1041 results about "Peak detection" patented technology

The invention relates to the field of hard disk data protection, in particular to a hard disk data protection method and device based on artificial intelligence. The method comprises the following steps: acquiring a hard disk history monitoring log; performing multi-period hard disk load analysis on the historical hard disk monitoring log, and performing load peak fluctuation fitting so as to construct a hard disk load peak fluctuation curve; carrying out peak value trend evolution analysis on the hard disk load peak value fluctuation curve, and carrying out abnormal load peak value detection to obtain a time period where an abnormal peak value point is located; carrying out current time period redundant data calculation on the hard disk according to the time period of the abnormal peak point, and carrying out dynamic global data backup decision, so as to generate an abnormal peak global backup strategy; and performing abnormal write behavior detection and attack node tracing positioning on the hard disk historical monitoring log, and marking a hostile attack tracing positioning point. According to the invention, safe and reliable hard disk data protection is realized.

The invention relates to an anti-deception-interference joint detection method and system based on space-time frequency, and the method comprises the steps: obtaining a signal source direction of a received satellite signal, and dividing the signal source direction into channels according to angles based on a beam forming criterion; based on a preset capture rule, initiating capture search for the signal in each channel, and synchronously performing correlation peak detection and array antenna anti-interference detection on the signal to obtain angle information of the signal in each channel and satellite PRN information in the capture process; null processing is carried out on the signal source directions of two or more satellite signals appearing at the same angle; outputting a capture result based on the capture rule, and tracking the capture result conforming to the capture rule; doppler frequency shift consistency detection and signal power calculation are carried out on the signals subjected to rotation tracking, and deception signals are recognized and fed back to the antenna array for filtering processing. The application has the effects of expanding the application range of the anti-deception jamming detection method and improving the anti-jamming capability of navigation positioning.

The invention relates to an ultrasonic flow metering device and a metering method thereof, and relates to the field of ultrasonic measuring devices. According to the metering device, through a cross-correlation operation circuit, a microcontroller, a double-threshold comparator, a peak detector, a gain adjustment circuit and a head wave threshold adjustment circuit, self-adaptive processing and adjustment of echo signals and head wave thresholds are achieved, and the accuracy and reliability of flow metering are improved. The metering method adopts an adaptive adjustment strategy of echo signal amplifier gain based on an output signal of an echo signal peak detection circuit, and a head wave threshold adaptive adjustment strategy based on a correlation operation value of echo signal sampling data and pre-stored reference wave data. According to the scheme, the stability of the echo signal and the rationality of the head wave threshold value are guaranteed, the method has adaptability to echo contour feature changes caused by transducer aging or contamination, and then the reliability of the flow metering process is guaranteed.

The invention relates to the technical field of power equipment state detection, in particular to an ultrahigh-frequency partial discharge online detection system, method and device and a medium, and the system comprises the steps: collecting an initial discharge signal in real time through an ultrahigh-frequency sensor array, and carrying out the preprocessing of the initial discharge signal, so as to obtain an ultrahigh-frequency discharge signal; carrying out peak detection on the ultrahigh-frequency discharge signal, and triggering a high-speed analog-to-digital converter to collect an original waveform when the amplitude exceeds a preset threshold value; performing multi-dimensional feature extraction on the original waveform by using a digital signal processor to obtain multiple groups of dimensional features; identifying and classifying the multiple groups of dimension features based on a random forest algorithm, generating discharge type labels and confidence coefficients, and storing the discharge type labels and the confidence coefficients in a dynamic database; carrying out spatial position calculation on the ultrahigh-frequency discharge signal by adopting a time difference method, and determining a three-dimensional coordinate of a discharge source; the dynamic database and the three-dimensional coordinates of the discharge source are subjected to space-time correlation and multi-dimensional analysis, a defect analysis result is generated, and high-precision online detection of the partial discharge defect of the high-voltage equipment is achieved.

