Abnormality detection method based on time domain joint features

Abstract

The invention discloses an anomaly detection method based on time-domain joint features, which comprises the following steps: sampling different types of signals, extracting time-domain features including extreme values, rise time, fall time and pulse width, forming a comparison set for abnormal signal detection, then identifying a to-be-detected signal, and determining whether the to-be-detected signal is abnormal or not. The abnormal signal detection method based on the time domain joint features is realized.

Description

Technical Field

[0001] The present invention belongs to the technical field of digital oscilloscopes. More specifically, it relates to an anomaly detection method based on joint time-domain features. Background Art

[0002] With the continuous development of electronic information technology, the complexity of signals and test requirements are constantly increasing, and the requirements for the diversity triggering and abnormal signal detection capabilities of oscilloscopes are becoming more and more stringent. Triggering, as a basic function in digital oscilloscopes for capturing waveforms of interest to users and enabling complete and stable display of waveforms, is an indispensable part of digital oscilloscopes. The traditional triggering function directly sends signals into the trigger module, which has the disadvantages of single triggering characteristics, overly cumbersome triggering, and lack of anomaly detection function. In the face of environmental noise, measurement errors, and the diversity of complex dynamic systems, the detection ability of the single-feature triggering mechanism is limited, making it difficult to meet diverse application requirements and unable to accurately and effectively detect the presence of abnormal signals from time-domain signals.

[0003] Traditional triggering methods mainly include edge triggering, level triggering, pulse-width triggering, etc. These methods only focus on certain aspects of signal characteristics, such as the rising edge or the threshold value of signal amplitude, ignoring the information of the overall dynamic changes of the signal, and are prone to "turn a blind eye" to complex abnormal signals. Moreover, some abnormal signals may exhibit multiple characteristics simultaneously, such as the simultaneous occurrence of pulse-width anomalies and extreme value overrun. Single triggering conditions are prone to false triggering and cannot effectively capture these abnormal signals.

[0004] At the same time, traditional triggering methods have high requirements for hardware circuit design. Signal noise and trigger edge distortion will bring false triggering effects, making it difficult to capture abnormal signals. When detecting multiple types of abnormal signals, it is usually necessary to set different triggering conditions one by one, resulting in frequent adjustment of trigger configurations, low efficiency, and complex operations.

[0005] In summary, there is no effective method to effectively detect abnormal signals. Therefore, it is particularly important to design an anomaly detection method applicable to oscilloscopes, which can improve the ability of digital oscilloscopes to capture abnormal signals in different scenarios to meet the increasing signal detection requirements. Summary of the Invention

[0006] The purpose of the present invention is to overcome the deficiencies of the prior art and provide an anomaly detection method based on joint time-domain features, which constructs triggering conditions through multi-dimensional time-domain features, thereby improving the ability of digital oscilloscopes in abnormal signal detection.

[0007] To achieve the above invention purpose, an anomaly detection method based on joint time-domain features of the present invention is characterized by including the following steps:

[0008] (1) Deploy a trigger detection module in the digital oscilloscope. The trigger detection module specifically includes: an extreme value trigger detection module, a pulse width trigger detection module, and a rise time and fall time trigger detection module;

[0009] (2) Signal acquisition;

[0010] Input different types of input signals into the digital oscilloscope in sequence, and use the digital oscilloscope to acquire the input signals. Each type of input signal is acquired in N segments. The j-th segment of the i-th type of input signal is denoted as X ij , X ij = {x ij1 , x ij2 , …, x ijk , …, x ijM}, where x ijk represents the k sampling points in X ij . k is the index of the sampling point number, and M is the number of sampling points;

[0011] Record the steady-state high level ij and steady-state low level of each segment of the sampled signal X and the sampling interval time Δt of each sampling point;

[0012] (3) Perform multi-dimensional time-domain feature extraction on each segment of the sampled signal X ij ;

[0013] (3.1) Extract extreme value features;

[0014] Traverse the sampling points in each segment of the sampled signal X ij to find all sampling points x ijk-1 whose amplitudes satisfy V ijk > V ijk+1 and store them in the maximum value sequence ijk ; find all sampling points whose amplitudes satisfy V ijk-1 > V ijk < V ijk+1 and store them in the minimum value sequence ; then calculate the extreme value interval range of each segment of the sampled signal:

