571results about "Digital function generators" patented technology

The invention relates to a method and a system for cyclically generating a high-speed waveform signal. The method comprises the following steps of: cyclically splicing waveform data by a waveform circulation module; the control module transmits a variable related to the waveform length to the waveform circulation module; the data storage module stores waveform data information; the waveform circulation module comprises an address circulation module which is responsible for data addressing and outputting a data address to a data storage module; the data circulation module receives the waveform data sent by the data storage module, completes circulation splicing of the data and sends the data to the data transmission interface, and the data transmission interface sends the waveform circulation data to the DAC. The waveform data length can be effectively reduced, and the requirement for the storage space is lowered; the method is suitable for any high-speed DAC (Digital-to-Analog Converter) using a JESD204 protocol interface, and can run most FPGA models of Xilinx JESD204 IP; the device can be adapted to an FPGA on-chip storage resource BRAM, and can also be adapted to an off-chip storage resource DDR and other devices; waveform circulation can be achieved according to needs, waveform infinite circulation can be completed, and circulation of specific times can also be completed.

The invention discloses an arbitrary digital signal synthesis method and system. The method comprises the following steps: analyzing a target parameter of a digital signal to be synthesized; generating a first multi-path baseband waveform by using a digital direct frequency synthesizer according to a modulation rate parameter in the target parameters, and reading pre-stored noise data to generate a second multi-path baseband waveform; selecting a plurality of appointed baseband waveforms, and modulating the plurality of baseband waveforms by using carrier frequency and bandwidth parameters in target parameters to obtain a plurality of instantaneous frequency curves; and integrating the multiple instantaneous frequency curves to obtain a phase stepping curve of each path, selecting a phase modulation curve according to a phase modulation type in the target parameters, and performing digital direct frequency synthesis on the phase stepping curve and the phase modulation curve of each path to generate multiple parallel digital synthesis signals. According to the invention, the synthesis process of the digital signals of different modulation types is uniformly decomposed, so that the coupling degree of the module and the complexity of the system are greatly reduced, and meanwhile, the resource consumption is reduced.

The invention discloses a radio frequency calculation method and system based on a time coding metasurface. The metasurface is composed of a metal patch layer, an F4B dielectric layer and a metal backboard layer. By applying a specific periodic control signal, the metasurface can modulate electromagnetic information of an incident wave. A baseband signal of the incident electromagnetic wave represents a time function needing to be operated, and a control signal of the metasurface corresponds to a coding sequence needing to be operated. After the electromagnetic wave interacts with the metasurface, a reflection echo carries information of an operation result, and Fourier transform and convolution operation can be directly carried out in a radio frequency domain. According to the method, conversion of radio frequency signals is omitted, and calculation is directly completed in a radio frequency domain, so that calculation time delay and energy consumption can be effectively reduced; meanwhile, due to the fact that a traditional radio frequency link is omitted, the radio frequency calculator constructed through the metasurface has low hardware complexity and cost and a simple structure on the premise that the calculation precision is guaranteed, and the method is suitable for the application fields of communication, radar, sensing and the like.

The embodiment of the invention provides a broadband signal dynamic generation method and system based on multiple clock sources, and the method comprises the steps: monitoring a target frequency in real time through a field programmable gate array, and automatically switching to a corresponding target clock source based on a frequency band to which the target frequency belongs; correcting the phase difference corresponding to the new and old target clock sources through a phase compensation register when the target clock sources are switched, and loading a compensation value to a phase offset register of a direct digital frequency synthesis chip; a fundamental waveform of a broadband signal is generated based on a direct digital frequency synthesis chip, multiple harmonic components are superposed through a fractional accumulation method, and each harmonic amplitude is independently controlled through a look-up table method; an output signal of the power amplifier is collected through the analog-to-digital converter, error information of the actually measured waveform and the theoretical waveform is extracted, the phase offset register and the harmonic amplitude look-up table coefficient are dynamically corrected, and a target broadband signal is output. Unification of broadband coverage, high resolution and quick response is realized.

The embodiment of the invention provides a data processing method and a data processing device, relates to the technical field of chips, and solves the problem of low performance of a processor. According to the specific scheme, a processor core obtains input data of a target function through an input interface; the processor core determines polynomial coefficients corresponding to the input data based on the first index and the second index. And the processor core performs polynomial calculation based on the data bits and the polynomial coefficients to obtain output data of the target function, and outputs the output data of the target function through the output interface. The embodiment of the invention is used for the calculation process of the target function.

