Disclosure of Invention
The invention aims to provide an active pixel swing expansion method for a CMOS image sensor, which solves the problem of swing attenuation of an 8T structure of the CMOS image sensor in the prior art in the reading process.
The active pixel swing expansion system for the CMOS image sensor comprises a photosensitive device, wherein one end of the photosensitive device is grounded, the other end of the photosensitive device is sequentially connected with a charge transfer device and a reset device and used for resetting the photosensitive device, the common end of the charge transfer device, which is connected with the reset device, is connected with the input end of a sampling buffer, the output end of the sampling buffer realizes the holding function through a capacitor, the other end of the capacitor is connected with a gating module, the two input ends of the gating module are respectively an external signal A and an external signal B, the output end of the gating module is connected to the inside of a pixel unit, the gating module, the external signal A and the external signal B are outside the pixel unit array, the external signal A is at a low level, and the external signal B is at a high level.
The photosensitive device is a photodiode, the active pixel unit converts photons into electrons through the photodiode, the electrons are converted into voltage signals at the input end of the sampling source follower device after passing through the charge transfer device, and the amplitude of the voltage signals is in direct proportion to the input of the photons.
The sampling phase of the gating module uses a correlated double sampling pattern.
The charge transfer device is an NMOS tube, the reset device is an NMOS tube, the sampling buffer is composed of an NMOS tube 160 and an NMOS tube, the sampling switch is composed of an NMOS tube and an NMOS tube, the energy storage elements are a capacitor A and a capacitor B, wherein the capacitor A is used for keeping reset voltage, and the capacitor B is used for keeping image signal voltage. The lower electrode plates of the capacitor A and the capacitor B are connected to the outside of the pixel unit array and are connected with an external signal through a switch, the output buffer is composed of a source electrode follower, a row selection tube and a tail current which are sequentially connected, and the column line of the tail current is the output signal of the final pixel unit.
An active pixel swing expansion method for a CMOS image sensor adopts an active pixel swing expansion system for the CMOS image sensor, and comprises a photosensitive device, wherein one end of the photosensitive device is grounded, the other end of the photosensitive device is sequentially connected with a charge transfer device and a reset device for resetting the photosensitive device, the common end of the charge transfer device connected with the reset device is connected with the input end of a sampling buffer, the output end of the sampling buffer realizes the holding function through a capacitor, the other end of the capacitor is connected with a gating module, the two input ends of the gating module are respectively an external signal A and an external signal B, the output end of the gating module is connected to the inside of a pixel unit, the gating module, the external signal A and the external signal B are both outside the pixel unit array, the external signal A is low level, the external signal B is high level, and the photosensitive device is a photodiode, the active pixel unit converts photons into electrons through a photodiode, the electrons are converted into voltage signals at the input end of a sampling source electrode follower device after passing through a charge transfer device, and the amplitude of the voltage signals is in direct proportion to the input of the photons; the sampling phase of the gating module uses a related double sampling mode, the charge transfer device is an NMOS tube, the reset device is an NMOS tube, the sampling buffer is composed of an NMOS tube 160 and an NMOS tube, the sampling switch is an NMOS tube and an NMOS tube, the energy storage elements are a capacitor A and a capacitor B, wherein the capacitor A is used for keeping reset voltage, and the capacitor B is used for keeping image signal voltage. The lower electrode plates of the capacitor A and the capacitor B are connected to the outside of the pixel unit array and are connected with an external signal through a switch, the output buffer consists of a source follower, a row selection tube and a tail current which are sequentially connected, and the column line of the tail current is the output signal of the final pixel unit;
the method is implemented according to the following steps:
step 1: after the exposure of the CMOS image sensor is finished, the pixel unit is led into a sampling phase, and an external signal sampling phase switching tube controls a signal to be high, so that a lower electrode plate of a holding capacitor is connected to an external signal;
