Pulse data type analysis method and device, electronic equipment and storage medium

Abstract

The invention provides a pulse data type analysis method and device, electronic equipment and a storage medium, and relates to the technical field of pulse data type analysis. In the application, firstly, acquired sample pulse data can be converted into sample pulse two-dimensional data; secondly, labeling the sample pulse two-dimensional data to form pulse type label data corresponding to the sample pulse two-dimensional data; then, training a candidate neural network based on the sample pulse two-dimensional data and the pulse type label data to form a target neural network corresponding to the candidate neural network; and finally, converting the obtained target pulse data into target pulse two-dimensional data, analyzing the target pulse two-dimensional data by using the target neural network, and outputting a target data type corresponding to the target pulse two-dimensional data. Based on the above content, the problem that the reliability of pulse data type analysis is relatively low in the prior art can be improved.

Description

Technical Field

[0001] The present application relates to the technical field of pulse data type analysis, and in particular to a pulse data type analysis method and device, electronic equipment and storage medium. Background Art

[0002] The analysis of pulse data types currently mainly includes algorithms such as the cumulative difference histogram method and the sequence difference histogram method. However, the inventors have found through research that these algorithms are unable to effectively sort low repetition rate signals (signals with a low frequency of repetition. Specifically, repetition rate refers to the frequency at which a signal recurs in the time domain. If a signal repeats at a low frequency (such as a few hertz or lower), it is called a low repetition rate signal. This signal may have an irregular, sparse time distribution, or be triggered by certain specific events. In some cases, low repetition rate signals may have a specific impact on the design of detection, analysis, or communication systems), resulting in a relatively low reliability problem in the analysis of pulse data types. Summary of the invention

[0003] In view of this, the purpose of the present application is to provide a pulse data type analysis method and device, electronic device and storage medium to improve the problem of relatively low reliability of pulse data type analysis in the prior art.

[0004] To achieve the above objectives, this application adopts the following technical solutions:

[0005] A pulse data type analysis method, comprising:

[0006] Converting the acquired sample pulse data into sample pulse two-dimensional data, wherein the coordinate system of the sample pulse two-dimensional data is a pulse repetition interval-pulse signal arrival time two-dimensional coordinate system;

[0007] Performing labeling processing on the sample pulse two-dimensional data to form pulse type label data corresponding to the sample pulse two-dimensional data, wherein the pulse type label data is used to reflect the data type of the sample pulse two-dimensional data, and the data type at least includes a low repetition rate pulse type;

[0008] Based on the sample pulse two-dimensional data and the pulse type label data, a candidate neural network is trained to form a target neural network corresponding to the candidate neural network;

[0009] The acquired target pulse data is converted into target pulse two-dimensional data, and the target neural network is used to analyze the target pulse two-dimensional data to output the target data type corresponding to the target pulse two-dimensional data, wherein the coordinate system of the target pulse two-dimensional data is the pulse repetition interval-pulse signal arrival time two-dimensional coordinate system.

[0010] In a preferred embodiment of the present application, in the above-mentioned pulse data type analysis method, the step of converting the acquired sample pulse data into sample pulse two-dimensional data includes:

[0011] Acquiring sample pulse data, wherein the sample pulse data includes a plurality of sample pulse signals;

[0012] Determine the pulse repetition interval of each sample pulse signal among the multiple sample pulse signals included in the sample pulse data, perform one-dimensional transformation processing on each of the pulse repetition intervals, and output multiple pulse repetition interval transformation values ​​corresponding to the multiple sample pulse signals;

[0013] The plurality of pulse repetition interval transformation values ​​are subjected to two-dimensional transformation processing, a plurality of pulse repetition interval two-dimensional transformation results are output, and corresponding sample pulse two-dimensional data are obtained based on the plurality of pulse repetition interval two-dimensional transformation results.

[0014] In a preferred embodiment of the present application, in the above-mentioned pulse data type analysis method, the step of determining the pulse repetition interval of each sample pulse signal in the multiple sample pulse signals included in the sample pulse data, and performing one-dimensional transformation processing on each of the pulse repetition intervals, and outputting multiple pulse repetition interval transformation values ​​corresponding to the multiple sample pulse signals, includes:

[0015] Determine a pulse repetition interval of each sample pulse signal among a plurality of sample pulse signals included in the sample pulse data;

[0016] For each of the pulse repetition intervals, based on a predetermined one-dimensional transformation factor, one-dimensional transformation processing is performed on the pulse repetition interval, and a one-dimensional transformation value corresponding to the pulse repetition interval is output; and based on a predetermined approximate function, the one-dimensional transformation value is approximated, and a pulse repetition interval transformation value corresponding to the pulse repetition interval is output.