System and method for determining valid heartbeats includes an external device and implantable pressure sensor (IPS) that senses pressure for a period of time and generates a pressure signal. The pressure signal is segmented into pseudo-systolic and pseudo-diastolic segments, and it is determined whether the pressure signal starts with one of the pseudo-systolic or pseudo-diastolic segments. In response to the pressure signal starting with a pseudo-diastolic segment, systolic content of at least one of the pseudo-systolic segments is modified based on diastolic content of at least one of the pseudo-diastolic segments. In response to the pressure signal starting with one of the pseudo-systolic segments, the systolic content of at least one of the pseudo-systolic segments is modified based on the systolic content of at least one of the pseudo-systolic segments, and valid heartbeats are determined based on the modified systolic content.

The invention relates to the technical field of fire control, in particular to an electrical fire control method and system based on multiple sensors, and the method comprises the steps: obtaining related data of electrical fire influence factors, and carrying out the preprocessing; extracting a current peak value by using an AMPD peak value detection algorithm according to the preprocessed current data, and constructing an arc fault strength index; combining the arc fault strength index and temperature data, analyzing the relationship between arc fault abnormal heat release and temperature change, constructing an electrical heat dissipation goodness, integrating the electrical heat dissipation goodness and the change rule of carbon monoxide and carbon dioxide concentration, constructing a fire risk index, and quantifying the risk degree of electrical fire occurrence; and correcting the SOS anomaly detection algorithm based on the fire risk index to obtain a corrected significant anomaly score, and preventing and controlling the electrical fire according to the significant anomaly score. According to the invention, the fire risk index is constructed in combination with various sensor data, and the electrical fire early warning accuracy and prevention and control efficiency are improved.

The invention relates to a direct current bias suppression device and method for a dual-active bridge and a dual-active bridge system, and the device comprises a high-frequency signal detection module which receives the primary side current and the secondary side current of a high-frequency transformer in the dual-active bridge, converts the primary side current and the secondary side current into voltage signals, and outputs the voltage signals to a control module; obtaining a primary side current detection signal and a secondary side current detection signal; the peak value detection module is used for detecting the maximum value and the minimum value of the primary side current in the primary side current detection signal and the maximum value and the minimum value of the secondary side current in the secondary side current detection signal; the sampling module is used for sampling the maximum value and the minimum value of the primary side current and the maximum value and the minimum value of the secondary side current; the control module is used for calculating the difference value between the sum value of the maximum value and the minimum value of the primary side current and zero and the difference value between the sum value of the maximum value and the minimum value of the secondary side current and zero, and obtaining the duty ratio adjustment amount of the primary side and the duty ratio adjustment amount of the secondary side by using the PI controller; and the direct-current bias current on the two sides of the dual-active bridge is suppressed according to the duty ratio adjustment amount.

The invention relates to the technical field of valve leakage detection, and particularly discloses an ultrahigh pressure gas valve leakage detection method and system. According to the method, a unified leakage feature vector is obtained through multi-source acquisition and processing; capturing a time sequence dependency relationship, and determining a micro leakage mode; carrying out abnormal peak detection to obtain a preliminary leakage probability; combining the initial leakage probability with real-time environmental parameters, and dynamically adjusting a judgment standard; comparing the unified leakage feature vectors through a judgment standard, fusing position coordinate triangulation positioning, judging whether micro crack leakage occurs or not, and obtaining leakage position estimation; the historical data sequence is fused, the micro crack development trend is predicted, and the early warning level is determined; and generating a corresponding response signal according to the early warning level, and outputting micro crack leakage early warning. Multi-source data fusion, dynamic threshold adjustment and position prediction can be carried out, the accuracy and robustness of micro crack leakage detection are remarkably improved, and efficient guarantee is provided for safe operation of a pipeline.

The invention relates to the technical field of digital control switching power supplies, and particularly discloses a switching power supply dynamic response adjusting system based on digital control, which comprises a digital controller, a multi-channel analog-to-digital conversion module, a parameter identification module, a self-adaptive control law adjusting module, a pulse width modulation driving module and a stability monitoring module. On-line identification of key parameters of a system is realized by constructing a nonlinear state space model and combining an extended Kalman filtering algorithm, dynamic compensation is carried out for non-minimum phase characteristics such as a right half plane zero point, and a self-adaptive control law adjustment module optimizes a control strategy in real time according to an updated model. A Lyapunov stability criterion and a predictive control mechanism are introduced, the robustness and response speed of the system are improved, and a stability monitoring module identifies an abnormal state through sliding window variance analysis and peak detection and triggers a fault-tolerant mechanism to guarantee safe operation of the system.

