Transient signal real-time capturing method based on frequency domain detection in digital oscilloscope

Abstract

The invention discloses a transient signal real-time capturing method based on frequency domain detection in a digital oscilloscope, and the method comprises the steps: carrying out the multi-channel spectrum analysis and processing of a signal through employing an improved WOLA structure, and carrying out the channel selection of the processed signal according to the initial frequency position of an effective signal and a signal boundary, and carrying out different capturing methods. Fourier transform and feature comparison are carried out on the signal containing the effective signal frequency band, and the frequency band not containing the effective signal is directly compared with the maximum value and the minimum value of the normal signal, so that transient signal capture of different frequency bands of the signal is realized.

Description

Technical Field

[0001] This invention belongs to the field of digital oscilloscope technology, and more specifically, relates to a method for real-time acquisition of transient signals based on frequency domain detection in a digital oscilloscope. Background Technology

[0002] With the rapid development of wireless communication, radar detection, and spectrum monitoring, the complexity of signals acquired by high-speed acquisition systems in the time and frequency domains is increasing. Modern communication systems widely employ frequency hopping, spread spectrum, linear frequency modulation, and various digital modulation techniques, resulting in signals exhibiting complex characteristics such as dynamic distribution, rapid switching, and wide bandwidth coverage in the frequency domain. Key features of these signals during transmission, such as transient bursts, spectral distortion, and the absence of specific modes, often carry important intelligence such as system status, interference information, or threat signals. Therefore, in high-speed acquisition systems, achieving real-time and accurate acquisition of transient signals has become a crucial guarantee for the system's functional integrity and reliability.

[0003] Traditional signal acquisition methods in digital oscilloscopes primarily rely on time-domain triggering mechanisms, such as edge triggering, pulse width triggering, or pattern triggering. These methods are suitable for simple signals with clear time-domain characteristics, but their acquisition capability is significantly insufficient for complex signals with substantial frequency-domain dynamic changes. For example, when receiving frequency-hopping signals, time-domain triggering cannot effectively follow the frequency jump patterns; when facing transient interference or in-band spectral changes, the time-domain waveform may not show significant distortion, causing critical signal segments to be ignored by the system.

[0004] Frequency domain analysis provides a new technical approach for transient signal acquisition. Through real-time spectrum analysis, the acquisition system can identify the frequency distribution, power spectral density, and time-varying patterns of the signal's spectrum. However, directly implementing full-bandwidth, high-resolution frequency domain processing in high-speed acquisition environments faces significant challenges: on the one hand, traditional FFT processing of full-bandwidth data involves enormous computational demands, making it difficult to meet real-time requirements; on the other hand, the frequency domain characteristics of transient signals often exhibit locality and time-varying nature, and simply using fixed thresholds or global template matching can easily lead to missed captures or false alarms. Existing signal acquisition schemes based on frequency domain detection typically struggle to balance processing speed, acquisition accuracy, and algorithmic flexibility in resource-constrained hardware environments.

[0005] Especially in high-sampling-rate, wide-bandwidth applications, the signal data throughput is extremely high. If all data is processed indiscriminately in the frequency domain, it will lead to excessive system load and response delay, making it difficult to capture transient or non-stationary signals in a timely manner. At the same time, the frequency domain structure of transient signals may evolve over time, and fixed-parameter spectrum analysis methods cannot adaptively track changes in signal characteristics, further limiting the practicality and robustness of the acquisition system.

[0006] Therefore, current frequency domain acquisition techniques for complex signals in high-speed acquisition systems of digital oscilloscopes still have significant shortcomings, lacking a solution that can achieve efficient, intelligent, and adaptive frequency domain analysis and acquisition with limited hardware resources. There is an urgent need in this field for a novel frequency domain acquisition method that can balance real-time processing capabilities, accurate acquisition performance, and resource efficiency to meet the pressing requirements of modern high-speed acquisition systems for real-time acquisition of transient signals. Summary of the Invention

[0007] The purpose of this invention is to overcome the shortcomings of the prior art and provide a real-time acquisition method for transient signals based on frequency domain detection in a digital oscilloscope. By using a parallel WOLA (Weigthed OverLap-Add) structure to perform multi-channel spectrum processing on the signal, the processed normal signal and the signal under test are compared in the frequency domain, and the signal is captured based on the differences in their spectral characteristics.