[0017] In a preferred embodiment of the present application, in the above-mentioned pulse data type analysis method, the step of performing two-dimensional transformation processing on the multiple pulse repetition interval transformation values, outputting multiple pulse repetition interval two-dimensional transformation results, and obtaining corresponding sample pulse two-dimensional data based on the multiple pulse repetition interval two-dimensional transformation results includes:

[0018] In the pulse repetition interval of each sample pulse signal in the plurality of sample pulse signals, a maximum value is determined to obtain a maximum pulse repetition interval, and a corresponding first interval is formed based on the maximum pulse repetition interval and a predetermined target parameter;

[0019] Determine, from the plurality of pulse repetition interval transformation values, each pulse repetition interval transformation value that is equal to any one of the first parameters in the first interval, and construct a corresponding pulse position set based on the pulse position of each determined pulse repetition interval transformation value;

[0020] Counting the number of pulse positions in the pulse position set to obtain a target number;

[0021] Based on the target quantity, determining a reference quantity;

[0022] Based on the target parameter and the reference quantity, forming a corresponding second interval;

[0023] For each second parameter in the second interval, based on the second parameter, determine a two-dimensional transformation result of a pulse repetition interval corresponding to the second parameter;

[0024] Based on the two-dimensional transformation result of the pulse repetition interval corresponding to each second parameter in the second interval, corresponding sample pulse two-dimensional data is formed.

[0025] In a preferred embodiment of the present application, in the above-mentioned pulse data type analysis method, the step of determining the two-dimensional transformation result of the pulse repetition interval corresponding to each second parameter in the second interval based on the second parameter includes:

[0026] For each second parameter in the second interval, a sum of the second parameter and the target parameter is calculated to obtain a target sum corresponding to the second parameter, and a corresponding third interval is formed based on the target sum and the target quantity, and, in the pulse position set, a pulse position corresponding to the second parameter is determined, and a pulse repetition interval change value corresponding to the pulse position is determined as a first pulse repetition interval change value, and, in the pulse position set, a pulse position corresponding to each third parameter in the third interval is determined respectively, and a pulse repetition interval change value corresponding to the pulse position is determined as a second pulse repetition interval change value;

[0027] For each second parameter in the second interval, based on the pulse position set, the pulse position corresponding to the second parameter is determined, and the pulse repetition interval corresponding to the pulse position is calculated, and the difference between each second pulse repetition interval transformation value corresponding to the second parameter and the first pulse repetition interval transformation value is calculated respectively, and, based on the pulse repetition interval and the difference, a two-dimensional transformation result of each pulse repetition interval corresponding to the second parameter is obtained.

[0028] In a preferred embodiment of the present application, in the above-mentioned pulse data type analysis method, the step of forming corresponding sample pulse two-dimensional data based on the pulse repetition interval two-dimensional transformation result corresponding to each second parameter in the second interval includes:

[0029] Determine the pulse repetition interval two-dimensional transformation result of the target row from the pulse repetition interval two-dimensional transformation results corresponding to each second parameter in the second interval, wherein the plurality of pulse repetition interval two-dimensional transformation results form a plurality of rows and columns;

[0030] Determine the filtering parameters corresponding to the target row, and perform filtering processing on the pulse repetition interval two-dimensional transformation result of the target row based on the filtering parameters;

[0031] After obtaining the filtering processing results of the pulse repetition interval two-dimensional transformation results of each row, corresponding sample pulse two-dimensional data are formed based on the filtering processing results of the pulse repetition interval two-dimensional transformation results of each row, wherein the sample pulse two-dimensional data includes the filtering processing results of the pulse repetition interval two-dimensional transformation results of each row.

[0032] In a preferred embodiment of the present application, in the above-mentioned pulse data type analysis method, the step of determining the filtering parameters corresponding to the target row and filtering the two-dimensional transformation result of the pulse repetition interval of the target row based on the filtering parameters includes:

[0033] Calculating a target row error based on a filtering result of a two-dimensional transformation result of a pulse repetition interval of a row preceding the target row and a two-dimensional transformation result of a pulse repetition interval of a row preceding the target row;

[0034] Determine a corresponding filter offset parameter based on the target row error, a pulse repetition interval two-dimensional transformation result of a row before the target row, and a pre-configured weight parameter;

[0035] Determining the filtering parameters corresponding to the target row based on the filtering offset parameters and the filtering parameters corresponding to the previous row of the target row;

[0036] Based on the filtering parameters, filtering processing is performed on the pulse repetition interval two-dimensional transformation result of the target row, wherein the filtering processing includes multiplying the filtering parameters and the pulse repetition interval two-dimensional transformation result of the target row.

[0037] The present application also provides a pulse data type analysis device, comprising:

[0038] A pulse data conversion module, used to convert the acquired sample pulse data into sample pulse two-dimensional data, wherein the coordinate system of the sample pulse two-dimensional data is a pulse repetition interval-pulse signal arrival time two-dimensional coordinate system;

[0039] A pulse type labeling module, used for labeling the sample pulse two-dimensional data to form pulse type label data corresponding to the sample pulse two-dimensional data, wherein the pulse type label data is used to reflect the data type of the sample pulse two-dimensional data, and the data type at least includes a low repetition rate pulse type;

[0040] A neural network training module, used for training a candidate neural network based on the sample pulse two-dimensional data and the pulse type label data to form a target neural network corresponding to the candidate neural network;

[0041] The pulse type analysis module is used to convert the acquired target pulse data into target pulse two-dimensional data, and use the target neural network to analyze the target pulse two-dimensional data, and output the target data type corresponding to the target pulse two-dimensional data, wherein the coordinate system of the target pulse two-dimensional data is the pulse repetition interval-pulse signal arrival time two-dimensional coordinate system.

[0042] Based on the above, the present application also provides an electronic device, including:

[0043] Memory for storing computer programs;

[0044] The processor connected to the memory is used to execute the computer program stored in the memory to implement the above-mentioned pulse data type analysis method.