The invention relates to the technical field of multi-channel audio synchronization, and discloses a multi-channel audio synchronization method and device for conference terminal equipment, equipment and a storage medium, and the method comprises the steps: collecting an audio data stream of each conference terminal equipment, and obtaining an embedded audio data stream; performing time-frequency domain conversion and peak detection to obtain an audio feature vector; calculating a cross-correlation function between the terminal devices to obtain a global time relation graph; constructing a two-layer cache structure, and adjusting the residence time of the audio data in the cache to obtain a buffer strategy; inputting the buffer strategy as a state into a multi-agent reinforcement learning model for communication resource allocation to obtain a resource allocation strategy; a global synchronization optimization problem is constructed based on a resource allocation strategy, an optimal synchronization parameter is solved through a gradient descent algorithm, time axis adjustment and content alignment are carried out on an embedded audio data stream, and a plurality of paths of synchronized audios are obtained, so that the audio quality and user experience of a multi-party teleconference are improved.

The invention relates to the technical field of high-precision anti-interference satellite navigation, and particularly discloses a detection and suppression method for suppressing deception combination interference of high-precision anti-interference navigation. One or at least two of space-frequency joint-multi-beam / zeroing-noise subspace projection, correlation peak detection and screening, true convex false deception jamming suppression detection, deception jamming recovery and cancellation, improved PAC and dense suppression / isolated suppression deception / dense suppression deception jamming subspace rapid solving methods are adopted for deep fusion; detection and suppression of combined interference are realized. According to the method, the problems of detection and suppression of combined interference of isolated suppression and weak cheating, dense suppression and weak cheating, isolated suppression and stronger cheating, dense suppression and stronger cheating, isolated suppression and strong cheating, dense suppression and strong cheating and the like are solved.

The invention belongs to the technical field of marine surveying and mapping, and discloses an airborne marine laser radar underwater weak echo signal detection method and system. According to the method, a waveform rising edge is determined by constructing a standard deviation curve, a water surface echo moment is accurately extracted, and two water surface geometric elements of a water surface point and a refraction vector are calculated on the basis; constructing a time-space constrained waveform superposition model, taking the Euclidean distance between the two water surface geometric elements as a neighbor discrimination standard, and screening and superposing neighbor waveforms by using a K neighbor judgment method to obtain a superposed pseudo waveform with a high signal-to-noise ratio; and finally, an improved normalized saliency peak detection method is utilized to judge the underwater echo position step by step and obtain the final underwater echo moment. After the model constructed by the invention is used for processing, the underwater weak echo detection of the airborne ocean laser radar is effectively realized, the maximum sounding performance of the airborne ocean laser radar is improved, and an effective technical support and a solution are provided for sounding data processing and application.

The invention discloses an ion chromatography wave crest boundary extraction method based on fitting residual and tangent judgment, and belongs to the technical field of ion chromatography analysis. Comprising the following steps: S1, reading an original waveform data sequence, and executing frequency reduction sampling according to a set step length; s2, calculating the slope of the data after underclocking sampling by using a difference method, and searching positive and negative slope sections by performing first-order derivation on the waveform after underclocking sampling; s3, noise interference is eliminated, slope areas are grouped, then a chromatographic peak detection limit value is set, and an effective peak left section and an effective peak right section are screened out; s4, combining the effective wave crest left side section and the effective wave crest right side section to obtain an effective wave crest and a preliminary left boundary point and a preliminary right boundary point of the effective wave crest; s5, fitting a straight line through the preliminary left boundary point and the preliminary right boundary point to position an error integral region; and S6, updating the optimal boundary point by traversing the evaluation indexes of all points in the error integral region. The boundary search range is more accurate, and the calculation efficiency is higher.

