240 results about "Acquisition time" patented technology

The invention discloses a circuit breaker on-line monitoring edge terminal and a high and low speed synchronous acquisition method. The edge terminal comprises a high-speed acquisition card, a low-speed acquisition card, a time synchronization module and an edge computing host. And the high-speed acquisition card and the low-speed acquisition card are respectively connected with the circuit breaker monitoring sensor and respectively acquire signals at different sampling frequencies. And the time synchronization module realizes time synchronization of the high-speed and low-speed acquisition cards through a trigger signal generation unit and a timestamp alignment unit. The edge computing host comprises a data processing unit, a data storage unit and a data uploading unit and is responsible for filtering, feature extraction, fault diagnosis, historical data storage and data uploading of collected signals. The invention further provides a high and low speed acquisition time synchronization method, high and low frequency signals can be synchronized in real time in a circuit breaker operation event, the accuracy and synchronism of data acquisition are ensured, and therefore the reliability and comprehensiveness of monitoring are improved.

The invention discloses a method for realizing data acquisition clock synchronization through Beidou time service and an FPGA phase-locked loop, and the method comprises the steps: obtaining a short message service signal of a Beidou satellite navigation system, and carrying out the analysis to obtain time service information; controlling a digital phase-locked loop to calibrate a local clock signal according to the time service information; monitoring the deviation between the time service information and the local clock signal, and adjusting the frequency and phase of the local clock signal through feedback control of the digital phase-locked loop; monitoring a clock synchronization state, setting a clock deviation threshold value, judging the synchronization state of the time service information and the local clock signal, if the deviation exceeds a threshold value range, triggering an alarm signal, and taking a recovery measure according to an abnormity type and duration time; a hardware timestamp function is integrated, clock synchronization with other acquisition nodes is carried out through a precise time protocol, and the acquisition time of seismic data is recorded.

The embodiment of the invention provides a foreground target segmentation and reconstruction method and device, equipment and a storage medium, and relates to the technical field of computer vision. The method comprises the following steps: directly inputting a video frame into a target segmentation model so as to generate a foreground target segmentation mask, associating the foreground target segmentation mask with a corresponding initial Gaussian point cloud as input data, inputting the input data into a space-time Gaussian splash model for data processing, and performing time slice division on the input data so as to obtain a time slice division result; and obtaining a slice Gaussian point cloud corresponding to each time slice, obtaining a two-dimensional Gaussian curved surface corresponding to each slice Gaussian point cloud, and obtaining a rendered image corresponding to each visual angle according to the plurality of two-dimensional Gaussian curved surfaces. And obtaining a foreground target rendering sequence at the processing moment according to the rendered image corresponding to each visual angle, and generating a foreground target reconstruction result according to the foreground target rendering sequences at all acquisition moments. While the dependence on external prior information is reduced, the stability of a time sequence can be ensured, and the accuracy of a rendering result is improved.

The invention belongs to the technical field of electromagnetic signal monitoring and analysis, and particularly relates to a multi-dimensional multi-channel broadband signal monitoring platform and an amplitude phase testing method.The monitoring platform is composed of a plurality of eight-channel receiving single machines, eight receiving single machines serve as a group, and any one of the eight receiving single machines serves as a basic function single machine in each group; a synchronous reference signal is generated through a parallel synchronous reference correction processor built in each functional single machine, and n * 8 parallel signals are received. The amplitude and phase test method comprises a frequency domain amplitude and phase consistency test and a time domain amplitude and phase consistency test. According to the invention, a synchronization reference rule correction processing mechanism and a frequency, amplitude and phase correction module are constructed, so that the technical problems that a long-term stability synchronization test cannot be realized only by adopting a step-by-step single-channel test in a traditional test, radio frequency phase shift introduced by a multi-channel synchronization test is inconsistent with an acquisition time sequence and the like are solved; and the signal parallel receiving and consistency analysis capabilities of 8 * n channels are realized.

The invention provides an information acquisition method, device, system and equipment, and the method comprises the steps: carrying out the spectrum analysis of a first pressure signal of a combustion chamber based on a plurality of target frequency bands corresponding to the combustion chamber for any combustion chamber of a gas turbine, and obtaining the signal feature value of the first pressure signal at each target frequency band; according to the signal characteristic value of each target frequency band, identifying a to-be-adjusted frequency band from the plurality of target frequency bands; adjusting a plurality of target frequency bands corresponding to the pressure sensor based on the signal characteristic value of the frequency band to be adjusted, and performing spectral analysis on a second pressure signal of the combustion chamber based on each adjusted target frequency band to obtain a signal characteristic value of the second pressure signal in each adjusted target frequency band; the acquisition time of the second pressure signal is later than that of the first pressure signal. Therefore, the frequency band corresponding to the combustion chamber can be automatically adjusted according to the signal characteristic value of the pressure signal in the combustion chamber in each target frequency band, and the accuracy of obtaining the characteristic information is improved.