[0045] On the basis of the above, the present application also provides a computer-readable storage medium, in which a computer program is stored, and when the computer program is run, each step of the above-mentioned pulse data type analysis method is executed.

[0046] The pulse data type analysis method and device, electronic device and storage medium provided by the present application can firstly convert the acquired sample pulse data into sample pulse two-dimensional data; secondly, the sample pulse two-dimensional data can be annotated to form pulse type label data corresponding to the sample pulse two-dimensional data; then, the candidate neural network can be trained based on the sample pulse two-dimensional data and the pulse type label data to form a target neural network corresponding to the candidate neural network; finally, the acquired target pulse data is converted into target pulse two-dimensional data, and the target neural network is used to analyze the target pulse two-dimensional data and output the target data type corresponding to the target pulse two-dimensional data. Based on the above content, since the sample pulse data is converted into the pulse repetition interval-pulse signal arrival time two-dimensional coordinate system to form the corresponding sample pulse two-dimensional data, the sample pulse two-dimensional data can be reliably annotated, and then the pulse type label data formed by the annotation can be reliably trained with a neural network, so that the formed target neural network has a higher pulse data type analysis capability, therefore, the target neural network can be used to reliably analyze the pulse data type of the target pulse two-dimensional data corresponding to the target pulse data, that is, the problem of relatively low reliability of pulse data type analysis existing in the prior art can be improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, preferred embodiments are specifically cited below and described in detail with reference to the attached drawings.

[0048] Figure 1 A structural block diagram of an electronic device provided in an embodiment of the present application.

[0049] Figure 2 A schematic diagram of a flow chart of a pulse data type analysis method provided in an embodiment of the present application.

[0050] Figure 3 A schematic diagram of sample pulse two-dimensional data before filtering provided in an embodiment of the present application.

[0051] Figure 4 A schematic diagram of filtered sample pulse two-dimensional data provided in an embodiment of the present application.

[0052] Figure 5 A block diagram of a pulse data type analysis device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0053] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. The components of the embodiments of the present application described and shown in the drawings here can be arranged and designed in various different configurations.

[0054] Therefore, the following detailed description of the embodiments of the present application provided in the accompanying drawings is not intended to limit the scope of the present application for which protection is sought, but merely represents selected embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without creative work are within the scope of protection of the present application.

[0055] like Figure 1 As shown, an embodiment of the present application provides an electronic device, wherein the electronic device may include a memory, a processor, and a pulse data type analysis device.

[0056] In detail, the memory and the processor are directly or indirectly electrically connected to realize data transmission or interaction. For example, the memory and the processor can be electrically connected through one or more communication buses or signal lines. The pulse data type analysis device includes at least one software function module stored in the memory in the form of software or firmware. The processor is used to execute an executable computer program stored in the memory, for example, the software function module and computer program included in the pulse data type analysis device, so as to implement the pulse data type analysis method provided in the embodiment of the present application.

[0057] Optionally, the memory can be, but is not limited to, random access memory (RAM), read only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable read-only memory (Electric Erasable Programmable Read-Only Memory, EEPROM), etc.

[0058] Furthermore, the processor may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), a system on chip (SoC), etc.; it may also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0059] Understandably, Figure 1 The structure shown is for illustration only, and the electronic device may also include Figure 1 More or fewer components as shown, or with Figure 1 The different configurations shown, for example, may also include a communication unit for exchanging information with other devices.

[0060] Combination Figure 2 The embodiment of the present application also provides a pulse data type analysis method applicable to the above electronic device, wherein the method steps defined in the process related to the pulse data type analysis method can be implemented by the electronic device.

[0061] The following will Figure 2 The specific process shown is explained in detail.

[0062] Step S110, converting the acquired sample pulse data into sample pulse two-dimensional data.

[0063] In the embodiment of the present application, the electronic device can convert the acquired sample pulse data into sample pulse two-dimensional data, and the sample pulse data can be one-dimensional data. The coordinate system of the sample pulse two-dimensional data is the pulse repetition interval-pulse signal arrival time two-dimensional coordinate system, such as Figure 3 As shown, the horizontal axis is the pulse repetition interval (PRI) and the vertical axis is the pulse signal arrival time (TOA). Figure 3 A white dot in the graph represents a two-dimensional data.

[0064] Step S120 , labeling the sample pulse two-dimensional data to form pulse type label data corresponding to the sample pulse two-dimensional data.

[0065] In an embodiment of the present application, after obtaining the sample pulse two-dimensional data, the electronic device may perform labeling processing on the sample pulse two-dimensional data to form pulse type label data corresponding to the sample pulse two-dimensional data. The pulse type label data is used to reflect the data type of the sample pulse two-dimensional data, and the data type at least includes a low repetition rate pulse type. For example, in an alternative embodiment, the pulse type label data may be a low repetition rate pulse type, or may not be a low repetition rate pulse type. In some other embodiments, in order to further improve the accuracy of type analysis, the low repetition rate pulse type may be further subdivided, such as a linear low repetition rate pulse type (such as Figure 3 The points in the vertical axis that show a straight line) and the curved low repetition rate pulse type (such as Figure 3 The data types of other points do not belong to the low repetition rate pulse type. Figure 3 It can include points of N rows and M columns, where the data types of points in the same column are the same. Therefore, a 1*M vector can be constructed as the pulse type label data. The vector parameter corresponding to the straight line point can be 1, the vector parameter corresponding to the curve point can be 2, and the vector parameter corresponding to the point that does not belong to the low repetition rate pulse type can be 0.