Embodiments described herein relate to computational multiplexing and applications thereof. The disclosed method and system for performing multiplexed diagnostic testing include providing a sample comprising nucleic acid units to a plurality of channels of a multichannel device; introducing a unique set of probes and reporter moieties into each of the channels; detecting a first indicator readout from a set of channels, wherein the first indicator readout is generated by interaction of a probe within the unique set with one of the nucleic acid units in the sample; generating a readout code for the sample based at least in part on the first indicator readout detected from the first set of channels; matching the generated readout code against a plurality of readout code entries of a lookup table; and generating a detection result of the multiplexed diagnostic testing for the sample based on the matched readout code.

The embodiments of the present disclosure provide a method for designing a high freedom parameterized frequency-modulated coded waveform, comprising: S1: obtaining a frequency-modulated coded vector; S2: selecting a coding model order to be utilized; S3: obtaining an parameterized instantaneous frequency function of a signal based on the frequency-modulated coded vector and the coding model order, integrating the parameterized instantaneous frequency function to obtain a phase of the signal, and generating the high freedom parameterized frequency-modulated coded waveform. Embodiments of the present disclosure provide a high freedom parameterized frequency-modulated coded waveform. The waveform has a continuous phase function, precise spectral control, and is configured to have a parameterized frequency-modulated waveform with high freedom.

The invention provides a dynamic residual amplification circuit based on charge sampling, and relates to the technical field of integrated circuits, the dynamic residual amplification circuit comprises an integrating circuit, a logic circuit, a charge pump and a zero-cross detection circuit, the integrating circuit is connected with the logic circuit and the zero-cross detection circuit, the integrating circuit converts a sampled input differential signal into a constant current, and the constant current is connected with the charge pump. The output node is continuously charged or discharged, so that the input differential signal is converted into a current integral quantity; the zero-cross detection circuit monitors the output voltage of the integrating circuit in real time, and when the output voltage is greater than or equal to the common mode voltage VCM, the moment when the output voltage is equal to the common mode voltage VCM is converted into a digital clock edge; the logic circuit is connected with the charge pump and outputs digital pulses to the charge pump according to the digital clock edge so as to control the reset time sequence of the whole dynamic residual amplification circuit; the charge pump is configured to output a residual voltage modulated by the digital pulse in response to the digital pulse. The application helps to improve the linearity of the residual signal.

The invention relates to the technical field of digital signal generation, in particular to a broadband sine wave digital signal generation method and system, and the method comprises the steps: obtaining an initial broadband sine wave digital signal; adopting a signal decomposition algorithm to obtain each modal component and residual component of the initial broadband sine wave digital signal; determining an interference influence factor of each modal component; determining a period difference degree and a superposition influence coefficient of each modal component; determining a relative difference factor and a relative influence coefficient of each modal component; determining a step parameter of each modal component; and in combination with an adaptive filtering algorithm, reconstructing the filtered modal component and residual component to obtain a de-noised broadband sine wave digital signal. According to the invention, the step length parameter of the adaptive filtering algorithm can be adjusted, the denoising effect of the adaptive filtering algorithm is improved, and then the high-quality broadband sine wave digital signal is generated.

The invention relates to transcendental function calculation, in particular to a resource reuse type transcendental function calculation device and method, and the device comprises a selector which receives input data, a data type and a control signal, and transmits the input data and the data type to a corresponding processing unit PE according to the control signal; the processing unit PE can process all types of transcendental functions and perform transcendental function calculation on input data according to data types; the result output unit is used for splicing the calculation results of the plurality of processing units PE in sequence to obtain output data; when all the processing units PE perform transcendental function calculation, the lookup table, the fixed-point multiplier, the floating-point multiplier and the floating-point adder are multiplexed; according to the technical scheme provided by the invention, the defect that the transcendental function is difficult to accurately calculate by using fewer hardware resources in the prior art can be effectively overcome.

A multi-bit linear-feedback shift register (LFSR) counter comprises a plurality of flip-flops serially coupled to one another with a separate flip-flop for each bit of a number of bits of the multi-bit counter, wherein each flip-flop has an individual binary value that, when combined with binary values of the other flip-flops, generates a unique sequence of bits representative of a unique state for the plurality of flip-flops; exclusive-or (XOR) logic between a pair of flip-flops based upon, at least in part, a plurality of taps for a primitive polynomial defined for the number of bits; and a look-up read-only table (LUT) mapping the unique sequence of bits generated by the plurality of flip-flops to a sequential decimal count value based upon, at least in part, an initial unique sequence of bits generated by the plurality of flip-flops.

