Abstract: The present disclosure relates to an integrated chip. The integrated chip includes a polysilicon layer arranged on an upper surface of a base substrate. A dielectric layer is arranged over the polysilicon layer, and an active semiconductor layer is arranged over the dielectric layer. A semiconductor material is arranged vertically on the upper surface of the base substrate and laterally beside the active semiconductor layer.

Abstract: A differential amplifier is provided. The differential amplifier includes a first load, a second load, a current source, a differential pair circuit, a first and a second switch circuit. The differential pair circuit includes a first transistor, a second transistor, a third transistor, and a fourth transistor. The first switch circuit controls the first and the second transistors, and the second switch circuit controls the third and the fourth transistors. Through the control and selection of the first and second switch circuits, a differential pair is selected in the differential pair circuit to receive and process a first input signal and a second input signal for signal.

Abstract: A memory-adaptive processing method for a convolutional neural network includes a feature map counting step, a size relation counting step and a convolution calculating step. The feature map counting step is for counting a number of a plurality of input channels of a plurality of input feature maps, an input feature map tile size, a number of a plurality of output channels of a plurality of output feature maps and an output feature map tile size for a convolutional layer operation. The size relation counting step is for obtaining a cache free space size in a feature map cache and counting a size relation. The convolution calculating step is for performing the convolutional layer operation with the input feature maps to produce the output feature maps according to a memory-adaptive processing technique, and the memory-adaptive processing technique includes a dividing step and an output-group-first processing step.

Abstract: A differential amplifier is provided. The differential amplifier includes a first load, a second load, a current source, a differential pair circuit, a first and a second switch circuit. The differential pair circuit includes a first transistor, a second transistor, a third transistor, and a fourth transistor. The first switch circuit controls the first and the second transistors, and the second switch circuit controls the third and the fourth transistors. Through the control and selection of the first and second switch circuits, a differential pair is selected in the differential pair circuit to receive and process a first input signal and a second input signal for signal.

Abstract: A harmonic densely connecting method includes an input step, a plurality of layer operation steps and an output step. The input step is for storing an original input tensor of the block into a memory. Each of the layer operation steps includes a layer-input tensor concatenating step and a convolution operation step. The layer-input tensor concatenating step is for selecting at least one layer-input element tensor of a layer-input set from the memory according to an input connection rule. When a number of the at least one layer-input element tensor is greater than 1, concatenating all of the layer-input element tensors and producing a layer-input tensor. The convolution operation step is for calculating a convolution operation to produce at least one result tensor and then storing the at least one result tensor into the memory. The output step is for outputting a block output.

Abstract: The present disclosure relates an integrated chip. The integrated chip includes a polysilicon layer arranged on an upper surface of a base substrate. A dielectric layer is arranged over the polysilicon layer, and an active semiconductor layer is arranged over the dielectric layer. A semiconductor material is arranged vertically on the upper surface of the base substrate and laterally beside the active semiconductor layer.

Abstract: An apparatus and method for providing a filtering false photon count events for each pixel in a DTOF sensor array are disclosed herein. In some embodiments, the apparatus includes: a light source configured to emit a modulated signal towards the object; a direct time of flight (DTOF) sensor array configured to receive a reflected signal from the object, wherein the DTOF sensor array comprises a plurality of single-photon avalanche diodes (SPADs); and processing circuitry configured to receive photon event detection signals from a center pixel and a plurality of pixels orthogonally and diagonally adjacent to the center pixel and output a valid photon detection signal, in response to determining whether a sum of the received photon event detection signals is greater than a predetermined threshold.

Abstract: A device includes a dielectric layer, an interlayer metal pad in the dielectric layer, a first capacitor over the interlayer metal pad, and a second capacitor over the dielectric layer. The first capacitor includes a first bottom capacitor electrode over and in contact with the interlayer metal pad, a first top capacitor electrode, and a first inter-electrode dielectric layer between the first bottom capacitor electrode and the first top capacitor electrode. The second capacitor includes a second bottom capacitor electrode over and in contact with the dielectric layer, a second top capacitor electrode, and a second inter-electrode dielectric layer between the second bottom capacitor electrode and the second top capacitor electrode.

Abstract: A self-calibration time-to-digital converter (TDC) integrated circuit for single-photon avalanche diode (SPAD) based depth sensing is disclosed. The circuit includes a SPAD matrix with a plurality of SPAD pixels arranged in m rows and n columns, the SPAD pixels in each column of SPAD pixels are connected by a column bus; a global DLL unit with n buffers and n clock signals; and an image signal processing unit for receiving image signals from the column TDC array. The circuit can also include a row control unit configured to enable one SPAD pixel in each row for a transmitting signal; a circular n-way multiplexer for circularly multiplexing n clock signals in the global DLL unit; a column TDC array with n TDCs, each TDC further comprises a counter and a latch, the latch of each TDC is connected to the circular n-way multiplexer for circular multiplexing.