[0015] Maximum value interval:

[0016] Minimum value interval:

[0017] (3.2) Extract rise time and fall time features;

[0018]

[0018] (3.2.1) Calculate the rise time where, is the moment of low level, is the voltage moment when the low level rises by 10%, is the voltage moment when the low level rises by 90%;

[0019] (3.2.2), Calculate the fall time Among them, is the moment of high level, is the voltage moment when the high level drops by 10%, is the voltage moment when the high level drops by 90%;

[0020] (3.2.3), For each segment of the sampled signal X ij Repeat the above operations to obtain N groups of rise times and fall times

[0021] (3.2.4), Extract the rise time and fall time range characteristics of the i-th type of input signal;

[0022] Rise time range characteristic:

[0023] Fall time range characteristic:

[0024] (3.3), Extract the positive pulse width and negative pulse width characteristics;

[0025] (3.3.1), Set the threshold value of each segment of the sampled signal X ij

[0026]

[0027] (3.3.2), Traverse the sampling points in each segment of the sampled signal X ij and find the sampling points that satisfy , and then mark these sampling points as zero-crossing points;

[0028] (3.3.3), Calculate the pulse width between two adjacent zero-crossing points;

[0029] If the amplitude V ij of the l-th zero-crossing point in the sampled signal X ijl satisfies: then calculate the positive pulse width

[0030]

[0031] Among them,​ denotes the sampling point number of the l-th zero crossing in the sampling signal X ij ;

[0032] If the amplitude V ij of the l-th zero crossing in the sampling signal X ijl satisfies: then calculate the negative pulse width between the l-th zero crossing and the (l + 1)-th zero crossing

[0033]

[0034] (3.3.4) Obtain the positive pulse width range and negative pulse width range of the sampling signal X ij :

[0035] Positive pulse width range:

[0036] Negative pulse width range:

[0037] (4) Store the multi-dimensional time-domain features as template comparison parameters in the trigger detection module. Among them, store the extreme value interval range as the extreme value template comparison parameter in the extreme value trigger detection module; store the rise time and fall time range features as the rise time and fall time template comparison parameters in the rise time and fall time trigger detection module; store the positive and negative pulse width ranges as the pulse width template comparison parameters in the pulse width trigger detection module;

[0038] (5) Input the signal to be measured into the digital oscilloscope, collect the sampling signal containing M sampling points, and then extract the characteristic values of the sampling signal, including the maximum value V max , the minimum value V min , the rise time T rise , the fall time T fall , the positive pulse width W pos , the negative pulse width W neg ; Send the extracted characteristic values to the trigger detection module for template comparison. If all the characteristic values are within the range of the template comparison parameters, it is determined that the signal to be measured is normal; otherwise, it is determined that the signal to be measured is abnormal.

[0039] The invention object of the present invention is realized as follows:

[0040] The abnormal detection method based on time-domain joint features of the present invention first samples different types of signals, extracts their time-domain features including extreme values, rise time, fall time, and pulse width, forms a comparison set for abnormal signal detection, and then identifies the signal to be measured to realize the abnormal signal detection method based on time-domain joint features.

[0041] Meanwhile, the anomaly detection method based on time-domain joint features of the present invention also has the following beneficial effects:

[0042] (1) Joint triggering of multi-dimensional features: By combining multiple time-domain features such as pulse width, extreme value, rise time, and fall time, and setting joint triggering conditions, the dynamic characteristics of signals can be described more comprehensively, improving the ability to capture abnormal signals.

[0043] (2) Enhanced anomaly detection accuracy: Through multi-feature joint triggering, the false triggering rate caused by a single feature is effectively reduced, while the sensitivity and detection accuracy for complex abnormal signals are improved.