There is described an artificial intelligence wearable ECG skin patch (400) to detect sudden cardiac arrest. The wearable ECG monitoring patch (400) with AI based predictive analytics and remote based cardiac monitoring (615) system that can detect cardiac arrhythmias automatically in real-time and make a diagnosis with AI models trained with acquired data. The wearable skin has a biocompatible polymer patch (400) which captures the electrical signal through a flexible printed electronic technology based conducting ink and a substrate. The microcontroller controls (201), store and transmit the data packets. The IoT connected signal transmission is capable of recording and transferring the data packets through wireless communication. The AI engine is capable of analysing, evaluating, testing and providing the data packets of sudden cardiac arrest through a peak detector algorithm. The ECG skin patch (400) to detect and measure the sudden cardiac arrest with the R-R interval time series to obtain heart rate variability.

The invention discloses a random access preamble and detection algorithm design based on low earth orbit satellite communication. Specifically, firstly, a novel lead code with a plurality of subsequences is provided, in order to enlarge the number of available lead codes in a wave beam, a ZCM sequence is introduced to form the subsequences, and orthogonal frequency division multiplexing is adopted for modulation and is sent to a channel. Subsequently, a timing advance (TA) detection algorithm is proposed. The first sub-sequence is used for decimal time delay detection, and the other sub-sequences are used for integer time delay detection. In addition, the peak detection ingeniously utilizes the variance of the power delay spectrum, so that the detection performance is improved. Finally, a simulation result shows that the lead code provided by the invention is almost not influenced by frequency offset, does not need pre-compensation, and is suitable for being applied to low-orbit satellite communication with relatively large frequency offset.

The invention belongs to the technical field of data processing, and particularly relates to a multi-dimensional holographic data analysis method and system for hazardous waste supervision, and the method comprises the steps: obtaining fluorescence spectrum data and mass spectrum data; preprocessing the acquired data, and carrying out peak detection by utilizing a first-order derivative of peak intensity based on a preprocessed fluorescence intensity matrix; determining the weight of each peak point in combination with the fluorescence spectrum of the standard sample to complete the matching of peak pairs, and calculating the peak matching distance and the distribution difference of the feature vectors; calculating a similarity score of the spectrum by integrating the peak matching distance and the distribution difference of the feature vectors and combining a matching ratio and a weight ratio; calculating weighted cosine similarity and Euclidean distance between the processed mass spectrum data and the mass spectrum data of the standard sample, and completing calculation of the similarity score of the mass spectrum between the samples; and tracing the source of the hazardous wastes according to the similarity score of the spectrum and the similarity score of the mass spectrum. And accurate traceability is realized through qualitative and quantitative analysis.

The invention discloses a non-contact drunk driving detection method and system adopting a large-linewidth tunable laser, and relates to the technical field of laser detection, and the method comprises the steps: collecting transmission light signals and corresponding electric signals under different linewidth conditions through experiments, synchronously recording experiment parameters, obtaining an original data set, carrying out the preprocessing of the original data set, and carrying out the detection of drunk driving. Generating a training set, a verification set and a test set, and training a BP (Back Propagation) neural network to obtain a pre-trained noise suppression model; the method comprises the following steps of: generating a wavelength-tunable modulated laser beam by adopting a large-linewidth tunable laser, directionally projecting the modulated laser beam into a vehicle cab, and enabling laser to penetrate through the cab space and carry alcohol gas concentration information to obtain a transmission light signal; focusing the transmission light signal to a photoelectric detector through an optical convergence device to obtain a noisy electric signal, inputting the noisy electric signal into a pre-trained noise suppression model, and outputting a de-noised electric signal; performing synchronous demodulation and harmonic analysis on the de-noised electric signal by using a phase-locked amplification detection technology, extracting a second harmonic signal amplitude associated with the modulation frequency, and determining a second harmonic signal characteristic amplitude through peak detection; according to the calibration relation between the second harmonic signal characteristic amplitude and the alcohol gas concentration, the concentration value of the alcohol gas in the cab is calculated in an inversion mode through a least square fitting algorithm, an alcohol concentration threshold value is set, the drunk driving condition of a driver is judged, a detection result is generated, and an alarm signal is output.

