73 results about "Large array" patented technology

An in-memory computing (IMC) accelerator using a high-density operation circuit and a low-power sense amplifier as a peripheral circuit includes a plurality of dynamic random-access memory (DRAM) banks each including a pair of cell arrays, a data supply logic, a memory, and a controller for IMC, a global SRAM, and a top-level controller, wherein the cell array includes a plurality of subarrays, each of the subarrays includes a DRAM array including a big array and a little array, and an arithmetic circuit configured to perform an operation, and the arithmetic circuit includes a sense amplifier configured to amplify a bit line voltage difference, and a compact multiply-accumulate (MAC)-single instruction multiple data (SIMD) unit (CMSU) for an MAC operation and an SIMD operation, so that functionality of an in-memory operation is diversified.

The liquid nitrogen testing Dewar comprises a cold end cold head, a variable-temperature cold head, a variable-temperature rotating block, a transition block, a loading cold head, the large-area-array detector, an optical window, a liquid nitrogen storage cavity and a shell. The cold energy of the liquid nitrogen storage cavity forms a solid conduction path through the cold-end cold head, the variable-temperature cold head and the loading cold head, and the variable-temperature cold head and the loading cold head adopt a low-heat-conduction hollow structure to construct a basic temperature difference; the variable-temperature rotating block is connected with the variable-temperature cold head and the transition block in a threaded fit mode, and the cooling capacity obtaining degree of the object carrying cold head is changed by means of position adjustment. According to the invention, flexible and accurate temperature control within the range of 80K-180K can be realized, the cooling capacity transmission is efficient, the use is convenient, and the device is suitable for testing and screening of large-area-array detectors with different temperature requirements.

The invention provides an avalanche photodiode with asymmetric optical resonance and a photosensitive array, and the avalanche photodiode comprises an electrode layer which is internally provided with a first windowing region; the top reflecting mirror is arranged in the first windowing area, the transmittance of the incident side of the top reflecting mirror to the light with the target wavelength is higher than a preset transmittance threshold value, and the reflectivity of the emergent side of the top reflecting mirror to the light with the target wavelength is higher than a preset reflectivity threshold value; the substrate is arranged on the emergent side of the top reflecting mirror; the buffer layer is arranged on one side, deviating from the top reflecting mirror, of the substrate; the active region is arranged on the side, away from the top reflector, of the buffer layer; and a bottom high reflective metal film. The top reflector and the bottom high-reflectivity metal film form an asymmetric optical resonant structure, and the purposes of filtering and absorption enhancement are achieved. And the bottom high-reflection metal film is low in process difficulty and suitable for preparation of large-array devices.

The invention provides an ultra-large array plane radar satellite decoupling vacuum thermal test method, which comprises the following steps of: arranging a satellite and radar antennas in a vacuum tank, detaching the radar antennas from the satellite, arranging the radar antennas on one side of the satellite side by side, and respectively arranging infrared simulation devices around the satellite to simulate external thermal flow; a wave absorbing device is arranged on an antenna array plane to absorb microwaves and simulate external heat flow, the power level of an infrared simulation device is adjusted according to the heat radiation level of the satellite surface, the power level of the wave absorbing device is adjusted according to the microwave emission power and the heat radiation level of the antenna array plane, and a vacuum thermal test is performed on the satellite and the antenna. According to the invention, vacuum thermal tests on a satellite and a radar antenna due to an overlarge antenna array plane can be avoided, a whole-satellite microwave wireless emission state test can be carried out, extremely high-risk operation caused by hanging and unfolding in an antenna tank is avoided after the antenna and the satellite are decoupled, the tank occupation period of links such as test state establishment and withdrawal is greatly shortened, and the test efficiency is improved. And the model development efficiency is improved.

