Synchronous capturing and collecting method for multi-node short-time burst signals
By using time-trigger mode and weighted moving average, autocorrelation processing and other technical means in a multi-node environment, the problem of short-term burst signal capture and synchronization is solved, and fast and efficient signal capture and extraction is achieved, which is suitable for signal positioning and tracking research in a multi-node environment.
Patent Information
- Application Number
- CN202510119932.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-01-25
- Publication Date
- 2025-05-09
AI Technical Summary
The prior art is difficult to quickly and efficiently capture and synchronize short burst signals in a multi-node environment, and has high requirements for the real-time acquisition bandwidth and processing speed of the device.
The data is synchronously collected using the time-triggered mode, the signal-to-noise ratio is improved through weighted moving average and autocorrelation processing, and the signal capture and extraction are completed using the peak threshold estimation method.
It realizes synchronous capture of short-time burst signals under multiple nodes, improves signal-to-noise ratio, reduces equipment requirements, and can quickly capture signals under low signal-to-noise ratio conditions, which is suitable for subsequent research on positioning and tracking.
Smart Images
Figure CN119966775A_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the technical field of short-time burst signal capture, and in particular relates to a synchronous capture and collection method for multi-node short-time burst signals. Background Art
[0002] Short-duration burst signals have a short duration, usually only in milliseconds, and appear at random times, making them difficult to capture and track in actual engineering. They are widely used in the field of covert communications, such as communications of military aircraft and submarines. If such signals can be captured synchronously and accurately in multiple nodes, the subsequent positioning and tracking of such signals will be of great significance.
[0003] There are many existing methods for burst signal capture and collection, including methods based on dual sliding window detection, methods based on ultra-wideband acquisition and then related detection processing, and methods based on signal maximum accumulation and dual threshold decision capture.
[0004] The main process of the double sliding window detection method is: the energy ratio of the two windows A and B is used as the decision variable , windows A and B are relatively static during the sliding process. When both windows contain only noise, the response is flat because the noise energy is ideally equal. When the edge of valid data reaches window A, the energy in A continues to increase until the beginning of the valid data is contained in window A. is the peak value of the signal, Figure 1 This position in corresponds to the sampling moment. After this point, window B begins to contain valid signal energy. As valid data enters window B, It gradually descends and finally returns to a flat surface. It can be regarded as a differentiator. When the energy of the input signal changes dramatically, the value will be relatively large. Exceeding the threshold This method assumes that the noise energy is equal in an ideal situation. In reality, there is a difference in noise energy. In addition, this method has a high requirement on the signal-to-noise ratio. When the amount of data is large, it is difficult to ensure the speed during the sliding process.
[0005] The method is based on ultra-wideband acquisition and then related detection processing. Its main process is: capture N radio frequency signals of different frequency bands; divide N different frequency bands into low frequency band, medium frequency band and high frequency band; capture short-time burst signals of the captured radio frequency signals of N different frequency bands in the following ways: identify the captured radio frequency signals of the low frequency band and capture short-time burst signals; move the captured radio frequency signals of the medium frequency band to the data processing frequency band, and identify and capture short-time burst signals of the radio frequency signals moved to the data processing frequency band; move the captured radio frequency signals of the high frequency band to the medium frequency band, move the radio frequency signals moved to the medium frequency band to the data processing frequency band, and identify and capture short-time burst signals of the radio frequency signals moved to the data processing frequency band. This method realizes the acquisition of ultra-wideband burst signals, and has very high requirements on the real-time acquisition bandwidth of the equipment. Due to the signal movement involved, the processing speed is difficult to match the requirements of data real-time performance.
[0006] The main process of the method based on maximum signal accumulation and double threshold judgment capture is as follows: first, perform FFT processing on the signal, then smooth and accumulate the maximum value of the signal spectrum in the frequency domain, smooth again to obtain the signal background, then subtract the signal background, use the double threshold judgment to detect the signal to obtain the signal frequency, and cache the original signal according to the time of signal detection to maintain the complete signal head and tail information. According to the signal frequency obtained from the detection result, the cached original signal is down-converted, filtered and extracted, and the complete signal is saved in the memory. This method involves a lot of related algorithm processing, including multiple smoothing processes, accumulation, down-conversion, filtering and other processes. These processes are time-consuming and difficult to meet the requirements in some scenarios with high speed requirements.