The invention discloses an oil and gas long-distance pipeline automatic control system fault identification method based on deep learning, and the method comprises the following steps: collecting and preprocessing a time sequence, and generating a standardized input sequence; constructing a physical constraint feature set, and generating a physical observable representation; encoding the valve opening instruction time sequence and the pump station frequency instruction time sequence to form a combined input sequence; obtaining Koopman state representation, and generating a linear evolution state sequence; a residual nonlinear compensation structure is introduced to generate residual features, and a comprehensive state sequence is formed; time scale splitting is carried out, and fast, medium-speed and slow time scale feature sequences are decomposed; identifying anomalies and generating a time scale anomaly feature set; leakage energy loss indexes are calculated and fused to form fault mode drift representation; judging the pipeline state, and outputting a fault recognition result. According to the invention, a deep Koopman operator structure is adopted, accurate identification of pipeline faults is realized, and the method has the advantage of high diagnosis reliability.

The invention discloses an early warning method and device for multi-device time sequence data and electronic equipment. The method comprises the steps of determining a time sequence data stream; dividing the time sequence data stream according to the model identifier to obtain a plurality of sub-time sequence data streams, sliding on the sub-time sequence data streams by using the time windows according to a preset step length from the starting timestamps of the sub-time sequence data streams, and determining the time sequence data under the time window which slides each time as a target time sequence data stream; performing timestamp alignment and data filling on time sequence data in the target time sequence data stream to obtain an aligned target time sequence data stream, and predicting a time sequence data predicted value corresponding to the aligned target time sequence data stream in a future time period through a prediction model corresponding to the aligned target time sequence data stream; and determining early warning information corresponding to the target time sequence data stream according to the time sequence data predicted value. The technical problem that the accuracy of an early warning result is affected by data time sequence dislocation caused by inconsistent acquisition time of time sequence data between devices is solved.

The embodiment of the invention provides a communication method and a communication device. The method comprises the steps of determining at least two pieces of channel information, such as latest channel information and at least one piece of historical channel information, and then determining input of a first AI model or determining a replaced second AI model according to a time interval between channel acquisition times corresponding to the channel information and at least one capability requirement of the first AI model. Therefore, the input of the AI model can be matched with the capability requirement of the AI model, so that the input of the AI model is more suitable, and the performance loss caused by taking improper channel information as the input of the AI model is avoided.

The invention discloses a pole piece slitting method, device and equipment and a storage medium, and relates to the technical field of batteries, and the pole piece slitting method comprises the steps that a first signal of the displacement of a slitting assembly in the pole piece slitting equipment in the axial direction of the slitting assembly changing along with the collection moment is obtained, and the axial direction is perpendicular to the pole piece slitting direction of the slitting assembly; based on the first signal, the current state of the slitting assembly is determined; and based on the current state of the slitting assembly, the pole piece slitting equipment is controlled. According to the invention, the first signal that the displacement of the slitting assembly in the axial direction in the pole piece slitting equipment changes along with the collection time can be obtained online in real time, and the current state of the slitting assembly is determined through the signal, so that the pole piece slitting equipment is controlled to be matched with the current state; and the condition that unqualified pole pieces cut by the pole piece cutting equipment flow into the next process can be effectively reduced, so that the performance and the service life of the finally produced battery are improved.

The invention provides a communication method and a communication device. The method comprises the following steps: a first communication device compresses N pieces of second channel information based on X pieces of first channel information to obtain M pieces of compressed information, for example, the ith compressed information is obtained by compressing Pi pieces of second channel information based on the jth piece of first channel information, and the acquisition time of the first channel information is earlier than the acquisition time of the second channel information; the first communication device sends the ith compressed information, and the first communication device also sends the jth first channel information; the second communication device restores Pi pieces of second channel information based on the i-th piece of compression information and the j-th piece of first channel information that may be received at different times. Through the mode, the channel information (namely the second channel information) can be fed back in time, and the feedback overhead can be reduced.