A miniaturized, automated method for controlled printing of large arrays of nano- to femtoliter droplets by actively transporting mother droplets over hydrophilic-in-hydrophobic (“HIH”) micropatches. The technology uses single or double-plate devices where mother droplets can be actuated and HIH micropatches on one or both plates of the device where the droplets are printed. Due to the selective wettability of the hydrophilic micropatches in a hydrophobic matrix, large nano- to femtoliter droplet arrays are created when mother droplets are transported over the arrays. The parent droplets are moved by various droplet actuation principles. Also, a method using two plates placed one top another while being separated by a spacer. One plate is dedicated to confirming and guiding parent droplets by using hydrophilic patches in a hydrophobic matrix, while the other plate contains HIH arrays for printing of the droplets. When the parent droplet guidance plate is rotated over the plate dedicated to printing of nano- to femtoliter droplets, the droplets are dispensed inside the HIH array utilizing their selective wettability. The methods allow the parent droplets to move over the HIH arrays many times, providing advantages for performing bio-assays or miniaturized materials synthesis in nano- to femtoliter sized droplets. With controlled evaporation of the dispensed droplets of solution, large arrays of printed material can be generated in seconds. The methods provide a nano- to femtoliter droplet printing technique for a wide variety of applications, e.g., protein- or cell-based bio-assays or printing of crystalline structures, suspensions of nanoparticles or microelectronic components.

This invention discloses a level shifting structure for an ultra-large area array ultra-low temperature infrared image sensor, comprising a conversion signal input transistor; a level clamping transistor whose gate receives a level VREF, and whose source is connected to the source of a feedback transistor, the gate of an inverting N-transistor, and the gate of an inverting P-transistor; the drain of the level clamping transistor connected to the source of the conversion signal input transistor; the drain of the conversion signal input transistor grounded; the gate of the conversion signal input transistor connected to IP; the drain of the feedback transistor connected to the source of a pull-up transistor, whose gate is connected to IP, and whose drain is connected to power supply VDD; the source of the inverting P-transistor connected to power supply VDD, whose drain is connected to the source of the inverting N-transistor, and whose drain is grounded; and a level VREF generated by a level VREF clamping matching voltage generation circuit. This ensures that the source-drain voltage of the conversion signal input transistor (low threshold transistor) does not exceed 1.2V, ensuring that the device operates within its normal voltage range and avoiding the risk of breakdown.

A method for sharing reconfigurable array modes in a MIMO frequency-modulated millimeter-wave radar array is proposed. Based on the echo signal data matrices obtained after analog-to-digital conversion of different modes, range-dimensional Fast Fourier Transform (FFT) and velocity-dimensional FFT are performed on them respectively to obtain the range-Doppler spectrum of different transmit / receive antenna combinations. Different modes with the same gain in the MIMO radar system are combined. Among the combinations with the same gain, the mode corresponding to the largest virtual array aperture in the horizontal and elevation directions under that gain is selected. The angles of these two modes are resolved and matched separately, and then combined to achieve the maximum virtual array aperture resolution. Digital beamforming is used to match the angle information obtained under the maximum aperture virtual array in the horizontal and elevation dimensions. This invention can effectively increase the maximum array aperture in the original modes with the same gain by utilizing the complementary effect between different reconfigurable array modes while using virtual array reconfigurable technology or MIMO technology.

The application relates to a fast demodulation system and method for a large-scale microcavity array on a chip, and the system comprises a frequency-sweeping laser, a first fiber coupler, a microcavity array transmission signal arm, an auxiliary interference reference arm, a same-frequency trigger arm and a signal processing device. The first fiber coupler is used for dividing a light source into a first light source and a second light source. The microcavity array transmission signal arm is used for receiving the first light source of the frequency-sweeping laser, and the first light source enters a large-array microcavity sensor device after polarization processing, and is converted into an electric signal after receiving the disturbance of an external sensing signal. The auxiliary interference reference arm is used for receiving the second light source of the frequency-sweeping laser, and two beams of coherent light with opposite phases are obtained after processing the second light source, and then the coherent light is converted into an electric signal. The same-frequency trigger arm is used for transmitting a same-frequency trigger signal emitted by the frequency-sweeping laser. The signal processing device is used for signal acquisition and processing. The application effectively suppresses the frequency-axis distortion and spectral line broadening caused by the frequency-sweeping nonlinearity of the laser, and can be applied to the accurate demodulation of a large-scale, high-Q microcavity sensing array.