[0007] In view of the deficiencies in the prior art, the present invention provides a method for synchronously capturing and collecting short-term burst signals of multiple nodes, which solves the problems of how to quickly and efficiently complete the capture of burst signals and how to synchronously capture in the case of multiple nodes. Summary of the invention
[0008] In order to solve the problems existing in the prior art, the present invention provides a synchronous capture and collection method for short-term burst signals of multiple nodes.
[0009] The present invention solves the technical problem by adopting the following technical solution: a method for synchronously capturing and collecting short-term burst signals of multiple nodes, comprising the steps of: S1. Each node uses the time-triggered mode to perform data synchronous acquisition: the device of each node uses the time-triggered mode to trigger acquisition, and preset the receiving channel parameters; S2. Weighted average processing of data collected by nodes: After collecting time domain data, each node performs weighted moving average processing. The calculation method of weighted average can be expressed as: ,in, is the weighted moving average at time t, It's in time The weight of i refers to the index value of the data point. It's in time The actual data value, M is the period of the moving average; S3. Data autocorrelation processing: The data after weighted moving average processing is processed according to the formula , perform autocorrelation processing, where N is the total number of data points, x[n] represents the value of the array corresponding to the index value n, and x[nm] represents the value of the array corresponding to the index value of the index value n shifted back by m points; S4. Peak threshold estimation: Setting the threshold , , is a coefficient between 0 and 1, is the peak estimate within the window; S5. Signal extraction: extracting data from the detected signal.
[0010] Further, step S1 comprises: S101, after starting, enter the preset receiving channel; S102, waiting for triggering; S103, after being triggered, sample collection is performed to obtain a sample data block with a timestamp, and the data block is sent; S104, determine whether it is the last collection, if not, repeat steps S101 to S104, if yes, end.
[0011] Further, in step S2, the weights satisfy .
[0012] Furthermore, the weight It is usually a decreasing sequence, and the choice of weights depends on the characteristics of the data and the needs of the analysis.
[0013] Further, in step S4, the sliding time window size is set to , for each signal segment after weighted moving average in the window, record its maximum value as an estimate of the peak value within this window.
[0014] Further, steps S1 to S5 are repeated to complete repeated acquisition of the signal.
[0015] Compared with the prior art, the present invention has the following beneficial effects: 1. Use time triggering to complete synchronous data acquisition, so that each node can synchronously collect the data segment containing burst signals; 2. The signal-to-noise ratio can be improved through weighted moving average and autocorrelation processing, which greatly simplifies the difficulty of signal detection and extraction, reduces the signal-to-noise ratio requirements for the signal, and can quickly capture and collect burst signals under low signal-to-noise ratio conditions; 3. After weighted moving average and autocorrelation data processing, combined with peak threshold estimation, only the signal data is extracted, which ensures that the amount of extracted data is small and facilitates subsequent data application transmission; In summary, the present invention ensures synchronous data collection on each node by adopting time triggering, then improves the signal-to-noise ratio of the target signal by weighted moving average and data autocorrelation processing, and then completes the capture and extraction of the signal by the peak threshold estimation method. This method effectively solves the problems of low timeliness and high equipment requirements for synchronous collection of short-term burst signals under multiple nodes, and has the characteristics of accuracy, high efficiency, and high speed, providing technical support for subsequent research on positioning and tracking of such signals. BRIEF DESCRIPTION OF THE DRAWINGS
[0016] Figure 1 It is the response diagram of the double sliding window method; Figure 2 It is a flow chart of the technical solution of the present invention; Figure 3 A flow chart for data collection using time triggering for each node; Figure 4 To synchronously collect the time domain waveform of the pulse signal using the traditional method; Figure 5 To synchronously collect the time domain waveform of the pulse signal using a time triggering method; Figure 6 It is the time domain waveform of the original pulse signal; Figure 7 It is the time domain waveform diagram of the pulse signal processed by this method; Figure 8 This is a schematic diagram of triggering acquisition using the traditional method; Fig. 9 This is a schematic diagram of time-triggered data acquisition; Figure 4-7 In the figure, the horizontal axis represents time and the vertical axis represents the amplitude of the signal. DETAILED DESCRIPTION
[0017] In order to facilitate the understanding of the present invention, the present invention is described in more detail below in conjunction with the accompanying drawings and specific embodiments. However, the present invention can be implemented in many different forms and is not limited to the embodiments described in this specification. On the contrary, the purpose of providing these embodiments is to make the understanding of the disclosure of the present invention more thorough and comprehensive.