The invention relates to the technical field of power monitoring, in particular to an online monitoring method and system for a power control cabinet. The method comprises the following steps: acquiring a time domain signal, tracking a fundamental wave frequency, adjusting a sampling window, and performing Fourier transform to obtain a harmonic amplitude and a phase; harmonic waves are divided into harmonic wave correlation groups according to association rules, harmonic wave phase consistency coefficients are calculated to weight the total energy in the groups, and weighted harmonic wave energy is obtained; total inter-harmonic energy is calculated, and an inter-harmonic structure factor is obtained based on the spectral entropy. And acquiring a real-time load rate and a change rate, and calculating a synchronous change coefficient and a load factor. And summing the weighted harmonic energy and the corrected inter-harmonic energy to obtain total distortion energy, and obtaining a comprehensive distortion index according to the total distortion energy. According to the scheme, the structured inter-harmonics possibly caused by the early failure of the equipment can be found earlier, and the requirements of early warning and predictive maintenance are met.

The invention relates to the technical field of multimode signal coverage, and discloses a multimode-based signal coverage amplification system and method. The system comprises a multi-source signal feature acquisition module, a real-time coverage situation generation module, a coverage deviation evaluation module, a resource scheduling module and a beam optimization execution module. The multi-source signal feature acquisition module acquires a multi-dimensional signal backtracking feature set containing first signal intensity distribution and first channel parameters at historical acquisition moments of a target area in a preset scanning period; a real-time coverage situation generation module obtains a real-time coverage situation map containing second signal intensity distribution and second channel parameters; a coverage deviation evaluation module constructs a coverage deviation matrix accordingly; a resource scheduling module retrieves a matched coverage blind area feature set, and when the dispersion reaches the standard, a resource release coefficient is calculated; and the beam optimization execution module executes coverage enhancement operation on the coverage resources according to the coefficient from the storage starting moment, so that the signal coverage quality and the resource utilization efficiency of the target area are effectively improved.

The invention provides a display device and a sound synthesis method, and the method comprises the steps: receiving sound data collected by a sound collector in response to an access instruction of the sound collector, and storing the sound data and corresponding collection time at a preset position; according to an input playing instruction, accompaniment audio data and original sound audio data of a target song are decoded, a start time identifier contained in the accompaniment audio data is detected, and after the start time identifier is detected, sound data are periodically read according to time corresponding to the start time identifier and collection time; when the sound data is read, generating target audio data according to the accompaniment audio data and the read sound data; and when the sound data is not read, generating target audio data according to the accompaniment audio data and the replacement audio data. According to the method, the missing sound data of the user can be filled in a mode of replacing the audio data, the integrity of the target audio is ensured, and the problem that the audio is incomplete in the audio recording process is solved.

The invention discloses a vehicle end data processing method and device, equipment and a storage medium, and relates to the technical field of vehicle end data processing, and the method comprises the steps: carrying out the data deduplication of an original CAN message of a target vehicle, and obtaining a to-be-analyzed CAN message; determining a current starting task based on the priority of the signal acquisition task; based on the rule information of the current starting task, determining a target vehicle signal in the analyzed vehicle signals corresponding to the CAN message to be analyzed, and collecting the target vehicle signal; and uploading the signal data acquired in the de-duplication acquisition time period to a cloud. Through the above mode, the received original CAN message is subjected to de-duplication, and the data acquisition time period is subjected to de-duplication before data uploading, so that repeated data uploading is avoided, repeated redundant data is reduced, vehicle end resource consumption is reduced, the priority is set between tasks, signal acquisition is performed according to the priority, the vehicle end is prevented from entering a high-load state, and the reliability of the vehicle end is improved. The treatment efficiency is improved.

The invention belongs to the field of wearable equipment, and particularly relates to a low-power-consumption wearable smart watch. Comprising an acquisition module, a processing module and a regulation and control module. The acquisition module is used for acquiring the signal strength of the current mobile network signal as first data according to a preset acquisition period, and associatively storing the acquisition time of the first data and the first data; the processing module is used for initializing the weakening times and weakening measurement, comparing the currently acquired first data with the first data acquired in the previous acquisition period, if the first data is weakened, increasing the weakening times, and accumulating the change value of the first data into the weakening measurement; if the first data becomes strong, whether the change value of the first data is larger than the weakening measurement or not is judged, if the change value of the first data is larger than the weakening measurement, the weakening frequency and the weakening measurement are initialized, and if the change value of the first data is not larger than the weakening measurement, the weakening measurement is started. According to the scheme, the problems of high power consumption and low user experience of the smart watch in a network environment with relatively weak signals are solved.