Breathing sensors, sensing methods, and vehicle occupant monitoring systems may employ ultrasonic transducers. One illustrative sensing method includes: sensing an acoustic transducer's responses to a series of acoustic bursts emitted in a vehicle's cabin; differencing pairs of said responses to obtain difference signals; converting the difference signals into magnitude signals; performing peak detection on each of the magnitude signals to obtain at least one peak amplitude; and providing a breathing signal based on the at least one peak amplitude. The breathing signal may be an occupancy detection flag, an occupant position, a breathing frequency, and / or any combination thereof.

The invention relates to the technical field of medical data processing, and provides a medical image data processing method based on deep learning, which comprises the following steps of: acquiring data according to an acquisition template, extracting a pulse time sequence by self-adaptive threshold peak detection, calculating an instantaneous phase according to linear interpolation, calculating a statistical magnitude, comparing a quantitative index with a preset threshold value, and calculating a pulse time sequence according to the statistical magnitude. Judging a steady state by combining a peak loss rate and an abrupt change detection rule, and calculating phase consistency between channels for verification; the method comprises the following steps of: splitting acquired data according to a concept entity to form a data relation model, implementing rapid global rigid estimation and applying affine transformation, estimating a pixel-level displacement field by adopting a pyramid dense optical flow network, applying the displacement field to an original pixel, and performing time domain fusion by taking optical flow confidence and a registration residual error as weights; and cutting the short-time image stabilization sequence after registration compensation, and outputting a pixel-level risk thermodynamic diagram, a candidate focus list and each output confidence interval by taking a hybrid network of a convolution front end and a space-time Transform backbone as a prediction model.

The invention discloses a channel-configurable liquid particle pulse amplitude detection method and a related device, and relates to the field of digital signal processing, and the method comprises the following steps: eliminating a background noise reference value of a digital pulse signal, presetting a dynamic threshold value, filtering a noise signal of which the amplitude value is lower than the dynamic threshold value in the digital pulse signal, and obtaining a channel-configurable liquid particle pulse amplitude detection result; obtaining a baseline correction digital pulse signal; when the amplitude of the baseline correction digital pulse signal exceeds a dynamic threshold for the first time, pulse counting is started, peak detection is carried out on the baseline correction digital pulse signal, and a pulse peak value and a peak value position corresponding to a pulse event are obtained; a channel threshold parameter is preset through an upper computer, a plurality of independent channels are divided, pulse events are loaded into the corresponding independent channels to be accumulated, a channel count value and corresponding particle size distribution data are obtained, and liquid particle pulse amplitude detection is completed. The problem that the number of channels and threshold parameters cannot be dynamically adjusted when a current hardware circuit is used for liquid particle pulse amplitude detection is solved.

A flyback-converter with synchronous-rectifier (SR) sense architecture is provided. A secondary side controller includes a SR-sense pin coupled through an external resistor to a drain of an SR on the secondary-side, a negative-sensing-detector, a peak-detector, a zero-crossing-detector, all coupled to the pin, and a resistor network (Rn) coupled between the pin and ground. The Rn includes a first resistor (R1) to couple the pin and to ground through a first switch (S1) during negative-sensing to divide a voltage (VSR_drain) coupled to the pin, and a second, higher resistance resistor (R2) to couple the pin to ground through a second switch (S2) during peak-detection to divide VSR_draincoupled to the pin. S1 and S2 are controlled by register-transfer-level circuit in the SSC. A line-feed-forward (LFF) circuit is coupled to the pin through an active diode to receive an undivided VSR_drain and mirrors diode current to control the converter in LFF mode.