step 2: sampling a reset voltage, disconnecting a reset device, when a grid control voltage of a sampling buffer jumps from a high level to a low level, disconnecting the sampling buffer, storing the reset voltage on a capacitor, wherein the upper polar plate of the capacitor A is the reset voltage, the lower polar plate is the low level voltage of an external signal, and storing the reset voltage;
and step 3: after the reset voltage is sampled, image signal voltage starts to be collected, the grid control voltage of the charge transfer device jumps from low to high level, so that electrons collected by the photodiode are transferred to the grid of the sampling buffer, when the charges are completely transferred, the charge transfer device is switched off, the sampling switch is switched off, the voltage of the upper polar plate 310 of the holding capacitor B is stored as the image signal voltage, and the image signal voltage is stored;
and 4, step 4: after the image signal voltage is sampled, the pixel unit enters a reading phase, an external signal is set to be a high level matched with an actual circuit, the voltage difference between two ends of a capacitor is kept unchanged, the output voltage of the source electrode follower is improved, and the swing amplitude of the output voltage is improved;
and 5: in the readout phase, the subsequent readout circuit can process the reset voltage. When the NMOS transistor is turned on, the upper plate of the holding capacitor B is connected to the source follower, and at this time, the array output column bus 250 outputs the image signal voltage, and the subsequent readout circuit can process the image signal voltage, so that the output of the reset voltage and the image voltage is completed, and the output swing of the NMOS transistor is improved.
In the sampling process of the active pixel unit, the lower polar plate of the energy storage element is connected to a low level through a gating module outside the area array.
In the process of reading the active pixel unit, the lower polar plate of the energy storage element is connected to a high level through a gating module outside the area array.
The invention has the advantages that the output voltage swing is promoted by configuring the lower plate voltage of the energy storage capacitor outside the pixel unit array, and for the pixel units in the array, no hardware consumption is increased, so that the filling factor and the quantum efficiency of the pixel units are not influenced, but a smart solution is provided for the large well-filled and high sensitivity due to the increase of the swing. By the technology, the output swing of the column line is greatly improved on the premise that the existing pixel unit array and the internal structure are not changed, and the method can be widely applied to the design of a high-dynamic and high-sensitivity CMOS image sensor.
Detailed Description
The present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.
An active pixel swing extension system for a CMOS image sensor comprises a photosensitive device 10, wherein one end of the photosensitive device 10 is grounded, the other end of the photosensitive device 10 is sequentially connected with a charge transfer device 20 and a reset device 30 and used for resetting the photosensitive device 10, the common end of the charge transfer device 20 connected with the reset device 30 is connected with the input end of a sampling buffer 40, the output end of the sampling buffer 40 realizes the holding function through an energy storage element 50, the other end of the energy storage element 50 is connected with a gating module 60, the two input ends of the gating module 60 are respectively an external signal A90 and an external signal B100, the output end of the gating module 60 is connected to the inside of a pixel unit, the gating module 60, the external signal A90 and the external signal B100 are outside the pixel unit array, the external signal A90 is at a low level, and the external signal B100 is at a high level.
The photosensitive device 10 is a photodiode, the active pixel unit converts photons into electrons through the photodiode, the electrons are converted into voltage signals at the input end of the sampling source follower device after passing through the charge transfer device, and the amplitude of the voltage signals is in direct proportion to the input of the photons.
The sampling phase of the gating module 60 uses a correlated double sampling pattern.
The charge transfer device 20 is an NMOS transistor, the reset device is an NMOS transistor, the sampling buffer is composed of an NMOS transistor 160 and an NMOS transistor, the sampling switch is an NMOS transistor and an NMOS transistor, and the energy storage elements are a capacitor a and a capacitor B, where the capacitor a is used to hold a reset voltage and the capacitor B is used to hold an image signal voltage. The lower electrode plates of the capacitor A and the capacitor B are connected to the outside of the pixel unit array and are connected with an external signal through a switch, the output buffer is composed of a source electrode follower, a row selection tube and a tail current which are sequentially connected, and the column line of the tail current is the output signal of the final pixel unit.