[0066] Step S130, training the candidate neural network based on the sample pulse two-dimensional data and the pulse type label data to form a target neural network corresponding to the candidate neural network.

[0067] In an embodiment of the present application, after obtaining the sample pulse two-dimensional data and the pulse type label data, the electronic device can train the candidate neural network based on the sample pulse two-dimensional data and the pulse type label data to form a target neural network corresponding to the candidate neural network. For example, the candidate neural network can be used to analyze the sample pulse two-dimensional data and output the corresponding sample data type. Then, the error (such as cross entropy loss, etc.) between the sample data type and the pulse type label data can be calculated. Finally, the network parameters of the candidate neural network can be adjusted in the direction of reducing the error until the error is reduced to the target error or the error reduction is less than the preset amplitude, and the training of the candidate neural network is completed, thereby forming a target neural network. In addition, the size of the pulse repetition interval and the type of the pulse repetition interval can be randomly changed, and the above operation can be repeated 1000 times independently to obtain 1000 sample pulse two-dimensional data, of which 70% can be used as a training set and 30% can be used as a test set. Furthermore, the candidate neural network may be a recurrent neural network, and the corresponding network structure may be composed of a sequence feature input layer, a bidirectional LSTM layer, a dropout layer, a fully connected layer, a softmax layer, and a classification layer. A more detailed network structure is not specifically limited or described herein.

[0068] Step S140, converting the acquired target pulse data into target pulse two-dimensional data, and using the target neural network to analyze the target pulse two-dimensional data, and outputting the target data type corresponding to the target pulse two-dimensional data.

[0069] In an embodiment of the present application, after training and forming a target neural network, the electronic device can convert the acquired target pulse data into target pulse two-dimensional data (as described above), and use the target neural network to analyze the target pulse two-dimensional data, and output the target data type corresponding to the target pulse two-dimensional data. The coordinate system of the target pulse two-dimensional data is a pulse repetition interval-pulse signal arrival time two-dimensional coordinate system.

[0070] Based on the above content, since the sample pulse data will be converted into the pulse repetition interval-pulse signal arrival time two-dimensional coordinate system to form the corresponding sample pulse two-dimensional data, the sample pulse two-dimensional data can be reliably labeled and processed, and then reliable neural network training can be performed based on the pulse type label data formed by the annotation, so that the formed target neural network has a higher pulse data type analysis capability. Therefore, the target neural network can be used to reliably analyze the pulse data type of the target pulse two-dimensional data corresponding to the target pulse data. In other words, the problem of relatively low reliability of pulse data type analysis existing in the prior art can be improved.

[0071] It should be noted that for step S110 , the specific method of converting the sample pulse data into the sample pulse two-dimensional data is not limited and can be selected according to actual needs.

[0072] For example, in an alternative implementation, in order to ensure the reliability of the two-dimensional data conversion, the above-mentioned step S110 may further include step S111, step S112 and step S113, and the specific content of each step is described as follows.

[0073] Step S111, acquiring sample pulse data.

[0074] In the embodiment of the present application, sample pulse data may be acquired first, wherein the sample pulse data includes a plurality of sample pulse signals, for example, sample pulse data with a time length of 10 seconds is acquired.

[0075] Step S112, determining the pulse repetition interval of each sample pulse signal in the multiple sample pulse signals included in the sample pulse data, performing one-dimensional transformation processing on each of the pulse repetition intervals, and outputting multiple pulse repetition interval transformation values ​​corresponding to the multiple sample pulse signals.

[0076] In the embodiment of the present application, after the sample pulse data is acquired, the pulse repetition interval of each of the multiple sample pulse signals included in the sample pulse data can be determined, and a one-dimensional transformation process is performed on each of the pulse repetition intervals, and multiple pulse repetition interval transformation values ​​corresponding to the multiple sample pulse signals are output. In other words, there is a one-to-one correspondence between the pulse repetition interval transformation value, the sample pulse signal, and the pulse repetition interval.

[0077] Step S113, performing two-dimensional transformation processing on the multiple pulse repetition interval transformation values, outputting multiple pulse repetition interval two-dimensional transformation results, and obtaining corresponding sample pulse two-dimensional data based on the multiple pulse repetition interval two-dimensional transformation results.

[0078] In an embodiment of the present application, after obtaining a plurality of pulse repetition interval transformation values, the plurality of pulse repetition interval transformation values ​​can be subjected to two-dimensional transformation processing (i.e., projected into a corresponding two-dimensional coordinate system), and a plurality of pulse repetition interval two-dimensional transformation results can be outputted. Based on the plurality of pulse repetition interval two-dimensional transformation results, corresponding sample pulse two-dimensional data can be obtained.

[0079] It can be understood that, in the above step S112, the specific method of outputting the pulse repetition interval change value is not limited and can be selected according to actual needs.

[0080] For example, in an alternative embodiment, in order to better aggregate similar pulses and thus reduce the error caused by data acquisition, the above-mentioned step S112 may further include step S112a and step S112b, and the specific content of each step is described as follows.