[0018] A synchronous capture and acquisition method for short-term burst signals of multiple nodes is proposed. The synchronous data acquisition is completed by using a time trigger method. Each node synchronously acquires the data segment containing the burst signal, and then improves the signal-to-noise ratio of the target signal through weighted moving average and data autocorrelation processing. The signal is captured and extracted by the peak threshold estimation method. The implementation flow chart is as follows: Figure 2 The specific steps are as follows: S1. To ensure the synchronous data collection of each node, the devices of each node are triggered to collect data by time triggering. The detailed time trigger collection flow chart is as follows: Figure 3 . The receiving channel parameters are preset before triggering because the tuning of the device and the establishment of the digital filter require time, and this time has microsecond differences between individual devices. If the channel parameters are set after the trigger conditions are met, the start time of sample collection by each node receiver will deviate, making it impossible to time-align the sample data. The microsecond difference is enough to make the data invalid. After the time trigger starts, you can choose to collect 1 second of data as a processing cycle. In actual projects, you can set the specific collection time according to the actual situation based on the bandwidth and period of the burst signal to be collected.
[0019] S2. After collecting the time domain data corresponding to the processing period, each node performs weighted moving average processing. Each data point is multiplied by a weight coefficient, and then the weighted values are summed and divided by the sum of the weights. The weights usually decrease exponentially as the data points age, but can also be based on other criteria, such as linear reduction or assigning fixed weights based on the relative importance of the data. The calculation method of weighted average can be expressed as: ,in, is the weighted moving average at time t, It's in time The weight of i refers to the index value of the data point, usually The value of That is, more recent data are given greater weight. It's in time The actual data value. M is the period of the moving average, that is, the number of recent data points included in the calculation. In practical applications, the weights It is usually a decreasing sequence, for example, linearly decreasing weights or exponentially decreasing weights can be used. The choice of weights depends on the characteristics of the data and the needs of the analysis. The weights satisfy , which ensures that the weighted moving average is the weighted average of the original data, and the sum of the weights is 1, so that the average value will not be magnified or reduced due to the weight.
[0020] S3. Data autocorrelation processing: The data after weighted moving average processing is processed according to the formula , perform autocorrelation processing, where N is the total number of data points, is the data value, indicating the value of the array corresponding to index value n. Similarly, x[nm] indicates the value of the array corresponding to index value n shifted back by m points. In the discrete point set, n and nm are the index values of the data points. This formula calculates the sum of the product of the value of the data point at different time points and its delayed version. The purpose is to further improve the signal-to-noise ratio of the signal and make subsequent signal detection and extraction easier.
[0021] S4, peak threshold estimation: perform peak estimation on the signal after weighted moving average processing. Assume the sliding time window size is , for each signal segment after weighted moving average in the window, record its maximum value As the peak value within the window is estimated. Set the threshold Set to a certain proportion of the peak value, that is ,in It is a coefficient between 0 and 1. In this scheme, due to the double processing of improving the signal-to-noise ratio in the early stage, the threshold setting is relatively simple. You can set a value at random first, and adjust it according to the situation if the detection result is not satisfactory.
[0022] S5. Extract data from the detected signal.
[0023] S6. Repeat the above steps to complete repeated acquisition of the signal.
[0024] The key points of the technical solution of the present invention include two: first, for multiple nodes, to ensure the synchronous data collection on each node, it is necessary to use a time-triggered method to collect data, and not to use the traditional collection method, that is, when the collection task is started, the collection instruction is synchronously sent to each node. This is because the response of different node devices is biased, which will cause the data to be unable to be aligned in time, such as Figure 8 This problem can be solved by using time triggering, such as Fig. 9 As shown; the second is to improve the signal-to-noise ratio through weighted moving average and autocorrelation processing, which can greatly simplify the difficulty of signal detection and extraction.
[0025] The present invention uses a time-triggered method to complete the synchronous data acquisition of each node, then improves the signal-to-noise ratio of the target signal through weighted moving average and data autocorrelation processing, and then completes the capture and extraction of the signal through the peak threshold estimation method. This method effectively solves the problems of low timeliness and high equipment requirements for the synchronous acquisition of short-term burst signals under multiple nodes, and has the characteristics of accuracy, high efficiency, and high speed, providing technical support for subsequent research on positioning and tracking of such signals.
[0026] Below, the beneficial effects of the technical solution of the present application are further illustrated by taking the application of the traditional method and the acquisition solution of the present application in a specific case as an example. A project requires the synchronous acquisition of short-term burst signal data from three nodes, and then the subsequent signal positioning and tracking and other processing.