An active pixel swing extension method for a CMOS image sensor adopts an active pixel swing extension system for the CMOS image sensor, which comprises a photosensitive device 10, wherein one end of the photosensitive device 10 is grounded, the other end of the photosensitive device 10 is sequentially connected with a charge transfer device 20 and a reset device 30 for resetting the photosensitive device 10, the common end of the charge transfer device 20 connected with the reset device 30 is connected with the input end of a sampling buffer 40, the output end of the sampling buffer 40 realizes the holding function through a capacitor 50, the other end of the capacitor 50 is connected with a gating module 60, the two input ends of the gating module 60 are respectively an external signal A90 and an external signal B100, the output end of the gating module 60 is connected to the inside of a pixel unit, the gating module 60, the external signal A90 and the external signal B100 are all outside the pixel unit array, and the external signal A90 is at a low level, the external signal B100 is high level, the photosensitive device 10 is a photodiode, the active pixel unit converts photons into electrons through the photodiode, the electrons are converted into voltage signals at the input end of the sampling source follower device after passing through the charge transfer device, and the amplitude of the voltage signals is in direct proportion to the input of the photons; the sampling phase of the gating module 60 uses a correlated double sampling mode, the charge transfer device 20 is an NMOS transistor, the reset device is an NMOS transistor, the sampling buffer is composed of an NMOS transistor 160 and an NMOS transistor, the sampling switches are an NMOS transistor and an NMOS transistor, and the energy storage elements are a capacitor a and a capacitor B, wherein the capacitor a is used for holding reset voltage and the capacitor B is used for holding image signal voltage. The lower electrode plates of the capacitor A and the capacitor B are connected to the outside of the pixel unit array and are connected with an external signal through a switch, the output buffer consists of a source follower, a row selection tube and a tail current which are sequentially connected, and the column line of the tail current is the output signal of the final pixel unit;
the method is implemented according to the following steps:
step 1: after the exposure of the CMOS image sensor is finished, the pixel unit is led into a sampling phase, and an external signal sampling phase switching tube controls a signal to be high, so that a lower electrode plate of a holding capacitor is connected to an external signal;
step 2: sampling a reset voltage, disconnecting the reset device 30, when the grid control voltage of the sampling buffer 40 jumps from a high level to a low level, disconnecting the sampling buffer 40, storing the reset voltage on the capacitor 50, wherein the upper plate of the capacitor A50 is the reset voltage, and the lower plate is the low level voltage of an external signal, and storing the reset voltage;
and step 3: after the reset voltage is sampled, image signal voltage starts to be collected, the grid control voltage of the charge transfer device 20 is changed from low to high level, so that electrons collected by the photodiode are transferred to the grid of the sampling buffer 40, when the charges are completely transferred, the charge transfer device 20 is switched off, meanwhile, the sampling switch is switched off, and at the moment, the voltage of the upper plate 310 of the holding capacitor B50 is stored as the image signal voltage, and the image signal voltage is stored;
and 4, step 4: after the image signal voltage is sampled, the pixel unit enters a reading phase, an external signal is set to be a high level matched with an actual circuit, the voltage difference between two ends of a capacitor is kept unchanged, the output voltage of the source electrode follower is improved, and the swing amplitude of the output voltage is improved;
and 5: in the readout phase, the subsequent readout circuit can process the reset voltage. When the NMOS transistor is turned on, the upper plate of the holding capacitor B is connected to the source follower, and at this time, the array output column bus 250 outputs the image signal voltage, and the subsequent readout circuit can process the image signal voltage, so that the output of the reset voltage and the image voltage is completed, and the output swing of the NMOS transistor is improved.
In the process of sampling of the active pixel unit, the lower electrode plate of the energy storage element (capacitor) is connected to a low level through a gating module outside the area array.