[0081] Step S112a, determining the pulse repetition interval of each sample pulse signal among a plurality of sample pulse signals included in the sample pulse data.

[0082] In the embodiment of the present application, the pulse repetition interval of each sample pulse signal in the plurality of sample pulse signals included in the sample pulse data may be determined first.

[0083] Step S112b, for each of the pulse repetition intervals, based on a predetermined one-dimensional transformation factor, perform one-dimensional transformation processing on the pulse repetition interval, output a one-dimensional transformation value corresponding to the pulse repetition interval, and based on a predetermined approximate function, perform approximate processing on the one-dimensional transformation value, and output a pulse repetition interval transformation value corresponding to the pulse repetition interval.

[0084] In an embodiment of the present application, after the pulse repetition interval is determined, for each of the pulse repetition intervals, a one-dimensional transformation process can be performed on the pulse repetition interval based on a predetermined one-dimensional transformation factor, and a one-dimensional transformation value corresponding to the pulse repetition interval can be output, and the one-dimensional transformation value can be approximated based on a predetermined approximate function, and a pulse repetition interval transformation value corresponding to the pulse repetition interval can be output. Exemplarily, the pulse repetition interval can be divided by the one-dimensional transformation factor, and then the result can be rounded to obtain the pulse repetition interval transformation value, so that the pulse repetition interval transformation value corresponding to each pulse repetition interval can be obtained. In addition, the specific value of the one-dimensional transformation factor is not limited, for example, it can be 1000. Among them, the specific formula for performing the one-dimensional transformation process can be as follows:

[0085] y i =round(PRI i / W);

[0086] Among them, y i Indicates the pulse repetition interval change value corresponding to the i-th PRI, round() indicates rounding, PRI i represents the i-th PRI (pulse repetition interval), and W represents a one-dimensional transformation factor.

[0087] It can be understood that, in the above step S113, the specific method of performing two-dimensional transformation processing to output sample pulse two-dimensional data is not limited and can be selected according to actual needs.

[0088] For example, in an alternative embodiment, in order to further enable similar pulses to be better aggregated, thereby reducing the error caused by data acquisition, the above-mentioned step S113 may further include step S113a, step S113b, step S113c, step S113d, step S113e, step S113f and step S113g, and the specific content of each step is described as follows.

[0089] Step S113a, determining a maximum value in the pulse repetition interval of each sample pulse signal in the plurality of sample pulse signals to obtain a maximum pulse repetition interval, and forming a corresponding first interval based on the maximum pulse repetition interval and a predetermined target parameter.

[0090] In the embodiment of the present application, after the pulse repetition interval is obtained, a maximum value can be determined in the pulse repetition interval of each sample pulse signal in the plurality of sample pulse signals to obtain a maximum pulse repetition interval, and a corresponding first interval is formed based on the maximum pulse repetition interval and a predetermined target parameter. Exemplarily, the target parameter can be 1, so that the first interval can be obtained as [1, PRI-MAX], where PRI-MAX represents the maximum pulse repetition interval.

[0091] Step S113b, among the multiple pulse repetition interval transformation values, determine each pulse repetition interval transformation value that is equal to any first parameter in the first interval, and construct a corresponding pulse position set based on the pulse position of each determined pulse repetition interval transformation value.

[0092] In the embodiment of the present application, after obtaining the first interval, each pulse repetition interval transformation value equal to any first parameter in the first interval can be determined from the plurality of pulse repetition interval transformation values, and based on the pulse position of each determined pulse repetition interval transformation value (i.e., the position of the corresponding sample pulse signal in the sample pulse data, such as the time sequence), a corresponding pulse position set is constructed. The specific implementation method can be:

[0093] for z=1:PRI_MAX

[0094] Index = find(y == z);

[0095] Among them, find(y==z) means finding the position of the number equal to z in y, that is, finding each pulse repetition interval transformation value that is equal to any first parameter (i.e. z) in the first interval among the multiple pulse repetition interval transformation values ​​(i.e. y), and Index represents the pulse position set.

[0096] Step S113c, counting the number of pulse positions in the pulse position set to obtain a target number.

[0097] In the embodiment of the present application, after forming a pulse position set, the number of pulse positions in the pulse position set may be counted to obtain a target number, which may be represented by Index_length.

[0098] Step S113d: determining a reference quantity based on the target quantity.

[0099] In the embodiment of the present application, after the target number is obtained, a reference number can be determined based on the target number, which can be represented by Index_length-1.

[0100] Step S113e: forming a corresponding second interval based on the target parameter and the reference quantity.

[0101] In the embodiment of the present application, after the reference number is obtained, a corresponding second interval, such as [1, Index_length-1], can be formed based on the target parameter and the reference number.

[0102] Step S113f, for each second parameter in the second interval, based on the second parameter, determine a two-dimensional transformation result of the pulse repetition interval corresponding to the second parameter.

[0103] In the embodiment of the present application, after the second interval is formed, for each second parameter in the second interval, a two-dimensional transformation result of the pulse repetition interval corresponding to the second parameter can be determined based on the second parameter.

[0104] Step S113g: Based on the two-dimensional transformation result of the pulse repetition interval corresponding to each second parameter in the second interval, corresponding sample pulse two-dimensional data is formed.