[0044] (3) Real-time and flexibility: Based on an oscilloscope to achieve joint feature triggering, it can respond to signal changes in real time and adapt to various signal detection scenarios through flexible condition settings. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 is a flowchart of the anomaly detection method based on time-domain joint features of the present invention;

[0046] Figure 2 is a schematic diagram of a sine sampling signal;

[0047] Figure 3 is a schematic diagram of an amplitude-modulated sampling signal;

[0048] Figure 4 is a flowchart of anomaly detection;

[0049] Figure 5 is a diagram of the anomaly recognition result of a sine signal;

[0050] Figure 6 is a diagram of the anomaly recognition result of an amplitude-modulated signal. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0051] The following describes the specific embodiments of the present invention with reference to the drawings, so that those skilled in the art can better understand the present invention. It should be particularly noted that in the following description, when the detailed description of known functions and designs may dilute the main content of the present invention, these descriptions will be omitted here.

[0052] Embodiment

[0053] In this embodiment, as Figure 1 shown, an anomaly detection method based on time-domain joint features of the present invention includes the following steps:

[0054] (1) Deploy a trigger detection module in a digital oscilloscope. The trigger detection module specifically includes: an extreme value trigger detection module, a pulse width trigger detection module, and a rise time and fall time trigger detection module;

[0055] Among them, the extreme value trigger detection module is used to store the extreme value template comparison parameters of different types of input signals; the pulse width trigger detection module is used to store the pulse width template comparison parameters of different types of input signals; the rise time and fall time trigger detection module is used to store the rise time and fall time template comparison parameters of different types of input signals;

[0056] (2) Signal acquisition;

[0057] Input different types of input signals into the digital oscilloscope in sequence, and use the digital oscilloscope to collect the input signals. 100 segments are collected for each type of input signal. Among them, the j-th segment of the i-th type of input signal is denoted as X ij , X ij ={x ij1 , x ij2 ,…, x ijk ,…, x ijM}, x ijk represents k sampling points in X ij . k is the number index of the sampling points, and M is the number of sampling points;

[0058] In this embodiment, connect the digital oscilloscope to the signal generator, and use the signal generator to generate different types of input signals; take the sine signal and the amplitude-modulated signal as detection examples, use the digital oscilloscope to collect the sine signal and the amplitude-modulated signal, and record the steady-state high level of each segment of the sine sampling signal steady-state low level and the sampling interval time Δt = 0.001 s of each sampling point; record the steady-state high level of each segment of the amplitude-modulated sampling signal steady-state low level and the sampling interval time Δt = 0.001 s of each sampling point, and obtain Figure 2 the sine sampling signal shown in Figure 3 and the amplitude-modulated sampling signal shown in

[0059] (3) Perform multi-dimensional time-domain feature extraction on each segment of the sampling signal X ij ;

[0060] (3.1) Extract extreme value features;

[0061] Traverse the sampling points in each segment of the sampling signal X ij to find all sampling points x ijk-1 whose amplitudes satisfy V ijk < V ijk+1 , and store them in the maximum value sequence ijk ; find all sampling points x whose amplitudes satisfy V ijk-1 > V ijk < V ijk+1The sampling points are stored in the minimum value sequence Then, count the extreme value interval range of each segment of the sampled signal:

[0062] Maximum value interval range:

[0063] Minimum value interval range:

[0064] (3.2) Extract the rise time and fall time features;

[0065] (3.2.1) Calculate the rise time Among them,

[0066] (3.2.2) Calculate the fall time Among them,

[0067] (3.2.3) For each segment of the sampled signal X ij Repeat the above operations to obtain N groups of rise times and fall times

[0068] (3.2.4) Extract the rise time and fall time ranges of the i-th type of input signal;

[0069] Rise time range:

[0070] Fall time range:

[0071] (3.3) Extract the positive pulse width and negative pulse width features;

[0072] (3.3.1) Set the threshold of each segment of the sampled signal X ij

[0073]

[0074] (3.3.2) Traverse the sampling points in each segment of the sampled signal X ij and find the sampling points that satisfy Then mark these sampling points as zero-crossing points;

[0075] (3.3.3) Calculate the pulse width between two adjacent zero-crossing points;

[0076] If the amplitude V of the l-th zero-crossing point in the sampled signal X ij satisfies: ijl Satisfy: ​Then calculate the positive pulse width between the l-th zero crossing point and the (l + 1)-th zero crossing point