The application discloses a method for obtaining and separating a scattering field of a large-scale array antenna, and the steps are as follows: obtaining a radiation field of each unit array of a subarray, and calculating a scattering field of an antenna mode item in each unit array; using the scattering field of the antenna mode item in each unit calculated by a pattern multiplication method to solve a mutual coupling correction matrix; using the mutual coupling correction matrix to correct the scattering field of the structure mode item calculated by the pattern multiplication method; adding the scattering field of the antenna mode item and the scattering field of the structure mode item to obtain a scattering field of the array; equivalently constructing the subarray and performing phase compensation to obtain a scattering field of each unit of the large-scale array antenna; and obtaining a total scattering field of the large-scale array antenna and the scattering field of the total antenna mode item and the total structure mode item through field superposition. The application introduces the mutual coupling correction matrix and the subarray theory, avoids directly performing matrix operation on the large-scale array antenna, improves the calculation speed, and solves the problem that a large amount of calculation time and resources are consumed when a scattering field of the large-scale array antenna is obtained in the prior art.

The invention discloses a region decomposition-based two-dimensional array resistivity inversion method, electronic equipment, a storage medium and a device. The method comprises the steps that area decomposition is conducted on a large array lateral logging response data set based on the layer interface position of a target area, a plurality of sub-areas are obtained, and every three adjacent stratums form one sub-area; the thickness and the center point position of each stratum are calculated, and then the thickness and the center point position of each sub-region are obtained; constructing a one-dimensional inversion model based on the stratum to perform one-dimensional array resistivity inversion; and constructing a two-dimensional inversion model based on the sub-regions to perform two-dimensional array resistivity inversion, and taking a one-dimensional inversion result as an initial parameter value of the two-dimensional inversion model. The method fully considers the influence of radial mud invasion and longitudinal surrounding rock layers, and strengthens the information constraint of the upper and lower well sections, thereby weakening the influence of surrounding rock, guaranteeing the extraction precision of the real resistivity of the stratum, and achieving the rapid and accurate imaging of the two-dimensional profile of the resistivity of the stratum.

The utility model discloses a surface array optical fiber collimator, is constituted by mould pressing lens array, a plurality of pigtail, a plurality of connecting pipe or ridge hole plate, one side of lens array is array convex surface, the other side is a plurality of ridge inclined plane, one side end surface of connecting pipe or ridge hole plate and lens array gluing, the other side end surface and pigtail gluing, since every optical fiber collimator is through the precise light adjustment between pigtail relative to plano-convex lens, and is fixed after the axial transition through connecting pipe or ridge hole plate, can guarantee the mechanical positioning accuracy index of every collimator and the optical performance index in the whole surface array optical fiber collimator, and the quality can be guaranteed, in addition, the utility model has compact structure and low material cost, and easy to assemble, the yield is high, is suitable for building large array.

The invention discloses a satellite infrared multi-probe metadata homogenization method based on variational self-encoding, which comprises the following steps of: acquiring multichannel observation brightness temperature data of a satellite large-array infrared detector, constructing a variational self-encoder model, designing a comprehensive loss function, adopting a dynamic weight scheduling mechanism, gradually adjusting the weight of each loss item in a training process, and carrying out uniformization on the multi-channel observation brightness temperature data of the satellite large-array infrared detector. And carrying out iterative training by adopting Adam and a piecewise learning rate, storing an optimal model, reconstructing observation data by utilizing the trained model, and outputting homogenized multi-detection-element hyperspectral brightness temperature data. The method can automatically extract and learn the complex nonlinear characteristics of the hyperspectral data, has higher accuracy and robustness compared with an empirical correction method, and improves the consistency and stability of the data of the large-array detector; and the method has relatively high expandability and universality.