The invention discloses a chip interface safety protection control circuit, which relates to the technical field of chip interface protection, and comprises a power supply control module which is used for carrying out rectification filtering and voltage stabilization processing on accessed electric energy, meeting power supply requirements, carrying out overcurrent detection, constant current regulation and overcurrent power-off protection control, and carrying out power-off protection on the power supply control module during power-off protection. The interface voltage of the chip module is lowered, the interface sampling module can carry out sampling and peak detection on pulse signals accessed or output by the chip module, the state detection module can judge the type of the pulse signals, namely square wave pulses or sine pulses, and the first sampling module and the second sampling module carry out peak sampling according to the type of the pulse signals. The micro-control module controls the threshold value setting module to set a comparison threshold value, then the abnormity detection module detects the peak value state of the pulse signal, and when the peak value state is larger than the normal peak value state, chip interface protection is carried out. The chip interface safety protection control circuit can automatically carry out interface signal detection and interface protection control.

The invention discloses a voice interactive chart dynamic generation method and system based on a large language model, and relates to the technical field of chart generation, and the method comprises the steps: collecting a user voice signal, carrying out the preliminary framing, calculating the short-time energy, and analyzing a mute segment set signal; frequency domain signals are extracted from the set signals, gain is calculated through a filter, noise reduction is carried out on the frequency domain signals, the zero-crossing rate and noise reduction short-time energy are calculated, effective voice frames are screened, peak detection is carried out, the average voice speed is calculated, voice speed normalization is carried out, and signal values of sampling points in the frames are extracted. According to the method, the noise power spectrum density is extracted through FFT on the mute section, a frequency domain model of background noise is effectively established, a follow-up filter can accurately act on an actually existing frequency band interference area, weakening of the voice main signal spectrum is avoided, the Mel filter bank is accessed after the power spectrum is calculated through the FFT after frame windowing, and the noise power spectrum density is calculated through the FFT after frame windowing. Energy can be redistributed on the logarithmic Mel scale according to human ear perception characteristics.

The invention provides a target angle measurement method of a sparse antenna array, a vehicle-mounted millimeter wave radar and a product, the sparse antenna array comprises n unit antenna groups which are arranged repeatedly, each unit antenna group comprises m unit antennas, and the distance between the unit antenna groups is not equal to the distance between the adjacent unit antennas in the unit antenna groups. Comprising the following steps: determining unit antenna sub-arrays corresponding to unit antennas respectively based on the unit antennas contained in each unit antenna group in a sparse antenna array; acquiring signals respectively received by each unit antenna sub-array, processing the signals to obtain a first angle spectrum corresponding to each unit antenna sub-array, and performing non-coherent accumulation on the first angle spectrum of each unit antenna sub-array to obtain a second angle spectrum; performing angle peak detection on the second angle spectrum to obtain a potential angle set; and determining a real target angle from the potential angle set according to the angle spectrum data of the unit antenna subarrays corresponding to the potential angles in the potential angle set.

The utility model discloses a low-cost laser pulse signal peak detection circuit, which relates to the field of detection circuits, and comprises a digital logic control part, an analog part and a digital part, the analog part comprises a plurality of subunits, each subunit comprises a peak detection circuit part and a low-speed ADC (Analog to Digital Converter) part, each subunit is connected with the digital part through a main switch WSEL, and the peak detection circuit part is composed of an error amplifier sharing a differential input tube, a peak level sampling read-out buffer, a holding capacitor Ch, a plurality of subswitches, a bias current source, a reset voltage source V1 and an output buffer drive circuit. According to the utility model, peak value measurement and quantification of narrow pulse width echo signals are realized by combining the peak value holding circuit and the low-speed ADC circuit, and the cost area and the power consumption of a laser ranging system are reduced.

The invention discloses a digital gain adjustment control circuit which comprises a power detector, a digital peak detector, a gain control module and a digital gain adjustment module. The power detector calculates the power of an input signal by adopting a smooth root-mean-square algorithm, and rapidly adjusts a gain coefficient according to different step lengths adapted to regions; the digital peak value detector adopts a sliding window maximum value method to detect the peak value of an input signal, carries out statistical counting on the overload and underload states of the peak value based on a two-stage method, and judges and adjusts the gain according to a comparison result with a threshold value, so that the influence of frequent gain refreshing on the AGC loop state is avoided, and the loop stabilization time is shortened. Flexible algorithm upgrading and parameter configuration are supported by adopting a full-digital framework, different communication standards are adapted through firmware updating, the limitation of single index control is solved through a dual-detection cooperative working mode, peak overshoot is effectively restrained while the average power of the system is kept stable, the system is conveniently integrated to an SoC chip through modular design, and the reliability of the system is improved. And the power consumption and the size are greatly reduced.