[0008] To achieve the above-mentioned objective, the present invention provides a method for real-time acquisition of transient signals based on frequency domain detection in a digital oscilloscope, characterized by comprising the following steps:

[0009] (1) Power on the digital oscilloscope and set it to normal acquisition mode;

[0010] (2) Input the normal signal into the digital oscilloscope and acquire it through the ADC in the acquisition system to obtain parallel D-channel sampling data. , This refers to the index of the sampling points in the sampled data;

[0011] (3) Channelized spectrum analysis processing of the sampled data is performed using the WOLA structure;

[0012] (4) Obtain a valid channel;

[0013] (5) Fast Fourier Transform;

[0014] Record the low-frequency signal output in the effective channel as The signal is then subjected to a Fast Fourier Transform to obtain its spectrum, and the maximum amplitude in the spectrum is recorded. Minimum amplitude and the corresponding frequency , ;

[0015] (6) Extract the amplitude range of the low-frequency signal output from the ineffective channel;

[0016] (7) Switch the digital oscilloscope to transient signal capture mode, then input the signal to be tested, and process it according to steps (2) to (6), and record the maximum amplitude of the spectrum obtained under the signal to be tested. Minimum amplitude and the corresponding frequency , ;

[0017] (8) Set the capture flag;

[0018] (9) When the capture flag appears, the digital oscilloscope enables transient signal capture and stores it in RAM. Finally, the transient signal is transmitted to the host computer for display.

[0019] The objective of this invention is achieved as follows:

[0020] This invention discloses a real-time acquisition method for transient signals based on frequency domain detection in a digital oscilloscope. It utilizes an improved WOLA structure to perform multi-channel spectrum analysis and processing on the signal. The processed signal selects a channel for different acquisition methods based on the starting frequency position of the effective signal and the signal boundary. Signals containing the effective signal frequency band undergo Fourier transform and feature comparison, while frequency bands not containing the effective signal are directly compared with the maximum and minimum values ​​of the normal signal, thereby achieving the acquisition of transient signals in different frequency bands of the signal.

[0021] Meanwhile, the real-time acquisition method for transient signals based on frequency domain detection in a digital oscilloscope of the present invention also has the following beneficial effects:

[0022] (1) Enhanced ability to capture complex transient signals: This method can effectively identify and capture transient signals with indistinct time-domain waveform characteristics, such as frequency hopping signals and spectrum change signals, through frequency domain analysis. It overcomes the limitations of traditional time-domain triggering mechanisms in capturing dynamically changing signals in the frequency domain and significantly improves the sensitivity and accuracy of digital oscilloscopes in detecting complex signals.

[0023] (2) Achieve efficient real-time frequency domain processing and acquisition: The efficient WOLA structure is used for channelized spectrum analysis, which divides the full bandwidth signal into multiple narrowband channels and reduces the sampling rate, greatly reducing the amount of data processing computation and system resource occupation. Thus, it can still maintain real-time processing capability in high-speed, high-bandwidth acquisition environment and meet the requirements for rapid acquisition of transient signals.

[0024] (3) It has an adaptive and intelligent acquisition and judgment mechanism: This method first learns the frequency domain characteristics of normal signals (such as effective channel, amplitude and frequency range, etc.), and then compares them with the signal to be tested in multiple dimensions to realize anomaly detection and triggering based on frequency domain characteristics. This adaptive judgment mechanism can dynamically track signal changes, reduce false triggering or missed triggering caused by environmental or signal fluctuations, and improve the robustness and applicability of the system. Attached Figure Description

[0025] Figure 1This is a flowchart of a method for real-time acquisition of transient signals based on frequency domain detection in a digital oscilloscope according to the present invention;

[0026] Figure 2 (a) is a traditional channelized digital downconversion structure;

[0027] Figure 2 (b) is a parallelized WOLA structure;

[0028] Figure 3 This is a schematic diagram of a parallelized WOLA structure implementing channelized downsampling of signals;

[0029] Figure 4 This is a normal signal waveform spectrum diagram;

[0030] Figure 5 Time-domain waveform and spectrum of 2.4G interference signal added;

[0031] Figure 6 Add the time-domain waveform and spectrum of the frequency abrupt change signal. Detailed Implementation

[0032] The specific embodiments of the present invention will now be described with reference to the accompanying drawings to enable those skilled in the art to better understand the invention. It should be particularly noted that in the following description, detailed descriptions of known functions and designs that might obscure the main content of the invention will be omitted here.