The invention discloses a memory low-voltage write circuit and method for high-bandwidth magnetic computing and a memory device. The memory low-voltage write-in circuit for high-bandwidth magnetic computing comprises a plurality of memory arrays, a source line multiplexing unit and inter-array buffer stage units, wherein the inter-array buffer stage units are arranged between two adjacent memory arrays; the storage array generates a write-in voltage within a preset low threshold range according to the control signal; the source line multiplexing unit multiplexes the source lines of the two adjacent storage arrays and the write-in driving unit; and the inter-array buffer stage unit generates and outputs a control signal of the next-stage storage array according to the bit line driving signal and the source line driving signal of the previous-stage storage array. Data writing is completed directly through logic voltage, the area overhead of a level converter in a traditional writing circuit is saved, and the problem of insufficient driving caused by parasitism in a large array is solved through the design of the buffer stages between the arrays.

The invention relates to the technical field of sensing product testing, in particular to a super-large array packaging conductivity testing device which comprises a wafer testing table, an open circuit testing plate and a short circuit testing plate, an objective table is installed at the top of the wafer testing table, and a swing arm positioning seat is installed at the top of the wafer testing table and on the back face of the objective table. The outer wall of the swing arm positioning seat is provided with a test probe, the top of the wafer test bench and the back surface of the swing arm positioning seat are provided with a main control unit, and the outer wall of the main control unit is connected with the swing arm positioning seat through a test line. According to the super-large array packaging conductivity testing device, the open circuit testing plate and the short circuit testing plate in the device are used for testing the to-be-tested product, and the problems that an existing testing device and an existing testing method cannot directly point-contact the two ends for measurement, the accuracy of a testing result is poor, the efficiency is low, and the product is prone to being damaged are solved.

The application relates to the technical field of data storage, in particular to an array data storage method, an electronic device and a medium, which comprises the following steps: S1, initializing an array A and recording corresponding default values D in the array A, wherein the total number N of elements in the array A is greater than a preset first threshold value, and the expected total number M of non-default value elements in the array A is less than a preset second threshold value; S2, initializing an empty tree structure T; S3, obtaining an array subscript of a non-default value to be newly added in the array A and a corresponding value; S4, storing the array subscript of the non-default value to be newly added as a key and the corresponding value as a value in a to-be-newly-added node corresponding to the tree structure T, inserting the to-be-newly-added node into the tree structure T according to subscript sequence, and updating the tree structure T; and S5, if all elements in the array A are processed, ending the process, or returning to step S3. The application improves the memory utilization rate of large array data storage.

The utility model discloses an array electro-acupuncture structure, an excitation circuit and medical beauty equipment, relates to the technical field of medical equipment, and discloses the array electro-acupuncture structure comprising a plurality of electro-acupuncture units; each electric needle unit is of a matrix structure composed of a plurality of electric needles, the plurality of electric needles of each electric needle unit are electrically connected, and each of rows and columns of each electric needle unit at least comprises two electric needles; each electro-acupuncture unit is used for sequentially receiving current excitation output energy; the electroacupuncture units are adjacently spliced to form the array electroacupuncture structure, and the array electroacupuncture structure is a matrix structure. A large array electro-acupuncture structure is divided into small electro-acupuncture unit matrixes for excitation, and the electro-acupuncture number of each electro-acupuncture unit is limited to ensure certain area dispersed current. The electric needles in each electric needle unit are electrically connected so as to receive unified current excitation and output energy. Due to the fact that the number of the electric needles excited at the same time is reduced, the difference between the electric needles is reduced, and therefore energy distribution can be more uniform through sequential excitation of the electric needle units.

The invention belongs to the field of target pose measurement, and particularly relates to a target pose measurement system and method based on curved surface features. The system comprises a laser emission module, a feature identification module and a processing module. The laser emission module and the feature recognition module are arranged towards a to-be-measured curved surface target, and M feature points are arranged on the to-be-measured curved surface target; the laser emission module is used for emitting N non-parallel laser beams to a to-be-measured curved surface target, N laser spots are formed on the to-be-measured curved surface target, and the mass centers of the N laser spots are not collinear; the view field of the feature recognition module covers all different pose envelope ranges of the to-be-detected curved surface target; the receiving end of the processing module is electrically connected with the sending end of the feature recognition module. According to the invention, the measurement range can be expanded through the expansion of the number of laser beams and the field of view of the detector, so that the pose measurement of an ultra-large array target can be realized, and the cost is low.