[0077]

[0078] wherein, represents the sampling point number of the l-th zero crossing point in the sampling signal X ij ;

[0079] If the amplitude V ij of the l-th zero crossing point in the sampling signal X ijl satisfies: Then calculate the negative pulse width between the l-th zero crossing point and the (l + 1)-th zero crossing point

[0080]

[0081] (3.3.4) Obtain the positive pulse width range and negative pulse width range of the sampling signal X ij :

[0082] Positive pulse width range:

[0083] Negative pulse width range:

[0084] (4) Store the multi-dimensional time-domain features as template comparison parameters into the trigger detection module. Among them, store the extreme value interval range as the extreme value template comparison parameter into the extreme value trigger detection module; store the rise time and fall time range features as the rise time and fall time template comparison parameters into the rise time and fall time trigger detection module; store the positive and negative pulse width ranges as the pulse width template comparison parameters into the pulse width trigger detection module;

[0085] In this embodiment, the template comparison parameters obtained from the sine sampling signal are as follows:

[0086] Rise time range: [0.045, 0.055]

[0087] Fall time range: [0.046, 0.056]

[0088] Maximum value interval range: [0.95, 1.05]

[0089] Minimum value interval range: [-1.05, -0.95]

[0090] Positive pulse width range: [0.095, 0.105]

[0091] Negative pulse width range: [0.095, 0.105]

[0092] The template comparison parameters for obtaining the amplitude-modulated sampling signal are as follows:

[0093] Rise time range: [0.008, 0.010]

[0094] Fall time range: [0.012, 0.016]

[0095] Maximum value interval range: [0.98, 1.07]

[0096] Minimum value interval range: [-1.02, -0.96]

[0097] Positive pulse width range: [0.090, 0.103]

[0098] Negative pulse width range: [0.085, 0.101]

[0099] (5) Input the signal to be measured into the digital oscilloscope, collect the sampling signal containing M sampling points, and then extract the characteristic values of the sampling signal, including the maximum value V max , the minimum value V min , the rise time T rise , the fall time T fall , the positive pulse width W pos , the negative pulse width W neg ; Send the extracted characteristic values to the trigger detection module for template comparison, and the detection process is as Figure 4 shown. If all the characteristic values are within the template comparison parameter range, it is determined that the signal to be measured is normal; otherwise, it is determined that the signal to be measured is abnormal. The abnormal signals to be measured are alternately cached using multiple FIFOs and sent to the upper computer for display, realizing the detection and capture of abnormal signals and implementing the abnormal signal detection method based on time-domain joint characteristics.

[0100] In this embodiment, the comparison results of the characteristic values extracted from the sine sampling signal and the template parameters are as follows:

[0101] Maximum value: 1.2 (abnormal)

[0102] Minimum value: -0.98 (normal)

[0103] Rise time: 0.032 s (abnormal)

[0104] Fall time: 0.025 s (abnormal)

[0105] Positive pulse width: 0.099 s (normal)

[0106] Negative pulse width: -0.098 s (normal)

[0107] The comparison results of the characteristic values extracted from the amplitude-modulated sampling signal and the template parameters are as follows:

[0108] Rise time: 0.036 s (abnormal)

[0109] Fall time: 0.052 s (abnormal)

[0110] Maximum value: 0.26 (abnormal)

[0111] Minimum value: -0.15 (abnormal)

[0112] Positive pulse width: 0.99 (normal)

[0113] Negative pulse width: 0.98 (normal)

[0114] According to the comparison results, it can be judged that an abnormal signal is detected. The detected abnormal signal is sent to the subsequent FIFO buffer and transmitted to the host computer for display. The abnormal signals intercepted from the sine sampling signal are as Figure 5 , and the abnormal signals intercepted from the amplitude-modulated sampling signal are as Figure 6 shown.

[0115] Although the above-described illustrative specific embodiments of the present invention have been described to facilitate the understanding of the present invention by those skilled in the art, it should be clear that the present invention is not limited to the scope of the specific embodiments. For those skilled in the art, as long as various changes are within the spirit and scope of the present invention defined and determined by the appended claims, these changes are obvious, and all inventions and creations using the concept of the present invention are within the scope of protection.

