5 results about "Signal Detection Analyses" patented technology

The invention discloses a complex signal detection analysis method and system based on a hybrid convolution window, and the method comprises the steps: carrying out the discrete sampling of a signal containing multiple complex harmonics in a power grid, and obtaining an original discrete sequence signal; processing the original discrete sequence signal by adding a mixed window to obtain a windowed discrete sequence signal; performing discrete Fourier transform on the windowed discrete sequence signal to obtain a discrete harmonic signal frequency spectrum value; obtaining a plurality of spectral lines which are at a preset distance from each harmonic frequency point in the discrete harmonic signal frequency spectrum value; according to a plurality of spectral lines having a preset distance from each harmonic frequency point, establishing a relationship between the spectral lines and a frequency spectrum value; obtaining an offset between the spectral line and a theoretical value through the relationship between the spectral line and the spectral value; and modifying the harmonic signal by using the offset to obtain a modified harmonic signal. And the civil harmonic detection precision is effectively improved.

The invention relates to the technical field of signal detection and analysis, and discloses a method for enhancing the generalization of an atrial fibrillation detection algorithm in combination with a GPT model, and the method comprises the following steps: S1, feature extraction: carrying out the multi-scale frequency domain decomposition of an input ECG signal through wavelet transform; a multi-head attention mechanism fused with Gaussian distribution is introduced, subspace feature expression is constructed, and multi-scale time-frequency features are obtained; s2, generalization enhancement is carried out, an atrial fibrillation model is constructed, dynamic sequence modeling and feature enhancement are carried out through a transfer learning mode and multi-scale time-frequency features, capturing of long-range dependence and tiny anomalies is achieved, fine tuning training is carried out in combination with atrial fibrillation detection task features, and a trained atrial fibrillation model is obtained; and S3, outputting a result, and outputting an atrial fibrillation detection result through the trained atrial fibrillation model. According to the electrocardiosignal atrial fibrillation detection strategy combining the large language model pre-training technology and the time sequence feature modeling method, the generalization ability and the detection precision of the model in diversified real scenes are improved.

The invention relates to a brain signal analysis method based on cross-frequency information fusion. The method is technically characterized by comprising the steps that a measurement normal form based on a magnetoencephalogram is established, and brain magnetic signals are collected; preprocessing the brain magnetic signals; calculating a weighted phase lag index between different channel signals in the preprocessed data, and constructing intra-layer connection of the cross-frequency multi-layer network; phase-amplitude coupling between channel signals under different frequency bands is calculated by adopting a modulation index measurement method, and interlayer connection of a frequency-crossing multi-layer network is constructed; and constructing a cross-frequency multi-layer brain function network according to the intra-layer connection and the inter-layer connection, and performing multi-dimensional feature extraction on the cross-frequency multi-layer brain function network. According to the method, the same-frequency and different-frequency connection characteristics are simultaneously integrated in the cross-frequency domain, so that the hierarchical organization structure of the brain function network can be more comprehensively described, the important topological characteristics of the multi-layer network can be stably extracted, the utilization efficiency of the cross-frequency coupling information in the brain magnetic signal is effectively improved, and the method can be widely applied to brain signal detection and analysis.

The invention discloses a portable precise hearing detection system based on adaptive sound field compensation, and relates to the field of hearing detection, and the system comprises an environment monitoring module which is used for obtaining an environment noise signal; the hearing monitoring module is used for acquiring an audio test signal and user feedback information; the data transmission module is used for setting an environment transmission channel protocol and a test transmission channel protocol corresponding to the data processing end for the environment end and the test detection end respectively, and transmitting corresponding data information; the environment compensation module is used for performing adaptive sound field compensation according to the environment noise signal to obtain an initial audio test signal; the test compensation module is used for performing adaptive compensation according to a comparison result corresponding to the initial audio test signal and the audio test signal to obtain a comprehensive audio test signal; the detection analysis module is used for performing analysis processing according to the user feedback information and the comprehensive audio test signal to obtain corresponding hearing detection data; the accuracy of the hearing detection process is improved.

The utility model relates to the field of embedded application technology discloses a kind of multi-channel logic signal detection analyzer system, comprising: multi-channel logic signal detection analyzer, multi-channel logic signal detection analysis work station and interface module;Multi-channel logic signal detection analyzer, interface module and multi-channel logic signal detection analysis work station are connected by integrated bus, and multi-channel logic signal detection analyzer is also connected to be detected equipment;Among them, the logic signal that the multi-channel logic signal detection analyzer is detected to equipment output is compressed to data, and the logic signal after processing is detected to obtain detection result, and detection result is sent to multi-channel logic signal detection analysis work station by integrated bus;Multi-channel logic signal detection analysis work station is used to display detection result. To solve the problem that channel expansibility is insufficient, and data storage and processing capacity are insufficient, and continuous long time signal analysis cannot be carried out.