Abstract: A method for testing semiconductor devices is disclosed, which includes: obtaining a result measured on a semiconductor device in one of a set of tests; comparing the result with a maximum value determined among respective results that were previously measured in one or more of the set of tests and a minimum value determined among respective results that were previously measured in one or more of the set of tests; determining, based on the comparison between the first result and the maximum and minimum values, whether to update the maximum and minimum values to calculate a delta value; comparing the delta value with a noise threshold value; determining based on the comparison between the delta value and the noise threshold value, whether to update a value of a timer; determining that the value of the timer satisfies a timer threshold; and determining that the semiconductor device incurs noise.

Abstract: A semiconductor device includes a source/drain diffusion area, a first doped region and a gate. The source/drain diffusion area, defined between a first isolation structure and a second isolation structure, includes a source region, a drain region and a device channel. The first doped region, disposed along a first junction between the device channel and the first isolation structure, is separated from at least one of the source region and the drain region. The first doped region has a dopant concentration higher than that of the device channel. The gate is disposed over the source/drain diffusion area. The first doped region is located within a projected area of the gate onto the source/drain diffusion area, the first isolation structure and the second isolation structure. A length of the first doped region is shorter than a length of the gate in a direction from the source region to the drain region.

Abstract: A differential amplifier is provided. The differential amplifier includes a first load, a second load, a current source, a differential pair circuit, a first and a second switch circuit. The differential pair circuit includes a first transistor, a second transistor, a third transistor, and a fourth transistor. The first switch circuit controls the first and the second transistors, and the second switch circuit controls the third and the fourth transistors. Through the control and selection of the first and second switch circuits, a differential pair is selected in the differential pair circuit to receive and process a first input signal and a second input signal for signal.

Abstract: Circuits and methods for protecting a device are provided. A first device to be protected includes a gate dielectric of a first thickness. A second device includes a gate dielectric of a second thickness that is less than the first thickness. A gate is shared by the first device and the second device.

Abstract: Some embodiments of the present disclosure are directed to a device. The device includes a substrate comprising a silicon layer disposed over an insulating layer. The substrate includes a transistor device region and a radio-frequency (RF) region. An interconnect structure is disposed over the substrate and includes a plurality of metal layers disposed within a dielectric structure. A handle substrate is disposed over an upper surface of the interconnect structure. A trapping layer separates the interconnect structure and the handle substrate.

Abstract: Some embodiments of the present disclosure are directed to a device. The device includes a substrate comprising a silicon layer disposed over an insulating layer. The substrate includes a transistor device region and a radio-frequency (RF) region. An interconnect structure is disposed over the substrate and includes a plurality of metal layers disposed within a dielectric structure. A handle substrate is disposed over an upper surface of the interconnect structure. A trapping layer separates the interconnect structure and the handle substrate.

Abstract: A semiconductor device comprises a source/drain diffusion area, and a first doped region. The source/drain diffusion area is defined between a first isolation structure and a second isolation structure. The source/drain diffusion area includes a source region, a drain region, and a device channel. The device channel is between the source region and the drain region. The first doped region is disposed along a first junction between the device channel and the first isolation structure in a direction from the source region to the drain region. The first doped region is separated from at least one of the source region and the drain region, and has a dopant concentration higher than that of the device channel. The semiconductor device of the present disclosure has low random telegraph signal noise and fewer defects.

Abstract: A method and a circuit for adaptive loop filtering in a video coding system are described. The method can include receiving a block of samples generated from a previous-stage filter circuit in a filter pipeline, the block of samples being one of multiple blocks included in a current picture, performing, in parallel, adaptive loop filter (ALF) processing for multiple target samples in the block of samples, while the previous-stage filter circuit is simultaneously processing another block in the current picture, storing, in a buffer, first samples each having a filter input area defined by a filter shape that includes at least one sample which has not been received, and storing, in the buffer, second samples included in the filter input areas of the first samples.

Abstract: A memory-adaptive processing method for a convolutional neural network includes a feature map counting step, a size relation counting step and a convolution calculating step. The feature map counting step is for counting a number of a plurality of input channels of a plurality of input feature maps, an input feature map tile size, a number of a plurality of output channels of a plurality of output feature maps and an output feature map tile size for a convolutional layer operation. The size relation counting step is for obtaining a cache free space size in a feature map cache and counting a size relation. The convolution calculating step is for performing the convolutional layer operation with the input feature maps to produce the output feature maps according to a memory-adaptive processing technique, and the memory-adaptive processing technique includes a dividing step and an output-group-first processing step.

Abstract: Circuits and methods for protecting a device are provided. A first device to be protected includes a gate dielectric of a first thickness. A second device includes a gate dielectric of a second thickness that is less than the first thickness. A gate is shared by the first device and the second device.

Abstract: Methods, kits, and compositions are provided herein for determining the eligibility of a subject to receive an anti-CD47 treatment based on a presence or absence of B-cells in the subject, and subsequently treating the eligible subject with the anti-CD47 treatment alone or in combination with one or more additional agent such as an anti-CD20 antibody.