[0105] In the embodiment of the present application, after obtaining the repetition interval two-dimensional transformation result, the corresponding sample pulse two-dimensional data can be formed based on the pulse repetition interval two-dimensional transformation result corresponding to each second parameter in the second interval. Exemplarily, the repetition interval two-dimensional transformation result can be directly used as the corresponding sample pulse two-dimensional data, or the repetition interval two-dimensional transformation result can be further processed, and then the processed result is used as the corresponding sample pulse two-dimensional data.

[0106] It can be understood that, in the above-mentioned step S113f, the specific method of determining the corresponding pulse repetition interval two-dimensional transformation result based on the second parameter is not limited. For example, in an alternative embodiment, in order to ensure the reliability of the determined pulse repetition interval two-dimensional transformation result, the above-mentioned step S113f can further include step f1 and step f2. The specific contents of each step are as follows.

[0107] Step f1, for each second parameter in the second interval, calculate the sum of the second parameter and the target parameter to obtain the target sum corresponding to the second parameter, and form a corresponding third interval based on the target sum and the target quantity, and, in the pulse position set, determine the pulse position corresponding to the second parameter, and determine the pulse repetition interval transformation value corresponding to the pulse position as the first pulse repetition interval transformation value, and, in the pulse position set, respectively determine the pulse position corresponding to each third parameter in the third interval, and determine the pulse repetition interval transformation value corresponding to the pulse position as the second pulse repetition interval transformation value.

[0108] In an embodiment of the present application, after the second interval is formed, the sum of the second parameter and the target parameter can be calculated for each second parameter in the second interval to obtain the target sum corresponding to the second parameter, and based on the target sum and the target quantity, a corresponding third interval is formed, and, in the pulse position set, the pulse position corresponding to the second parameter is determined, and the pulse repetition interval transformation value corresponding to the pulse position is determined as the first pulse repetition interval transformation value, and, in the pulse position set, the pulse position corresponding to each third parameter in the third interval is determined respectively, and the pulse repetition interval transformation value corresponding to the pulse position is determined as the second pulse repetition interval transformation value.

[0109] Step f2, for each second parameter in the second interval, based on the pulse position set, determine the pulse position corresponding to the second parameter, and the pulse repetition interval corresponding to the pulse position, and respectively calculate the difference between each second pulse repetition interval transformation value corresponding to the second parameter and the first pulse repetition interval transformation value, and, based on the pulse repetition interval and the difference, obtain the two-dimensional transformation result of each pulse repetition interval corresponding to the second parameter.

[0110] In an embodiment of the present application, for each second parameter in the second interval, based on the pulse position set, the pulse position corresponding to the second parameter is determined, and the pulse repetition interval corresponding to the pulse position, and the difference between each second pulse repetition interval transformation value corresponding to the second parameter and the first pulse repetition interval transformation value is calculated respectively, and, based on the pulse repetition interval and the difference, each pulse repetition interval two-dimensional transformation result corresponding to the second parameter is obtained (for example, the pulse repetition interval transformation value corresponding to the pulse repetition interval is used as the horizontal coordinate of the pulse repetition interval two-dimensional transformation result, and the difference is rounded down as the vertical coordinate of the pulse repetition interval two-dimensional transformation result).

[0111] Based on this, in a specific application, the specific implementation of the above steps f1 and f2 can be as follows:

[0112]

[0113] Among them, i represents the second parameter, i+1 represents the target and value, j represents the third parameter, the third interval is [i+1, Index_length], floor represents rounding down, PRI Indexj Indicates the second pulse repetition interval change value, PRI Indexi represents the first pulse repetition interval transformation value, YY represents the ordinate of the pulse repetition interval two-dimensional transformation result, and XX represents the abscissa of the pulse repetition interval two-dimensional transformation result.

[0114] It can be understood that, in the above-mentioned step S113g, the specific method of forming the corresponding sample pulse two-dimensional data based on the pulse repetition interval two-dimensional transformation result is not limited. For example, in an alternative embodiment, in order to further improve the reliability of the sample pulse two-dimensional data and reduce the interference of noise data, the above-mentioned step S113g can further include step g1, step g2 and step g3, and the specific content of each step is described as follows.

[0115] Step g1: determining a two-dimensional transformation result of the pulse repetition interval of a target row from the two-dimensional transformation results of the pulse repetition interval corresponding to each second parameter in the second interval.

[0116] In an embodiment of the present application, the pulse repetition interval two-dimensional transformation result of the target row can be determined from the pulse repetition interval two-dimensional transformation result corresponding to each second parameter in the second interval. Wherein, a plurality of the pulse repetition interval two-dimensional transformation results form multiple rows and multiple columns. Exemplarily, each row can be used as a target row in turn, or some rows can be used as target rows (so that the data of other rows can be directly retained, thereby ensuring the diversity and richness of the sample data). For example, the first 100 rows can be directly retained, that is, starting from row 101 as the target row.

[0117] Step g2, determining the filtering parameters corresponding to the target row, and filtering the pulse repetition interval two-dimensional transformation result of the target row based on the filtering parameters.

[0118] In an embodiment of the present application, the filtering parameters corresponding to the target row can also be determined, and based on the filtering parameters, the two-dimensional transformation result of the pulse repetition interval of the target row can be filtered, so that the interference of noise data on subsequent network training can be reduced.