The invention relates to a chromatographic peak detection method and device, computer equipment, a storage medium and a computer program product, and relates to the technical field of chromatographic analysis, the method in one embodiment comprises the following steps: obtaining chromatographic detection data, the chromatographic detection data comprising reaction time and corresponding signal intensity; traversing the chromatographic detection data, and identifying chromatographic peaks in the chromatographic detection data; based on the preset peak appearance time of the first component, determining a first component chromatographic peak corresponding to the determined first component in the chromatographic peaks; determining the appearance time of a second component associated with the first component based on the peak value time of the chromatographic peak of the first component and the relationship between the appearance time of the first component and the appearance time of the second component; and determining a second component chromatographic peak corresponding to the second component in the chromatographic peaks based on the appearance time of the second component. By adopting the scheme of the embodiment, the efficiency of determining the components of the chromatographic peak in the chromatographic peak detection process is improved.

The invention discloses an LC-MS peak detection method, device, equipment and medium, and the method comprises the steps: firstly obtaining an original data set of an environment sample collected by an LC-MS instrument, extracting and adjusting peak detection parameters, and determining a plurality of peak regions and characteristic parameters thereof; generating a synthetic peak region and synthetic characteristic parameters which are consistent with the real peak region in characteristic parameter distribution by using a conditional generative adversarial network, and then constructing an LC-MS peak detection model comprising a peak region classification model and a peak boundary defining model through a lightweight multi-task convolutional neural network based on the characteristic parameters and the synthetic characteristic parameters; and finally, inputting to-be-detected sample data into the model to obtain a detection result. According to the method, peak detection parameters are directly extracted from original data and dynamically adjusted, classification of peak regions and peak boundary integration are synchronously completed in combination with the improved lightweight multi-task convolutional network, the real-time detection requirement is met, and finally high-precision detection of real positive peaks is achieved and the noise amount is remarkably reduced.

The invention provides a digital receiver circuit and a control method thereof. The digital receiver circuit comprises a frame synchronization detection module, a symbol synchronization detection module and a symbol decoding module, the frame synchronization detection module comprises a first matched filter and a first peak detection module; the first matched filter is used for performing convolution operation on a guide code sequence in the received synchronization signal to obtain a convolution operation result; the first peak value detection module is used for determining the position of a convolution peak value according to the convolution operation result; determining the end position of the guide code sequence according to the position of the convolution peak value; the symbol synchronization detection module is used for detecting positioning information of an end position of each symbol in effective data connected with the guide code sequence; according to the positioning information, determining a pulse sequence corresponding to a judgment moment of symbol decoding of the effective data; and the symbol decoding module is used for performing symbol decoding on the effective data according to the judgment time and the pulse sequence. The technology can improve the accuracy of signal detection and decoding.

The invention discloses a chromatographic data optimization processing method for a liquid chromatograph, and relates to the technical field, and the method comprises the following steps: collecting an original signal flow, carrying out preliminary peak detection, and generating an original peak list; extracting modal exclusive characteristics of each chromatographic peak; converting the modal exclusive features into a unified multi-dimensional semantic feature vector; calculating a time interval weight, and calculating a semantic relevancy weight based on the multi-dimensional semantic feature vector; correcting the time interval weight, generating a network edge weight, and constructing a peak correlation topology network by taking the multi-dimensional semantic feature vector as a node; identifying node clusters connected with high edge weights, and aggregating the node clusters into candidate compound entities; tracking a node signal intensity change track, and adding a conflict mark; conflict resolution arbitration is carried out on the candidate compound entities with the added conflict marks, dominant detector evidence is output, and an analysis report is generated. According to the method, the problems of poor data collaboration of multiple detectors, inaccurate peak identification and association and isolated data processing flow are effectively solved.