In the process of reading the active pixel unit, the lower electrode plate of the energy storage element (capacitor) is connected to a high level through a gating module outside the area array.
The active pixel swing extension method for the CMOS image sensor comprises a photon-to-electron conversion photosensitive device 10, a charge transfer device 20, a reset device 30, a sampling buffer 40, an energy storage element (capacitor) 50, a gating module 60, two input ends 90 and 100 of the gating module, an output buffer 70 and a signal line 80 output by a pixel unit. Wherein one end of the photosensitive device 10 is grounded, the other end is connected to the charge transfer device 20, and the other end of the charge transfer device 20 is connected to the reset device 30, for resetting the photosensitive device 10. Meanwhile, the common terminal of the charge transfer device 20 and the reset device 30 is connected to the input terminal of the sampling buffer 40, and the output terminal of the sampling buffer 40 performs the holding function through the capacitor 50. The other end of the capacitor 50 is connected to the gating module 60, two input ends of the gating module are the external signal 90 and the external signal 100, respectively, wherein one end 100 of the capacitor is inside the pixel unit, the gating module 60, the external signal 90 and the external signal 100 are outside the area array, and the external signal 90 is at a low level, generally a signal ground, and the external signal 100 is at a high level, and needs to be set according to the lowest voltage operation limit of the subsequent output buffer.
The key point of the present invention is that when the sampling buffer 40 samples the reset signal and the image signal, the gating module gates the external signal 90, so that the capacitor 50 can be ensured to perform normal sampling, and the sampling voltage is the difference between the voltages of the upper plate 120 and the lower plate 110 of the capacitor 50; when the output buffer 70 starts to operate, the gating module 60 connects the lower plate 110 of the capacitor 50 to the high level of the external signal 100, and at this time, due to the function of the capacitor 50, the voltage of the upper plate 120 of the capacitor 50 is correspondingly increased, but the difference between the voltages of the upper plate 120 and the lower plate 110 of the capacitor 50 is not changed. After the conversion, the input voltage of the output buffer 70 is raised, so that the input range of the output buffer is raised, and therefore the swing of the output, i.e. the output signal 80 of the pixel unit, is raised.
Fig. 1 is a method for increasing pixel unit swing according to the present invention, fig. 2 is an application example of the method in an 8T pixel unit structure, and fig. 3 is a schematic diagram of an example timing control relationship of an 8T pixel unit.
As shown in fig. 1, in order to implement the active pixel swing extension method for CMOS image sensor disclosed in the present invention, the voltage configuration of the sampling stage and the output stage outside the area array is used to increase the input voltage of the output buffer of the pixel unit, so as to increase the voltage swing of the output column line of the area array.
As shown in fig. 2, an application example of the method in an 8T structure is shown, wherein the photosensitive device is a photodiode 130, the charge transfer device is an NMOS transistor 140, the reset device is an NMOS transistor 150, the sampling buffer is composed of an NMOS transistor 160 and an NMOS transistor 170, the sampling switches are an NMOS transistor 180 and an NMOS transistor 190, the energy storage elements are a capacitor 200 and a capacitor 210, wherein the capacitor 210 is used for holding a reset voltage, and the capacitor 200 is used for holding an image signal voltage. The lower plates 260 of the capacitor 200 and the capacitor 210 are connected to the outside of the pixel unit array, and are connected to an external signal 290 through a switch 270 and an external signal 300 through a switch 280. The output buffer is composed of a source follower (NMOS transistor 220), a row select transistor (NMOS transistor 230), and a tail current 240, wherein the column line 250 is the output signal of the final pixel unit.
The timing control of the specific operation process is shown in fig. 3, for convenience, the waveform numbers in fig. 3 represent the control voltages of the devices with the numbers corresponding to those in fig. 2, and it is assumed that the switch 270 and the switch 280 are at high level, the switch is turned on, the external signal 290 is at low level, and the external signal 300 is at high level.