Claims

1. An anomaly detection method based on time-domain joint features, characterized in that, Including the following steps: (1). Deploy a trigger detection module in the digital oscilloscope. The trigger detection module specifically includes: an extreme value trigger detection module, a pulse width trigger detection module, and a rise time and fall time trigger detection module; Among them, the extreme value trigger detection module is used to store the extreme value template comparison parameters of different types of input signals; the pulse width trigger detection module is used to store the pulse width template comparison parameters of different types of input signals; the rise time and fall time trigger detection module is used to store the rise time and fall time template comparison parameters of different types of input signals; (2). Signal acquisition; Input different types of input signals into the digital oscilloscope in sequence, and use the digital oscilloscope to collect the input signals. Each type of input signal is collected in N segments. Among them, the j-th segment of the i-th type of input signal is denoted as X ij , X ij = {x ij1 , x ij2 , …, x ijk , …, x ijM}, where x ijk represents k sampling points in X ij . k is the number index of the sampling points, and M is the number of sampling points; Record each segment of the sampled signal X ij of the steady-state high level steady-state low level and the sampling interval time Δt of each sampling point; (3) Perform multi-dimensional time-domain feature extraction on each segment of the sampled signal X ij ; (3.1). Extract extreme value features; Traverse each sampling signal X ij for the sampling points, and find all the sampling points x ijk-1 where V ijk <V ijk+1 >V ijk , and store them in the maximum value sequence Find all the sampling points where V ijk-1 >V ijk <V ijk+1 , and store them in the minimum value sequence Then, count the extreme value interval range of each sampling signal: Maximum value interval: Minimum value interval: (3.2). Extract rise time and fall time features; (3.2.1), Calculate the rise time Among them, is the low-level time, is the voltage time when the low level rises by 10%, is the voltage time when the low level rises by 90%; (3.2.2), Calculate the fall time Among them, is the high-level time, is the voltage time when the high level drops by 10%, is the voltage time when the high level drops by 90%; (3.2.3), for each segment of the sampled signal X ij Repeat the above operations to obtain N sets of rise times and fall times j = 1, 2…, N; (3.2.4). Extract the rise time and fall time range features of the i-th type of input signal; Rise time range characteristic: Fall time range characteristic: (3.3). Extract positive and negative pulse width features; (3.3.1) Set the threshold value of each segment of the sampling signal X ij ​ (3.3.2) Traverse each sampling point in the sampling signal X ij to find the sampling points that satisfy and then mark these sampling points as zero-crossing points; (3.3.3). Calculate the pulse width between two adjacent zero-crossing points; If the amplitude V of the l-th zero crossing in the sampling signal X ij satisfies: ijl then calculate the positive pulse width between the l-th zero crossing and the (l + 1)-th zero crossing Among them, represents the sampling point number of the l-th zero crossing in the sampling signal X ij ; If the amplitude V of the l-th zero crossing in the sampling signal X ij satisfies: ijl Then calculate the negative pulse width between the l-th zero crossing and the (l + 1)-th zero crossing ​ (3.3.4) Obtain the sampling signal X ij Positive pulse width range and negative pulse width range of Positive pulse width range: Negative pulse width range: (4). Store the multi-dimensional time-domain features as template comparison parameters in the trigger detection module. Among them, store the extreme value interval range as the extreme value template comparison parameter in the extreme value trigger detection module; store the rise time and fall time range features as the rise time and fall time template comparison parameters in the rise time and fall time trigger detection module; store the positive and negative pulse width ranges as the pulse width template comparison parameters in the pulse width trigger detection module; (5) Input the signal to be measured into a digital oscilloscope, collect a sampling signal containing M sampling points, and then extract the characteristic values of the sampling signal, including the maximum value V max , the minimum value V min , the rise time T rise , the fall time T fall , the positive pulse width W pos , the negative pulse width W neg ; Send the extracted characteristic values to the trigger detection module for template comparison. If all the characteristic values are within the template comparison parameter range, it is determined that the signal to be measured is normal; otherwise, it is determined that the signal to be measured is abnormal.

Images