The invention relates to the technical field of data processing, and discloses a digital feedback dynamic adjustment method and system for a neutron generator control signal, and the method comprises the steps: calling a reconfigurable waveform parameter set corresponding to a target neutron from a mapping relation according to the yield and beam spot characteristic demands of the target neutron; generating a basic high-frequency pulse sequence of the neutron tube, and performing pre-distortion filtering on the basic high-frequency pulse sequence to obtain a high-frequency pulse sequence of the neutron tube; collecting an anode current and a target voltage signal of the neutron tube applied with the high-frequency pulse sequence during operation, and converting the anode current and the target voltage signal into a digital feedback current; comparing the digital feedback flow with a reference waveform in the reconfigurable waveform parameter set in real time to obtain an error correction amount of the high-frequency pulse sequence; dynamically adjusting the high-frequency pulse sequence according to the error correction; applying the dynamically adjusted high-frequency pulse sequence to a corresponding electrode in a neutron tube through power; the neutron tube control signal processing efficiency can be improved.

A high speed transmitter system using a common mode controlled serialization stage embedded in an output stage is disclosed. In some embodiments, a transmitter comprises: a serialization circuit configured to convert parallel data into serial data using one or more serialization stages; a logic circuit configured to connect each input of a last one of the one or more serialization stages of the serialization circuit to a respective dedicated output stage of the output circuit via a respective logic function; and an output circuit configured to implement an output stage using an NMOS transistor to generate a signal based on the received input. Furthermore, the logic circuit is configured to a desired voltage value (e.g., a common mode of the output circuit) such that the switching point of the NMOS gm stage can be reduced and the gm stage can be activated for a reduced time.

Systems, methods, and other embodiments associated with high-performance arctangent computation at arbitrarily high precision are described. In one embodiment, an example method brackets an angle to a working range of an arctangent approximation polynomial. A closest index of a lookup table to the bracketed angle is determined. An angle shift for the bracketed angle is generated that is configured to move the bracketed angle to a high-precision segment of the range segments. A shifted angle is generated based on the bracketed angle and the angle shift. The arctangent approximation polynomial is evaluated at the shifted angle to produce an estimated arctangent of the shifted angle. A pre-computed arctangent corresponding to the closest index in the lookup table is retrieved from the lookup table in proximate memory. An augmented-precision arctangent is then generated from the estimated arctangent and the pre-computed arctangent.

A method for an automated speech recognition (ASR) model for unifying streaming and non-streaming speech recognition including receiving a sequence of acoustic frames. The method includes generating, using an audio encoder of an automatic speech recognition (ASR) model, a higher order feature representation for a corresponding acoustic frame in the sequence of acoustic frames. The method further includes generating, using a joint encoder of the ASR model, a probability distribution over possible speech recognition hypothesis at the corresponding time step based on the higher order feature representation generated by the audio encoder at the corresponding time step. The audio encoder comprises a neural network that applies mixture model (MiMo) attention to compute an attention probability distribution function (PDF) using a set of mixture components of softmaxes over a context window.

Systems, computer-implemented methods, and computer program products are provided that facilitate efficiently approximating a range of sideband frequencies. According to one embodiment, the system may include a processor executing computer-executable components stored in a memory. The computer-executable components may include a wave splitting component that generates a plurality of waveform segments using a definition of an intended waveform, the plurality of waveform segments being phase shifted. The computer-executable components may further include a rotation component that assigns a phase rotation to apply to at least one waveform segment of the plurality of waveform segments, the phase rotation being out of phase with a previous waveform segment of the plurality of waveform segments.

A processing element is configured to approximate a transcendental function. The processing element comprises an input storage and a look-up storage. The processing element obtains floating-point input data from the input storage representing having an input exponent value and an input mantissa value. The processing element looks up approximation parameters and an output exponent value from the look-up storage, wherein each group of approximation parameters and output exponent value are stored in the look-up storage in association with a respective range of a plurality of ranges that are defined by the input exponent value and the input mantissa value. The ranges cover values of the input exponent value and input mantissa value such that the output exponent value associated with each range does not change by more than a predetermined number. An approximation function is evaluated that approximates the transcendental function based on the looked-up approximation parameters and output exponent.

