Abstract:
The circuit includes a first transistor; a second transistor whose first terminal is connected to a gate of the first transistor for setting the potential of the gate of the first transistor to a level at which the first transistor is turned on; a third transistor for setting the potential of a gate of the second transistor to a level at which the second transistor is turned on and bringing the gate of the second transistor into a floating state; and a fourth transistor for setting the potential of the gate of the second transistor to a level at which the second transistor is turned off. With such a configuration, a potential difference between the gate and a source of the second transistor can be kept at a level higher than the threshold voltage of the second transistor, so that operation speed can be improved.
Abstract:
Systems formed by a multi-bit three-transistor (3T) memory cell (i.e., dynamic-analog RAM) are provided. The 3T memory cell includes: a read-access transistor M1 in electrical communication with a read bitline; a switch transistor M2 in electrical communication with the read-access transistor M1; a write-access transistor M3 in electrical communication with the read-access transistor M1 and a write bitline; and a memory node MEM in electrical communication between the read-access transistor M1 and the write-access transistor M3, wherein the memory node MEM is configured to store a 4-bit weight WE. An array of the 3T memory cells (i.e., dynamic-analog RAMs) may form a computing-in-memory (CIM) macro, and further form a convolutional neural network (CNN) accelerator by communicating with an application-specific integrated circuit (ASIC) which communicates with a global weight static random access memory and an activation static random access memory.
Abstract:
A shift register unit, a display panel including the shift register unit and a display device including the display panel are provided. The shift register unit includes a driving module, an output module, a first transistor, and a second transistor. By connecting a second electrode of the first transistor in the shift register unit with an output terminal of the shift register unit, even if a channel width of the second transistor is considerably smaller than a theoretical design value, abnormal output of the shift register unit can be avoided.
Abstract:
A low density One-Time Programmable (OTP) memory is disclosed to achieve low gate count and low overhead in the peripheral circuits to save the cost. A maximum-length Linear Feedback Shift Register (LFSR) can be used to generate 2n−1 address spaces from an n-bit address. The registers used in the address generator can have two latches. Each latch has two cross-coupled inverters with two outputs coupled to the drains of two MOS input devices, respectively. The inputs of the latch are coupled to the gates of the MOS input devices, respectively. The sources of the MOS input devices are coupled to the drains of at least one MOS device(s), whose gate(s) are coupled to a clock signal and whose source(s) are coupled to a supply voltage. The two latches can be constructed in serial with the outputs of the first latch coupled to the inputs of the second latch.
Abstract:
A gate drive circuit includes a plurality of stages connected one after another to each other. Each of the stages includes a charging section, a driving section, a discharging section, a holding section and a holding control section. The driving section pulls up a high level of a first clock signal to output a gate signal. The discharging section discharges a voltage potential of a first node to a first off-voltage. The holding section holds a voltage potential of the first node to the first off-voltage. The holding control section receives the first clock signal and a second clock signal. The holding control section holds a voltage potential of the holding section to a second off-voltage through a second node in accordance with the second clock signal to prevent floating of the holding section.
Abstract:
A sample-and-hold device provides output charge pairs which represent samples of a continuous-time differential input voltage. The device uses charge-coupled device elements in a symmetrical structure for splitting a constant input charge into a signal-dependent output charge pair. It is capable of operation at higher speed and with higher dynamic range than similar prior-art devices.
Abstract:
An image sensor operable to vary the output spatial resolution according to a received light level while maintaining a desired signal-to-noise ratio. Signals from neighboring pixels in a pixel patch with an adjustable size are added to increase both the image brightness and signal-to-noise ratio. One embodiment comprises a sensor array for receiving input signals, a frame memory array for temporarily storing a full frame, and an array of self-calibration column integrators for uniform column-parallel signal summation. The column integrators are capable of substantially canceling fixed pattern noise.
Abstract:
An analog delay circuit which includes an analog memory circuit wherein a plurality of memory cells each including a memory capacitor and a selection switch for the memory capacitor are arranged in a matrix includes row switches provided for the individual columns for individually being driven by row selection signals. A same clock signal from a clock generation circuit is supplied commonly to an X direction scanning circuit and a Y direction scanning circuit. The number of stages of registers of the X direction scanning circuit and the number of stages of registers of the Y direction scanning circuit are set so that they have no common divisor other than 1. Consequently, when the memory cells are to be selectively scanned, a same selection condition can be provided to all of the memory cells without relying upon the positions of the memory cells, and the parasitic capacitance connected to a signal write/read terminal is reduced.
Abstract:
An image sensor operable to vary the output spatial resolution according to a received light level while maintaining a desired signal-to-noise ratio. Signals from neighboring pixels in a pixel patch with an adjustable size are added to increase both the image brightness and signal-to-noise ratio. One embodiment comprises a sensor array for receiving input signals, a frame memory array for temporarily storing a full frame, and an array of self-calibration column integrators for uniform column-parallel signal summation. The column integrators are capable of substantially canceling fixed pattern noise.