This invention discloses an extra-large array-type flange processing equipment, relating to the field of flange processing equipment. It includes a milling machine base, a milling machine body movably mounted on the milling machine base, a mounting seat mounted on the milling machine body, and a milling cutter movably mounted on the mounting seat. It should be noted that, in this embodiment of the invention, when a problem occurs with the milling cutter, a start switch can be triggered, causing four avoidance cylinders to move the clamping plate downwards and disengage from the milling cutter via four lowering rods. This also causes the two shielding side boxes to merge and shield the flange material plate. Therefore, through rapid detection, risk avoidance, and physical shielding, the impact damage to the expensive flange material plate by the milling cutter is prevented to the greatest extent possible. Furthermore, avoiding collisions also protects the expensive milling cutter, reduces damage to key machine tool components such as the spindle and guide rails caused by severe impacts, extends equipment life, reduces maintenance costs, improves the safety of the entire processing process, and reduces the potential for safety accidents.

The application provides a broadband near-field beam training method, device and medium, and belongs to the technical field of Internet communication transmission. The method comprises the following steps: acquiring a system model of a broadband super-large array communication system; determining a first beam splitting effect of a sparse array in the broadband in a spatial domain and a second beam splitting effect of the broadband in a frequency domain based on the system model; and performing super-resolution broadband near-field beam training based on the first beam splitting effect and the second beam splitting effect. The technical scheme can realize super-resolution broadband beam training with extremely low overhead.

The application discloses a kind of millimeter wave rectangular waveguide slot antenna structure device, including two groups of thinner and parallel first metal plate, second metal plate and two first metal conductive layer with rectangular waveguide transmission line, second metal conductive layer, the first metal plate, second metal plate and first metal conductive layer, second metal conductive layer are sequentially crossed and stacked, the present application provides a new type of rectangular waveguide slot antenna structure device, with cost advantage, can be used for electromagnetic wave emission or reception, or for both, with very good antenna performance, can work at different frequencies, such as 30GHz-300GHz range, but also outside this range, while can be used as independent antenna can also be integrated with other components can also be assembled into larger array with multiple waveguide slot antenna groups.

The invention provides a common excitation amplitude sparse reconfigurable array synthesis method, and relates to an information transmission and processing technology. The method comprises the following steps: 1) constructing a mathematical optimization model through a radiation characteristic requirement and a null requirement of a to-be-synthesized wave beam, and solving the mathematical optimization model by using particularity of conjugate symmetric excitation and a continuous convex approximation algorithm; 2) utilizing a feasible point tracking algorithm, introducing a slack variable to help to solve a feasible point, and overcoming a non-feasible solution caused by non-convex constraint; the method has the advantages that a plurality of radiation directional diagrams can be formed by one sparse array, so that a plurality of different application scenes can be met; the number of array elements is reduced through the design of a sparse array, so that the system cost is further reduced; due to the characteristic of excitation weight conjugate symmetry, optimization variables are remarkably reduced, and the advantages of the method on a large-scale array are highlighted; the DRR constraint is added, so that the dynamic range and the quantization precision required by the attenuator are reduced, and the hardware complexity is simplified.

This invention discloses a terahertz receiver and a focal plane camera device based on a variable capacitor. The receiver includes a connected zigzag inductor and interdigitated capacitors; each finger of the interdigitated capacitor is composed of closely arranged double-cross-shaped repeating units, enabling it to efficiently absorb terahertz waves without the need for an additional absorber. Terahertz radiation changes the dielectric constant of the dielectric, causing changes in capacitance and resonant frequency; detection is achieved by detecting these changes. The focal plane camera device includes an array of the aforementioned receivers. Each receiver is coupled to a zigzag-shaped center feed line via a coupling feed line, and each unit has a different resonant frequency, achieving frequency division multiplexing readout. This invention combines absorption and resonance functions into one, greatly simplifying the structure, achieving high absorption efficiency, possessing dual polarization and intrinsic frequency division multiplexing capabilities, and easily expanding into large arrays. It can be widely applied in fields such as high-sensitivity terahertz passive imaging and spectral analysis.