The invention discloses a carrier synchronization control method, which is applied to an energy storage converter and comprises the following steps: recording a first timestamp corresponding to a slave carrier wave starting point of a local machine; judging whether a first broadcast message sent by the host at the host carrier starting point is obtained within a first preset time or not; if not, determining that the host has a fault, and if the local meets a preset condition, switching the local to be a new host of the system; if so, determining that the acquisition time of the first broadcast message is a second timestamp; based on second preset time, the first timestamp and the second timestamp, calculating a time deviation between the carrier starting point of the local computer and the carrier starting point of the host computer; and adjusting the carrier starting point of the local machine to be synchronized with the carrier starting point of the host based on the time deviation. The carrier synchronization can be realized without adding extra equipment, and when the host fails, the slave can compete for the new host, so that the cost of the energy storage system can be reduced, and the stability of the system is improved.

The invention discloses a GIS insulation defect identification method and system, a terminal and a medium, and the method comprises the following steps: employing an adaptive pulse sequence threshold determination method based on a quartile distance, and carrying out the adaptive extraction of a plurality of GIS internal partial discharge pulse signals including HFCT, UHF and optical detection equipment; performing dimensionless energy binarization feature extraction on the HFCT, UHF and AE partial discharge signals within the fixed acquisition time to generate feature vectors; establishing a multi-scale insulation defect partial discharge real-time online incremental database, and automatically adjusting parameters of an incremental support vector machine model by adopting a Bayesian optimization algorithm; and inputting the feature vector into an SVM (Support Vector Machine) model subjected to Bayesian optimization of hyper-parameters, and carrying out multi-parameter partial discharge identification.

The invention discloses an intelligent bracelet integrated with photoelectric volume pulse wave detection and a data processing method thereof, and belongs to the field of intelligent bracelets.The intelligent bracelet comprises a host, a display screen is arranged on the upper surface of the host, an acquisition sensor is arranged on the lower surface of the host, and a control knob is arranged in front of the outer surface of the host; through the detachable structure of the watchband, the mounting groove and the inner groove are formed in the left end and the right end of the host and matched with the clamping block, the pinch bar, the spring and the button of the watchband to form a quick dismounting mechanism, the pinch bar can be driven to be separated from the clamping groove of the clamping block by pressing the button, the watchband can be dismounted by hands without tools, and the operation time is shortened to be within 3 seconds; in order to solve the problem that a traditional watchband is tedious to replace and easy to damage, the method comprises the following steps: in a signal preprocessing stage, combining band-pass filtering and adaptive Kalman filtering, and synchronously introducing motion data of an acceleration sensor and a PPG signal signal-to-noise ratio evaluation mechanism, when SNRlt; and the acquisition time is automatically prolonged or a user is prompted to stabilize the hand at 10dB.

The invention provides a sensor acquisition time synchronization method and system based on ZYNQ. A unified clock source is arranged in the FPGA, and it is guaranteed that no accumulated drift exists in the time difference. Hardware triggering is carried out by adopting an IMU signal as a reference, camera triggering at a precise moment is carried out according to the triggering moment in combination with factors such as exposure time and transmission delay of an image, and it is ensured that IMU data transmitted to a memory and a camera image are from the same real moment. Sensor data are calibrated through a software algorithm, and a result is fed back to the sensor trigger signal offset processing module to form a complete calibration loop. And closed-loop feedback is introduced, so that automatic and self-adaptive high-precision synchronization is realized. The IMU signal is used as a camera image acquisition triggering means, so that the image frame frequency and the IMU frequency are in a multiple relationship, and the parameters of the offset processing module can be modified according to the required frame frequency and the existing IMU frequency in the actual operation process to adjust the frame rate.

The invention relates to the technical field of power equipment monitoring, in particular to an intelligent electric energy meter operation state abnormity monitoring management method. Obtaining a fluctuation characteristic value according to local data at any moment in the load data sequence and the corresponding fitting data; filtering according to the fluctuation characteristic value to obtain a smooth load sequence; obtaining different load time periods according to the smooth load sequence, carrying out clustering, and obtaining a power utilization fluctuation value according to fluctuation characteristic values at all moments in a time period cluster; acquiring an acquisition time period according to the time period distribution characteristics of the time period cluster; the relative fluctuation degree is obtained according to the electricity utilization fluctuation values of the user and other users in the collection time period and the distance characteristics of the electric energy meter; and obtaining a frequency adjustment coefficient according to the load data of the user, the temperature sequence and the relative fluctuation degree. The monitoring data of the electric energy meter are adaptively acquired according to the frequency adjustment coefficient, and the monitoring accuracy of potential faults of the electric energy meter is improved while the storage pressure of the monitoring data is reduced.