[0119] Step g3, after obtaining the filtering processing result of the pulse repetition interval two-dimensional transformation result of each row, forming corresponding sample pulse two-dimensional data based on the filtering processing result of the pulse repetition interval two-dimensional transformation result of each row.

[0120] In the embodiment of the present application, after obtaining the filtering processing result of the pulse repetition interval two-dimensional transformation result of each row, corresponding sample pulse two-dimensional data can be formed based on the filtering processing result of the pulse repetition interval two-dimensional transformation result of each row. The sample pulse two-dimensional data includes the filtering processing result of the pulse repetition interval two-dimensional transformation result of each row.

[0121] It is understandable that in the above step g2, in order to further ensure the reliability of the filtering process, the filtering process can be performed in the following manner:

[0122] First, the target row error may be calculated based on the filtering result of the pulse repetition interval two-dimensional transformation result of the row before the target row and the pulse repetition interval two-dimensional transformation result of the row before the target row, such as performing difference calculation;

[0123] Secondly, the corresponding filtering offset parameter can be determined based on the target row error, the pulse repetition interval two-dimensional transformation result of the row before the target row, and the pre-configured weight parameter. For example, the product of the target row error, the pulse repetition interval two-dimensional transformation result of the row before the target row, and the pre-configured weight parameter can be calculated as the filtering offset parameter;

[0124] Then, based on the filtering offset parameter and the filtering parameter corresponding to the previous row of the target row, the filtering parameter corresponding to the target row can be determined. For example, the sum of the filtering offset parameter and the filtering parameter corresponding to the previous row of the target row can be calculated as the filtering parameter.

[0125] Finally, based on the filtering parameters, the two-dimensional transformation result of the pulse repetition interval of the target row can be filtered, wherein the filtering process includes multiplying the filtering parameters and the two-dimensional transformation result of the pulse repetition interval of the target row. For example, the result of the multiplication calculation can be used as the result after filtering.

[0126] For example, in a specific application, filtering processing can be implemented based on the following method:

[0127]

[0128] Among them, PRI2D i represents the result of two-dimensional transformation of the pulse repetition interval of the i-th row, yn_l i represents the filtering result of the two-dimensional transformation result of the pulse repetition interval of the i-th row, w represents the filtering parameter, e i represents the error of the i-th row, and 2*0.02 represents the weight parameter. It should be noted that in the above implementation content, "for" represents a loop, and ":" represents the value from one parameter to another parameter. Based on this, through the above filtering process, Figure 3 The pulse repetition interval two-dimensional transformation result shown is converted into the corresponding filtering processing result to interfere with more noise data and retain all the data of the first 100 rows, thereby fully ensuring the diversity of the sample pulse two-dimensional data. In addition, the specific value of the initial value of w is not limited and can be configured according to needs, for example, it can be 0.

[0129] Combination Figure 5 The embodiment of the present application also provides a pulse data type analysis device applicable to the above electronic device. The pulse data type analysis device may include a pulse data conversion module, a pulse type labeling module, a neural network training module and a pulse type analysis module.

[0130] The pulse data conversion module is used to convert the acquired sample pulse data into sample pulse two-dimensional data, wherein the coordinate system of the sample pulse two-dimensional data is the pulse repetition interval-pulse signal arrival time two-dimensional coordinate system. In the embodiment of the present application, the pulse data conversion module can be used to perform Figure 2 As shown in step S110, the relevant contents of the pulse data conversion module can refer to the above description of step S110.

[0131] The pulse type labeling module is used to label the sample pulse two-dimensional data to form pulse type label data corresponding to the sample pulse two-dimensional data, wherein the pulse type label data is used to reflect the data type of the sample pulse two-dimensional data, and the data type at least includes a low repetition rate pulse type. In the embodiment of the present application, the pulse type labeling module can be used to perform Figure 2 As shown in step S120, the relevant contents of the pulse type labeling module can refer to the above description of step S120.

[0132] The neural network training module is used to train the candidate neural network based on the sample pulse two-dimensional data and the pulse type label data to form a target neural network corresponding to the candidate neural network. In the embodiment of the present application, the neural network training module can be used to perform Figure 2 As shown in step S130, the relevant contents of the neural network training module can refer to the above description of step S130.

[0133] The pulse type analysis module is used to convert the acquired target pulse data into target pulse two-dimensional data, and use the target neural network to analyze the target pulse two-dimensional data, and output the target data type corresponding to the target pulse two-dimensional data, wherein the coordinate system of the target pulse two-dimensional data is the pulse repetition interval-pulse signal arrival time two-dimensional coordinate system. In the embodiment of the present application, the pulse type analysis module can be used to perform Figure 2 As shown in step S140, the relevant contents of the pulse type analysis module can refer to the above description of step S140.

[0134] In an embodiment of the present application, corresponding to the pulse data type analysis method applied to the electronic device, a computer-readable storage medium is also provided, in which a computer program is stored, and when the computer program is run, each step of the pulse data type analysis method is executed. Among them, each step executed when the aforementioned computer program is run will not be described one by one here, and reference can be made to the explanation of the pulse data type analysis method in the previous text.

