 1, C 2 represents the two capacitors in the differential capacitor amplifier. This is the comparator bias current.

9. The method for determining exponentially distributed delay based on light intensity and event threshold control according to claim 1, characterized in that, The parameters used in the model calculation include at least illumination conditions, process parameters, capacitance parameters, and comparator parameters.

10. The method for determining exponentially distributed delay based on light intensity and event threshold control according to claim 9, characterized in that, The illumination conditions include the photocurrent corresponding to the initial light intensity. I ph Rate of change of light intensity The process parameters include the subthreshold slope factor of the NMOS transistor. κ n Thermoelectricity U T The capacitance parameters include the parasitic capacitance of the PD. C PD Photodiodes (PDs) and V pr Parasitic capacitance of nodes C m The comparator parameters include the comparator load capacitance. C L Channel length modulation coefficient λ and comparator bias current I event .

Images