The application relates to a method for deducting the background noise of a large-array CMOS detector in the digital domain, in which, firstly, a dark field image L1 generated by the large-array CMOS detector is set, and the dark field image is aligned and blocked, the mean value of the blocked image is counted, and the image block mean value matrix corresponding to the dark field image L1 is obtained; the detector is set to normally image, and a photosensitive image L2 is obtained; a radiation calibration dark field image L0 is obtained through a detector radiation calibration method, and the radiation calibration dark field image data is blocked according to the same division mode as the dark field image data, the mean value of the blocked image is counted, and the image block mean value matrix corresponding to the radiation calibration dark field image L0 is obtained; the variances of the two image block mean value matrices are calculated respectively, and the sizes of the two variances are compared to determine whether the dark field image is effective. The method realizes accurate deduction of the background of a large-array CMOS detector with a large number of pixels, improves the adaptability in the imaging process of the detector, and improves the signal-to-noise ratio index of the image.

Disclosed is an electromagnetic-circuit co-design approach for massively reconfigurable, multifunctional, and high-speed programmable metasurfaces with integrated chip tiling. The ability to manipulate the incident electromagnetic fields in a dynamically programmable manner and at high speeds using integrated chip tiling approach is also disclosed. The scalable architecture uses electromagnetic-circuit co-design of metasurfaces where each individual subwavelength meta-element is uniquely addressable and programmable. The disclosed device comprises a large array of such meta-elements. The design relies on integrated high frequency switches designed in conjugation with meta-element for massive reconfigurability of incident amplitude and phase. The disclosed chip is multi-functional and can perform beamforming, high speed spatial light modulation, dynamic holographic projections, and wavefront manipulation.

The application relates to a kind of MIMO millimeter wave radar two-dimensional super-resolution angle measurement methods using space-time virtual conversion, comprising: using the orthogonality of transmitting signal, MIMO sparse array is carried out MIMO virtual array transformation, obtain space domain virtual array, obtain the received signal data of space domain virtual array, using the independence of received time domain multiple fast shot data, then carry out time domain virtual transformation, obtain space-time virtual array;Under the antenna layout of space domain virtual array, if the received signal data is single fast shot data, then two-dimensional super-resolution angle measurement is realized using orthogonal matching pursuit (OMP) sparse reconstruction method, if the received signal data is a small amount of fast shot data, then two-dimensional super-resolution angle measurement is realized using sparse bayesian learning (SBL) sparse reconstruction method;If the received signal data is multiple fast shot data, under the antenna layout of space-time virtual array, super-resolution angle measurement is realized using OMP method.The method of the application can increase the effective aperture of array, and significantly improve the angle measurement accuracy of pitch and azimuth dimension.

A hierarchical activation-broadcast network for column-oriented neural-network compute arrays is disclosed. Activation values corresponding to each processing row are distributed concurrently to all columns of processing elements through a hierarchical latch tree that fans out the signals synchronized within one clock period. The broadcast network maintains precise timing alignment among columns while minimizing capacitive loading and clock skew. The distribution fabric may be fabricated on a separate die or interposer and connected to the compute die by high-density hybrid bonds. The approach enables wide, low-skew activation delivery across large arrays, reducing energy consumption and improving frequency scalability for CASCADE architectures.

The invention provides a front-end driving compensation circuit and method for a super-large array pixel sensor, and solves the technical problems that the performance of a pixel sensor chip is poor and the improvement of a frame rate is limited due to the non-ideal effect of an existing super-large array pixel sensor. The circuit comprises a buffer chain, a delay detection module and a delay compensation module. The delay detection module is used for storing the delay time of the to-be-calibrated signal in the calibration state when the to-be-calibrated signal is transmitted in the buffer chain; and the delay compensation module is used for reading delay time, compensating the row driving signal in the buffer chain during normal work, and synchronizing the row driving signal of the buffer chain. The circuit can be widely applied to the technical field of integrated circuits.