[0033] Example

[0034] In this embodiment, as Figure 1 As shown, the present invention discloses a method for real-time acquisition of transient signals based on frequency domain detection in a digital oscilloscope, comprising the following steps:

[0035] (1) In this embodiment, the transient signal real-time acquisition method is divided into two modes: normal acquisition mode and acquisition mode. After powering on the digital oscilloscope, we first set it to normal acquisition mode.

[0036] (2) In normal acquisition mode, the normal signal is input to the digital oscilloscope and acquired by the ADC in the acquisition system to obtain parallel D-channel sampling data. , This refers to the index of the sampling points in the sampled data;

[0037] In this embodiment, the normal signal is assumed to be a 100MHz sine wave, and the sine wave needs to be sampled for a sufficient duration (i.e., at least 3 sampling periods).

[0038] Let the sampling rate of the ADC be... The FPGA's operating frequency is 20 GSPS. The frequency is 312.5MHz, therefore the number of parallel paths D is: ;

[0039] (3) Channelized spectrum analysis processing of the sampled data is performed using the WOLA structure;

[0040] The full name of the WOLA structure is Weighted Overlay Addition Structure. This structure is a polyphase discrete Fourier transform (DFT) filter bank used in channelized receivers. It optimizes parameter design flexibility by shifting the signal instead of the window function. Figure 2 (a) is a traditional channelized digital downconversion structure. Figure 2 (b) is a parallelized WOLA structure, which has the following advantages:

[0041] (s1) Figure 2 (a) shows that the traditional structure in the channelization filter must process high-frequency signals, usually requiring a bandpass filter design with a steep transition band and high order to separate adjacent channels, resulting in high computational complexity. In contrast, the parallelized WOLA structure only needs to perform signal channelization processing through weighting, grouping summation, and DFT. The weighting coefficients only need to use the coefficients of the prototype low-pass filter, and the transition band requirement is relaxed, the order is low, and it is easy to implement.

[0042] (s2) For high-speed parallel data, directly implement Figure 2 The high-performance band channelization filter (traditional channelization structure) shown in (a) is extremely difficult to construct. The efficient WOLA structure achieves channelization through parallel weighting, group summation and DFT processes, resulting in high throughput and stable performance.

[0043] (s3) In Figure 2 In the traditional channelization architecture shown in (a), the anti-mirror filter is usually independent of the channelization filter, increasing system complexity. Figure 2 In the parallelized WOLA structure shown in (b), only weighted summation and DFT are required, without the need for additional independent anti-mirror filters and decimation modules.

[0044] Finally, the WOLA structure of DFT is used for channelized spectrum analysis of broadband digital signals. This structure can divide the full-bandwidth signal into multiple narrow-bandwidth digital channels in the digital domain, and then shift the sub-signals of each channel to the low-frequency band through digital domain signal processing. At the same time, the sampling rate of the signal is greatly reduced by digital decimation, thereby greatly reducing the computational load of subsequent processing and realizing real-time data processing. The steps are as follows:

[0045] (3.1) Set sampling data truncation length ; Sample data Shift D data points backward, starting from the first... The data will be truncated starting at a length of [length missing]. Sampling data , ;

[0046] (3.2) Sampling data Perform channelized digital filtering;

[0047] Let the coefficients of the prototype low-pass filter be... The corresponding bandwidth is ,in, The sampling rate of the acquisition system, The number of channels is given. In this embodiment, the sampling rate of the acquisition system is set to 20 GSPS and the number of channels is 128. Therefore, the passband cutoff frequency of the prototype low-pass filter is 78.125 MHz and the stopband start frequency is 156.25 MHz.