After the exposure of the CMOS image sensor is finished, the pixel unit enters a sampling phase, and the sampling phase switching tube 270 controls the signal to be high, so that the lower plates of the holding capacitor 200 and the holding capacitor 210 are connected to the external signal 290. Since the 8T pixel cell can implement correlated double sampling inside the pixel cell, it is necessary to sample the reset voltage and the image signal voltage separately at the sampling phase. During the period when the gate voltage of the reset transistor 150 is at the high level, the gate voltage of the NMOS transistor 160 is reset to the high level, and both the sampling transistor 180 and the sampling transistor 190 are in the on state, so that the reset voltage can be transmitted to the holding capacitor 200 and the holding capacitor 210. When the gate control voltage of the sampling tube 190 changes from high level to low level, the sampling tube 190 is turned off, and the reset voltage is stored in the capacitor 210, at this time, the upper plate of the capacitor 210 is the reset voltage, and the lower plate is the low level voltage of the external signal 290. It should be noted here that the reset tube 150 must be disconnected before the sample tube 190 is disconnected, as shown in fig. 3, so that the influence of charge injection and clock feed-through of the reset tube 150 can be suppressed. After the reset voltage is sampled, the image signal voltage starts to be collected, and what is shown in fig. 3 is that the gate control voltage of the charge transfer device 140 changes from low to high, so that the electrons collected by the photodiode 130 are transferred to the gate of the NMOS transistor 160, which is specifically shown in that the gate voltage of the NMOS transistor decreases, and the decreasing amplitude is related to the number of electrons collected by the photodiode 130. After the charge is completely transferred, the charge transfer device 140 is turned off, and the sampling switch 180 is turned off, at this time, the voltage of the upper plate 310 of the holding capacitor 200 holds the image signal voltage, so that the sampling phase finishes sampling and holding the reset voltage and the image signal voltage, the reset voltage is held on the capacitor 210, the image signal voltage is held on the capacitor 200, and at this time, the lower plates of the two capacitors are both connected to the external signal 290.
After the sampling phase is finished, the pixel unit enters a reading phase, the key point at this time is that the switch tube 270 is disconnected, the switch tube 280 is connected, so that the lower electrode plates of the holding capacitor 200 and the holding capacitor 210 are connected to the external signal 300, because the external signal 300 can be configured to be a high level matched with an actual circuit, the voltage difference between two ends of the holding capacitor is kept unchanged, the voltage of the upper electrode plates of the holding capacitor 200 and the holding capacitor 210 can be raised, the gate voltage of the source follower 220 is raised, the output voltage of the source follower is raised, the source follower and the output tail current are ensured to work in a saturation region, and the amplitude of the output voltage is raised. The specific implementation process is shown in figure 3,
in the readout phase, the gate control terminal of the row select transistor 230 goes high, thereby connecting the array output column bus 250 to the pixel cell source follower 220. At this time, the upper plate 320 of the holding capacitor 210 is directly connected to the gate of the source follower 220, so that when the row select transistor 230 is turned on, the array output column bus 250 outputs a reset voltage, and a subsequent readout circuit can process the reset voltage. When the sampling tube 190 is turned on, the upper plate 310 of the holding capacitor 200 is connected to the source follower 220, and at this time, the image signal voltage is output to the array output column bus 250, and the subsequent readout circuit can process the image signal voltage. Therefore, the output of the reset voltage and the image voltage is completed, and the output swing amplitude of the reset voltage and the image voltage is improved.
In summary, the active pixel swing extension method for the CMOS image sensor provided by the present invention realizes the carrying capacity of the full-well charge inside the active pixel, effectively improves the voltage swing of the conventional circuit, and provides a solution for the design of the high dynamic and high sensitive CMOS image sensor.
While the invention has been described in further detail with reference to specific preferred embodiments thereof, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined by the appended claims.