The invention discloses an arbitrary waveform signal generation method and device, equipment, a medium and a product, and relates to the field of signal processing, and the method comprises the steps: obtaining a frequency control word, and obtaining target waveform sampling data of a target waveform type; obtaining an output phase value for the frequency control word through a phase accumulation method based on a preset data length; wherein the preset data length is the length of the target waveform sampling data, and the length of the output phase value is an integral multiple of the preset data length; obtaining a target waveform amplitude value through a waveform look-up table method according to the output phase value and the target waveform sampling data; and performing analog-to-digital conversion on the target waveform amplitude value to obtain a first waveform signal. According to the invention, the waveform signal of any output wave can be prevented from generating periodic jitter.

To provide methods, systems, and apparatuses, including computer-readable media for performing neural network computations.SOLUTION: An exemplary process for generating output of a neural network layer in a system executing a neural network on a hardware circuit includes a host that receives a batch of input to the neural network layer. Each input is stored in a memory location identified by an address. The system identifies one or more duplicate addresses in a listing of addresses for one or more pieces of input. For each duplicate address: the system generates a unique identifier that identifies the duplicate address in the listing of addresses. The system obtains first input from memory locations identified by addresses corresponding to the unique identifiers and generates output of the layer from the obtained first input.SELECTED DRAWING: Figure 6

The invention relates to a square wave generation method and system based on a DDS signal transmitter, and belongs to the technical field of digital electronic signal synthesis. The method comprises the following steps: calculating a digital analog signal sampling frequency based on a square wave output frequency set by a user, and obtaining a digital fundamental frequency phase signal according to a phase-locked loop precision range; mapping the digital fundamental frequency phase signal into a sinusoidal amplitude signal according to a digital phase accumulation mechanism to obtain a digital-analog conversion clock, binarizing the sinusoidal amplitude signal, and outputting a square wave output period; dividing a square wave output period into a low level region and a high level region according to the precision range of the phase-locked loop, reading a square wave output frequency set by a user through a random storage address, and outputting a non-offset phase-locked square wave; and constructing an error accumulation self-checking architecture to calculate a frequency control word, and when detecting that the frequency control word has a decimal discarding error, recalculating the frequency control word.

A hardware acceleration circuit, a data processing acceleration method, a chip, and an accelerator are provided. The hardware acceleration circuit includes: an exponential function module, configured to obtain exponential function values of a plurality of data elements in a data set; an adder, configured to obtain an addition operation result of the exponential function values; a first processing circuit, configured to perform preset processing on the addition operation result, to process the addition operation result into at least first data and second data; a second processing circuit, configured to perform preset processing on at least the first data and the second data, to obtain a reciprocal of the addition operation result; and a third processing circuit, configured to perform preset processing on an exponential function value of an ith data element in the data elements and the reciprocal, to obtain a specific function value of the ith data element.

The invention discloses a Depthwise volume storage internal calculation device and method based on dynamic table lookup accumulation, and the device comprises a lookup table memory, a multi-address concurrent lookup table cache module, a dynamic table lookup accumulation module, a data write-back module, a feature map memory, a data acquisition module and a cache switching control module. Based on the design of a multi-access concurrent lookup table cache module, parallel access to a plurality of pre-stored lookup table values in a single period is supported, and dynamic space alignment, multi-channel parallel lookup and accumulation operation of an input feature map and the lookup table values are realized in cooperation with a dynamic lookup accumulation module; according to the method, a lookup table pre-generation technology of receptive field decoupling is adopted, and exponential compression of lookup table storage overhead is realized by separating a convolution kernel space dimension mapping relation. According to the method, the multiplication and addition operation of the traditional convolution is converted into a dynamic table look-up accumulation process, a high-complexity multiplier of the traditional Depthwise convolution is avoided, the calculation efficiency is remarkably improved, and meanwhile, the hardware resource consumption and the power consumption are greatly reduced.

The invention relates to the technical field of underground pipeline detection, in particular to a Hertz signal digitization method and system for underground pipeline detection. A digital detection excitation signal with stable Hertz frequency is generated by adopting a phase accumulation mechanism of a DDS (direct digital synthesis) technology and digital interpolation translation; a controllable random jitter frequency division strategy is introduced to generate multiple groups of single-frequency-point voltage signals to reduce spectrum aliasing and background noise; a response signal is sampled and filtered through a high-precision ADC and a self-adaptive FIR low-pass filter; and carrying out detection analysis and alarm on the target underground pipeline in combination with the frequency band energy normalization analysis and feature vector construction of the wavelet sub-band. According to the invention, digital improvement can be carried out on a 512 Hz analog signal detected by a traditional underground pipeline, and the sensitivity, the anti-interference capability and the signal processing precision of underground pipeline detection are greatly improved.