[0048] Record No. The sampled data after D data shifting and truncation in each channel is: ; Use channelized digital filters to process the sampled data Perform weighted filtering to obtain the filtered signal. :

[0049] ;

[0050] (3.3) Filter the signal according to Increasing order, with Points are grouped together and divided into The vector is then summed by taking the components in the same column to obtain a vector of length . signal , ;

[0051] (3.4) Regarding the signal DFT and time-domain modulation are performed to obtain the channelized low-frequency signal. ;

[0052] In this embodiment, channelized downsampling of the signal is implemented in the FPGA through a WOLA structure, and the specific implementation steps are as follows: Figure 3 As shown, the input signal The data is grouped by a decimation factor, with 64 data points per group. Input signal. Shift into the data in sequence using 64 data samples as the basic unit. middle, Every 64 data points are updated, the WOLA structure performs one data processing step, resulting in 128 time-domain samples corresponding to a specific moment in each of the 128 channels. A prototype low-pass filter is then used. right The data is weighted, that is and Multiplying corresponding points together yields a 512-dot product sequence. Next, the weighted sequence according to The data is grouped into four 128-dimensional vectors, each group consisting of 128 points in ascending order. The corresponding components of these four vectors are then summed to obtain 128 sums. A 128-point DFT is performed on these 128 sums, and the resulting sequence is then multiplied by... Then you can get the final output.

[0053] (4) Obtain a valid channel;

[0054] Calculate the frequency range of each channel using the following formula:

[0055] ;

[0056] Extract the frequency of the input normal signal. Determine frequency Record the corresponding channel number within the frequency range of the channel it is located in, and set that channel as a valid channel;

[0057] In this embodiment, when the bandwidth of the channelized digital filter is At 156.25MHz, the effective channel for normal signals is located in channel 1.

[0058] (5) Fast Fourier Transform;

[0059] Record the low-frequency signal output in the effective channel as The signal is then subjected to a Fast Fourier Transform to obtain its spectrum, and the maximum amplitude in the spectrum is recorded. Minimum amplitude and the corresponding frequency , ;

[0060] In this embodiment, when performing channel selection for a low-frequency signal, since the signal is located in channel 1, the output of that channel is subjected to a Fast Fourier Transform with 128 points, resulting in... Figure 4 The waveform spectrum shown;

[0061] (6) Extract the amplitude range of the low-frequency signal output from the ineffective channel;

[0062] low frequency signal The amplitude range is obtained through the extreme value detection module: ;

[0063] (7) Switch the digital oscilloscope to transient signal capture mode, then input the signal to be tested, and process it according to steps (2) to (6), and record the maximum amplitude of the spectrum obtained under the signal to be tested. Minimum amplitude and the corresponding frequency , ;

[0064] (8) Set the capture flag;

[0065] (8.1) Boundary crossing anomaly judgment;

[0066] Frequency comparison: If and If the frequency is within the specified range, then the output flag will be set. Otherwise, the frequency is judged to be out of bounds. ;

[0067] Amplitude comparison: If and If the amplitude is within the limit, then the output flag is set. Otherwise, the judgment range is out of bounds. ;

[0068] When the frequency exceeds the limit Or the amplitude exceeds the limit When this happens, set the capture flag. ,otherwise, ;

[0069] (8.2) Determining the amplitude fluctuation range of the signal under test;

[0070] Under the signal to be measured, the amplitude range of the low-frequency signal output in each ineffective channel is traversed. The amplitude range of the signal is compared with that of the corresponding channel under normal signal conditions. If the amplitude range of a channel under the test signal exceeds this range, the capture flag is set to... Otherwise, the capture flag is set to ;

[0071] (9) Transient signal capture;

[0072] when or At that time, the digital oscilloscope activates transient signal capture and stores it in RAM, and finally transmits the transient signal to the host computer for display.