[0027] Comparative example: The traditional method is used to perform three-node synchronous acquisition. The three nodes are placed at the same location, so theoretically the positions of the signals in the three nodes are consistent. Figure 8 As shown, the actual collected data time is not aligned; Figure 4 The pulse signal time domain waveform is shown in the figure; after processing, the corresponding signal starting position index value is obtained, as shown in Table 1. It can be seen that the positions of the signals in the three nodes are quite different.
[0028] Table 1 Synchronous acquisition of detection signal index values using traditional methods
[0029] Embodiment: According to the steps of the technical solution of the present application, firstly, step S1 is performed to place the three nodes at the same location, so theoretically the positions of the signals in the three nodes are consistent, such as Fig. 9 As shown, the time trigger method is used for synchronous acquisition. This scheme uses 1 second of data acquisition as a processing cycle to obtain the following Figure 5 The pulse signal time domain waveform is shown in Table 2, and the corresponding signal starting position index value is obtained. It can be seen that the positions of the signals in the three nodes are basically the same.
[0030] Table 2 Synchronous acquisition of detection signal index values using time triggering method
[0031] The time domain waveform of the original pulse signal is as follows: Figure 6 As shown, it can be seen that the signal-to-noise ratio of the pulse signal is very low, which is not conducive to the extraction of signal data in steps S4 and S5. In this embodiment, the method of the present application is built into the computer software, the time triggers the signal acquisition, and the computer obtains the formula in the operation model to perform weighted moving average processing and autocorrelation processing, and finally obtains the following Figure 7 The time domain waveform of the processed pulse signal is shown in the figure, and it can be seen that the signal-to-noise ratio is significantly improved. Therefore, this acquisition scheme has low requirements on the signal-to-noise ratio, and can complete the rapid capture and acquisition of burst signals under low signal-to-noise ratio conditions.
[0032] In step S4, take In step S5, only the signal data is extracted, the data volume is small, the extraction difficulty is greatly reduced, the extraction is fast, the accuracy is high, and the amount of extracted data is small, which is convenient for subsequent data application transmission, and provides technical support for subsequent research on positioning and tracking of such signals.
[0033] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.
Claims
1. A method for synchronously capturing and collecting short-term burst signals of multiple nodes, characterized in that: Includes steps: S1. Each node uses the time-triggered mode to perform data synchronous acquisition: the device of each node uses the time-triggered mode to trigger acquisition, and preset the receiving channel parameters; S2. Weighted average processing of data collected by nodes: After collecting time domain data, each node performs weighted moving average processing. The calculation method of weighted average can be expressed as: ,in, is the weighted moving average at time t, It's in time The weight of i refers to the index value of the data point. It's in time The actual data value, M is the period of the moving average; S3. Data autocorrelation processing: The data after weighted moving average processing is processed according to the formula , perform autocorrelation processing, where N is the total number of data points, x[n] represents the value of the array corresponding to the index value n, and x[nm] represents the value of the array corresponding to the index value of the index value n shifted back by m points; S4. Peak threshold estimation: Setting the threshold , , is a coefficient between 0 and 1, is the peak estimate within the window; S5. Signal extraction: extracting data from the detected signal.
2. The synchronous capture and collection method of multi-node short-time burst signals according to claim 1 is characterized in that: Step S1 includes: S101, after starting, enter the preset receiving channel; S102, waiting for triggering; S103: after being triggered, sample collection is performed to obtain a sample data block with a timestamp, and the data block is sent; S104, determine whether it is the last collection, if not, repeat steps S101 to S104, if yes, end.
3. The synchronous capture and collection method of multi-node short-term burst signals according to claim 1, characterized in that: In step S2, the weights satisfy .
4. The synchronous capture and collection method of multi-node short-time burst signals according to claim 1, characterized in that: Weight It is usually a decreasing sequence, and the choice of weights depends on the characteristics of the data and the needs of the analysis.
5. The synchronous capture and collection method of multi-node short-time burst signals according to claim 1, characterized in that: In step S4, the sliding time window size is set to , for each signal segment after weighted moving average in the window, record its maximum value as an estimate of the peak value within this window.
6. The synchronous capture and collection method of multi-node short-time burst signals according to any one of claims 1 to 5, characterized in that: Repeat steps S1 to S5 to complete repeated acquisition of the signal.