The present application relates to an audio data processing system, method, device, electronic device and storage medium, and belongs to the field of terminal technology. The method includes: negotiating a target time with a first Bluetooth device; determining a time offset of the acquisition time of the acquired audio data relative to the target time; sending the audio data and the time offset to the first Bluetooth device, and the first Bluetooth device is connected to the second Bluetooth device via a Bluetooth communication link. Through this solution, the acquisition of audio data by multiple second Bluetooth devices is achieved, making the acquired audio data more three-dimensional.

The invention provides an echo signal processing method, an integrated circuit chip and an ultrasonic flowmeter, and is suitable for the technical field of echo signal processing. The method comprises the following steps: acquiring an echo signal, and controlling an analog-to-digital converter to acquire the echo signal based on an acquisition starting moment and a preset acquisition time interval of each acquisition period to obtain N pieces of acquisition data; based on a predetermined storage rule, storing the collected data of each collection period until the collected data of the M collection periods are stored to corresponding positions of the cache region; and arranging the collected data in sequence based on the positions of the collected data of the M collection periods in the cache region to obtain a reduced waveform. The echo signal processing method provided by the embodiment of the invention can be suitable for a built-in analog-to-digital converter to carry out data acquisition, and the problems that an external high-speed analog-to-digital converter has high requirements on circuit design and high cost, and signal interference exists in high-speed transmission are solved.

The invention discloses a high-frequency electrotome negative plate impedance detection method and system and a high-frequency electrotome, and the detection method comprises the steps: an impedance detection signal preprocessing step: determining an acquisition time interval of an impedance detection signal according to a current working mode and a corresponding optimal acquisition time interval; only the impedance detection signals in the acquisition time interval are collected; the acquisition time interval is one or more carrier periods in which the signal-to-noise ratio of the impedance detection signal is higher than the average value of the signal-to-noise ratio in a unit modulation period T under the current working mode; and the optimized impedance detection signal is filtered and amplified and then is input into an impedance calculation module to calculate and obtain a real-time impedance value of the negative plate of the high-frequency electrotome. Interference is filtered out by monitoring information such as power and impedance intervals when the high-frequency electrotome works and combining noise waveform characteristics of different working modes.

The invention provides a non-contact life signal processing method in a strong clutter environment, and belongs to the field of data processing.The method comprises the steps that extreme points in a signal data sequence of a user and first and last signal data are recorded as to-be-corrected data; obtaining an amplitude adjustment parameter and a slope adjustment parameter of each piece of to-be-corrected data according to a numerical value difference and an acquisition time difference between each piece of to-be-corrected data and the to-be-corrected data on the left side and the right side of the to-be-corrected data; and combining the numerical value of the to-be-corrected data to obtain a corrected value of the to-be-corrected data, thereby obtaining upper and lower envelope lines. The invention aims to solve the problem that the signal data of a user is processed by an EMD (empirical mode decomposition) algorithm due to the fact that an upper envelope line and a lower envelope line which are obtained through extreme points in a signal data sequence are greatly influenced by noise because the noise in an environment where the user is located has a great influence on the signal data of the user. And the accuracy of the obtained heart rate signal and respiration signal is low.

The present disclosure relates to a method, device, storage medium and terminal device for determining a wake-up delay, the method comprising: periodically collecting multiple audio data through a collection thread, and obtaining the collection time of the multiple audio data; when the wake-up engine receives the audio data, starting the wake-up thread, the wake-up thread is used to record the number of sampling points of the audio data received by the wake-up engine; when it is detected that the wake-up engine is in a wake-up state, determining the wake-up time of the wake-up engine; determining the target number of sampling points recorded by the wake-up thread when the wake-up engine wakes up, and determining the target collection time corresponding to the collection thread according to the target number of sampling points; determining the wake-up delay corresponding to the terminal device according to the target collection time and the wake-up time. In this way, the wake-up delay corresponding to the terminal device can be obtained according to the audio data collected in real time. Compared with the wake-up delay obtained by a special signal in a laboratory, the accuracy of the wake-up delay obtained by the present disclosure is higher.