[0073] In this embodiment, the spectral distribution characteristics of a certain acquired signal are as follows: Figure 5 and Figure 6 As shown, where, as Figure 5 When a 2.4 GHz interference signal is added to a normal signal, based on the spectral information of the ineffective channel, the spectral amplitude of the synthesized signal is significantly higher than the spectral range of the normal signal. Therefore, the capture flag is set to... The system will capture the signal collected at this moment, store it, and display it.

[0074] like Figure 6 When performing a frequency jump operation on a signal, the frequency coordinates corresponding to the maximum and minimum amplitudes of the synthesized signal in the same channel are significantly different from those of the normal signal, indicating a frequency out-of-bounds error. Therefore, the capture flag is set to... The system will capture the signal collected at this moment, store it, and display it.

[0075] Although the illustrative specific embodiments of the present invention have been described above to enable those skilled in the art to understand the invention, it should be understood that the invention is not limited to the scope of the specific embodiments. For those skilled in the art, various changes are obvious as long as they are within the spirit and scope of the invention as defined and determined by the appended claims, and all inventions utilizing the concept of the present invention are protected.

Claims

1. A method for real-time acquisition of transient signals based on frequency domain detection in a digital oscilloscope, characterized in that, Includes the following steps: (1) Power on the digital oscilloscope and set it to normal acquisition mode; (2) Input the normal signal to the digital oscilloscope and acquire it through the ADC in the acquisition system to obtain D-channel sampling data. , This refers to the index of the sampling points in the sampled data; (3) Channelized spectrum analysis processing of the sampled data is performed using the WOLA structure; (3.1) Set sampling data truncation length ; Sample data Shift D data points backward, starting from the first... The data will be truncated starting at a length of [length missing]. Sampling data , ; (3.2) Sampling data Perform channelized digital filtering; Let the coefficients of the prototype low-pass filter be... The corresponding bandwidth is ,in, The sampling rate of the acquisition system, Number of channels; Record No. The sampled data after D data shifting and truncation in each channel is: ; Using channelized digital filters to process sampled data Perform weighted filtering to obtain the filtered signal. : ； (3.3) Filter the signal according to Increasing order, with Points are grouped together and divided into The vector is then summed by taking the components in the same column to obtain a vector of length . signal , ; (3.4) Regarding the signal Parallel DFT and time-domain modulation are performed in the digital domain to obtain the channelized low-frequency signal. ; (4) Obtain a valid channel; Calculate the frequency range of each channel using the following formula: ； Extract the frequency of the input normal signal. Determine frequency Record the corresponding channel number within the frequency range of the channel it is located in, and set that channel as a valid channel; (5) Fast Fourier Transform; Record the low-frequency signal output in the effective channel as The signal is then subjected to a Fast Fourier Transform to obtain its spectrum, and the maximum amplitude in the spectrum is recorded. Minimum amplitude and the corresponding frequency , ; (6) Extract the amplitude range of the low-frequency signal output from the ineffective channel; low frequency signal The amplitude range is obtained through the extreme value detection module: ; (7) Switch the digital oscilloscope to transient signal capture mode, then input the signal to be tested, and process it according to steps (2) to (6), and record the maximum amplitude of the spectrum obtained under the signal to be tested. Minimum amplitude and the corresponding frequency , ; (8) Set the capture flag; (8.1) Boundary crossing anomaly judgment; Frequency comparison: If and If the frequency is within the specified range, then the output flag will be set. Otherwise, the frequency is judged to be out of bounds. ; Amplitude comparison: If and If the amplitude is within the limit, then the output flag is set. Otherwise, the judgment range is out of bounds. ; When the frequency exceeds the limit Or the amplitude exceeds the limit When this happens, set the capture flag. ,otherwise, ; (8.2) Determining the amplitude fluctuation range of the signal under test; Under the signal to be measured, the amplitude range of the low-frequency signal output in each ineffective channel is traversed. The amplitude range of the signal is compared with that of the corresponding channel under normal signal conditions. If the amplitude range of a channel under the test signal exceeds this range, the capture flag is set to... Otherwise, the capture flag is set to ; (9) Transient signal capture; when or At that time, the digital oscilloscope activates transient signal capture and stores it in RAM, and finally transmits the transient signal to the host computer for display.

Images