Disclosed is a system-on-chip. The system-on-chip including at least one operation processing unit configured to process an instruction based on a driving voltage and a clock signal, a temperature management module configured to estimate a cooling coefficient and a heating coefficient associated with the at least one operation processing unit based on a temperature measured at the at least one operation processing unit, and determine an operating frequency of the clock signal to be sent to the at least one operation processing unit based on the cooling coefficient and the heating coefficient, and a dynamic voltage and frequency scaling (DVFS) module configured to output a control signal for adjusting the driving voltage and the clock signal based on the operating frequency.

The present invention relates to a pop count-based deep learning neural network computation method, a multiply accumulator, and a device thereof. The computation method according to an exemplary embodiment of the present invention is a computation method for a deep learning neural network, including a step of generating one-hot encoding codes according to the type of first multiplication result values for a multiplication (first multiplication) of weights (W) and input values (A); a step of performing a pop-count for each generated code; and a step of accumulating result values for a constant multiplication (second multiplication) between each type of the first multiplication result value and each count value of the pop-count which are different constant values.

The invention relates to the technical field of signal generation, and discloses a signal generation system and method, and the system comprises a first signal generation module; the second signal generation module comprises an analog signal generation unit and a signal copying unit; the signal mixing module comprises an analog frequency mixing unit, a digital frequency mixing unit and a signal switching unit; the output end of the analog signal generation unit is connected with the input end of the analog frequency mixing unit; the first end of the signal copying unit is connected with the clock signal output end of the analog signal generating unit, the second end is connected with the mixing input end of the analog signal generating unit, and the third end is connected with the first end of the signal switching unit; the second end of the signal switching unit is connected with the output end of the analog frequency mixing unit, and the third end is connected with the input end of the digital frequency mixing unit. Therefore, a direct digital frequency synthesizer in the analog signal generation unit is calibrated when the clock signal is switched to, and spurious signals are eliminated based on a calibration result when the clock signal is switched to the analog local oscillator signal.

The utility model relates to a signal generation device, an analog-digital hybrid test board card and a test machine, a read-write control module reads an original waveform from a storage device according to a received read request instruction, and returns the read original waveform to an arbitrary waveform generator; the arbitrary waveform generator receives the original waveform and outputs a reconstructed waveform matched with the frequency of a sampling clock of the digital-to-analog conversion device to the first communication module. And the second communication module is connected with the first communication module, and the reconstructed waveform is received by the second communication module and then is transmitted to the digital-to-analog conversion device. According to the technical scheme, the reconstructed waveform can be generated by resampling the original waveform, the external clock chip outputs the sampling clock with the fixed frequency to the digital-to-analog conversion device, digital-to-analog conversion can be completed, the external clock chip does not need to adjust the frequency of the sampling clock, the hardware requirement for the external clock chip and the digital-to-analog conversion device is low, and the design cost is reduced.

The embodiments of the present application relate to the technical field of chips. Provided are a data processing method and a data processing apparatus, which ameliorate the problem of the poor performance of a processor. The specific solution involves: a processor core acquiring input data of a target function by means of an input interface; on the basis of a first index and a second index, the processor core determining a polynomial coefficient corresponding to the input data; and on the basis of a data bit and the polynomial coefficient, the processor core performing polynomial calculation to obtain output data of the target function, and outputting the output data of the target function by means of an output interface. The embodiments of the preset application are used for a calculation process of a target function.

Aspects of the disclosure provide methods and apparatuses for point cloud compression and decompression. In some examples, an apparatus for point cloud compression / decompression includes processing circuitry. The processing circuitry determines to use a prediction mode for coding (encoding / decoding) information associated with a current point in a point cloud. In the prediction mode, the information associated with the current point is predicted based on one or more neighbor points of the current point. The processing circuitry calculates, using integer operations, a distance-based weighted average value based on distances of the one or more neighbor points to the current point, and determines the information associated with the current point based on the distance-based weighted average value.

A compensation lookup table generator includes a maximum power voltage determiner, a rush power reduction amount operator, an allowable current increase amount operator and a lookup table modifier. The maximum power voltage determiner is configured to determine a maximum power voltage based on a first lookup table, a peak gain setting value and a power voltage curve. The rush power reduction amount operator is configured to operate a rush power reduction amount based on the maximum power voltage and a rush power limit. The allowable current increase amount operator is configured to operate an allowable current increase amount based on the maximum power voltage and the rush power reduction amount. The lookup table modifier is configured to generate a second lookup table by modifying the first lookup table based on the peak gain setting value and the allowable current increase amount.