The invention relates to the technical field of plastic pipe cutting, and provides a multi-angle band saw operation control method and system.The method comprises the steps that vibration data, current data and rotor angles at different collection moments are collected in the plastic pipe cutting process, and the target collection moment is marked; constructing a vibration sequence, a current sequence and a rotor angle sequence at a target acquisition moment; determining a cutting misalignment degree, a cutting misalignment degree sequence and a cutting misalignment degree predicted value sequence at the target acquisition moment, and determining a notch form offset degree at the target acquisition moment according to the difference between the cutting misalignment degree sequence at the target acquisition moment and the cutting misalignment degree predicted value sequence; and an integral gain adjustment value at the target acquisition moment is calculated, and real-time adjustment of operation of the multi-angle band saw for plastic pipe cutting is achieved in combination with the rotor angle at the target acquisition moment. According to the method, the integral gain is adaptively valued, and the cutting track offset is reduced.

The invention discloses a residual current real-time remote protection method based on edge calculation, which relates to the technical field of power failure protection, and comprises the following steps: executing time sorting and differential calculation based on residual current signal data acquired by current sensors deployed on power distribution branches and corresponding acquisition timestamps; generating a residual current signal record containing the time interval value and the amplitude difference value; extracting a signal arrival delay value and a time jitter value of each record according to the residual current signal record and a locally stored model version number, and fitting the two types of data to generate a time drift compensation parameter; performing version consistency comparison based on the time drift compensation parameter and a cloud model version number returned by the cloud interface; according to the method, a synchronous verification and arbitration updating mechanism of the local and cloud residual current anomaly judgment result is constructed in an edge computing environment, so that the consistency of the judgment result can still be kept under the conditions that the network is unstable and the model is frequently updated.

The invention relates to the technical field of image communication, in particular to an active tracking control method of a high-precision PTZ camera. The method comprises the following steps: acquiring a Rocker Cmd object; the speed values of all the shafts are judged based on a Rocker Cmd object, and corrected rocker input values are obtained; performing protocol mapping on the corrected rocker input value, and caching the corrected rocker input value to obtain an instruction queue; removing invalid stop instructions in the instruction queue to obtain an optimized continuous control instruction sequence; and performing acquisition time and direction judgment on the optimized continuous control instruction sequence to obtain a corrected control instruction set. Through intelligent processing, protocol mapping and caching, logic analysis and state feedback of rocker input signals, high-precision, smooth, reliable and durable closed-loop control of the PTZ camera is realized, and the control error-tolerant rate is remarkably improved.

The application discloses a method, device and equipment for identifying a movement time point of a surface electromyogram signal and a medium. The method comprises the following steps: acquiring a to-be-detected force platform signal corresponding to a to-be-detected surface electromyogram signal; wherein the acquisition time points of the to-be-detected surface electromyogram signal and the to-be-detected force platform signal are synchronous; determining a positive slope cumulative function according to the to-be-detected force platform signal, and performing first-order derivation on the positive slope cumulative function to obtain a positive slope change rate function; determining a movement time point of a movement event corresponding to the to-be-detected surface electromyogram signal according to the positive slope change rate function and a movement change rate threshold; wherein the independent variable and the dependent variable of the positive slope cumulative function are the acquisition time point of the to-be-detected force platform signal and the cumulative positive slope respectively, and the movement time point is a starting time point or an ending time point. The embodiment of the application solves the problem that the traditional identification method is susceptible to noise interference of the surface electromyogram signal, and improves the identification accuracy of the movement time point.

The present application relates to wireless communication signal processing technical field, specifically for wideband wireless signal time-frequency classification and positioning method based on graphic processor acceleration, including the following steps: obtaining the complex IQ sequence of receiver, association center frequency, actual sampling rate and collection time, obtaining frequency resolution unit width, determining frequency domain transformation point number and calling graphic processor transformation scheme, executing batch frequency domain transformation in graphic processor and generating time-frequency visual input, forming detection batch by time-frequency visual input of multiple center frequencies, obtaining candidate frame, category identification and confidence, screening candidate frame according to confidence and category identification, and outputting classification and positioning result according to candidate frame horizontal coordinate, image scaling parameter, center frequency and actual sampling rate.The present application keeps the corresponding relationship of category identification, frequency calculation and degradation path output by unified collection metadata, frequency domain processing, batch detection and coordinate back mapping.