[0135] In summary, the pulse data type analysis method and device, electronic device and storage medium provided by the present application can, first, convert the acquired sample pulse data into sample pulse two-dimensional data; secondly, the sample pulse two-dimensional data can be annotated to form pulse type label data corresponding to the sample pulse two-dimensional data; then, the candidate neural network can be trained based on the sample pulse two-dimensional data and the pulse type label data to form a target neural network corresponding to the candidate neural network; finally, the acquired target pulse data is converted into target pulse two-dimensional data, and the target neural network is used to analyze the target pulse two-dimensional data and output the target data type corresponding to the target pulse two-dimensional data. Based on the above content, since the sample pulse data is converted into the pulse repetition interval-pulse signal arrival time two-dimensional coordinate system to form the corresponding sample pulse two-dimensional data, the sample pulse two-dimensional data can be reliably annotated, and then the pulse type label data formed by the annotation can be reliably trained with a neural network, so that the formed target neural network has a higher pulse data type analysis capability, therefore, the target neural network can be used to reliably analyze the pulse data type of the target pulse two-dimensional data corresponding to the target pulse data, thereby improving the problem of relatively low reliability of pulse data type analysis existing in the prior art.

[0136] In several embodiments provided in the embodiments of the present application, it should be understood that the disclosed device and method can also be implemented in other ways. The device and method embodiments described above are merely schematic, for example, the flowcharts and block diagrams in the accompanying drawings show the possible architecture, functions and operations of the device, method and computer program product according to multiple embodiments of the present application. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of a code, and the module, a program segment or a part of a code contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of boxes in the block diagram and / or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or action, or can be implemented with a combination of dedicated hardware and computer instructions.

[0137] In addition, the functional modules in the various embodiments of the present application may be integrated together to form an independent part, or each module may exist separately, or two or more modules may be integrated to form an independent part.

[0138] If the function is implemented in the form of a software function module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or the part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, an electronic device, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a disk or an optical disk. It should be noted that, in this article, the term "include", "include" or any other variant thereof is intended to cover non-exclusive inclusion, so that the process, method, article or device including a series of elements includes not only those elements, but also includes other elements that are not explicitly listed, or also includes elements inherent to such a process, method, article or device. Without more constraints, an element defined by the phrase "comprising a..." does not exclude the existence of other identical elements in the process, method, article or apparatus comprising the element.

[0139] The above description is only the preferred embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A pulse data type analysis method, characterized in that: include: Converting the acquired sample pulse data into sample pulse two-dimensional data, wherein the coordinate system of the sample pulse two-dimensional data is a pulse repetition interval-pulse signal arrival time two-dimensional coordinate system; Performing labeling processing on the sample pulse two-dimensional data to form pulse type label data corresponding to the sample pulse two-dimensional data, wherein the pulse type label data is used to reflect the data type of the sample pulse two-dimensional data, and the data type at least includes a low repetition rate pulse type; Based on the sample pulse two-dimensional data and the pulse type label data, a candidate neural network is trained to form a target neural network corresponding to the candidate neural network; The acquired target pulse data is converted into target pulse two-dimensional data, and the target neural network is used to analyze the target pulse two-dimensional data to output the target data type corresponding to the target pulse two-dimensional data, wherein the coordinate system of the target pulse two-dimensional data is the pulse repetition interval-pulse signal arrival time two-dimensional coordinate system.

2. The pulse data type analysis method according to claim 1, characterized in that: The step of converting the acquired sample pulse data into sample pulse two-dimensional data comprises: Acquiring sample pulse data, wherein the sample pulse data includes a plurality of sample pulse signals; Determine the pulse repetition interval of each sample pulse signal among the multiple sample pulse signals included in the sample pulse data, perform one-dimensional transformation processing on each of the pulse repetition intervals, and output multiple pulse repetition interval transformation values ​​corresponding to the multiple sample pulse signals; The plurality of pulse repetition interval transformation values ​​are subjected to two-dimensional transformation processing, a plurality of pulse repetition interval two-dimensional transformation results are output, and corresponding sample pulse two-dimensional data are obtained based on the plurality of pulse repetition interval two-dimensional transformation results.

3. The pulse data type analysis method according to claim 2, characterized in that: The step of determining the pulse repetition interval of each sample pulse signal among a plurality of sample pulse signals included in the sample pulse data, performing one-dimensional transformation processing on each of the pulse repetition intervals, and outputting a plurality of pulse repetition interval transformation values ​​corresponding to the plurality of sample pulse signals comprises: Determine a pulse repetition interval of each sample pulse signal among a plurality of sample pulse signals included in the sample pulse data; For each of the pulse repetition intervals, based on a predetermined one-dimensional transformation factor, one-dimensional transformation processing is performed on the pulse repetition interval, and a one-dimensional transformation value corresponding to the pulse repetition interval is output; and based on a predetermined approximate function, the one-dimensional transformation value is approximated, and a pulse repetition interval transformation value corresponding to the pulse repetition interval is output.

4. The pulse data type analysis method according to claim 2, characterized in that: The step of performing two-dimensional transformation processing on the plurality of pulse repetition interval transformation values, outputting a plurality of pulse repetition interval two-dimensional transformation results, and obtaining corresponding sample pulse two-dimensional data based on the plurality of pulse repetition interval two-dimensional transformation results comprises: In the pulse repetition interval of each sample pulse signal in the plurality of sample pulse signals, a maximum value is determined to obtain a maximum pulse repetition interval, and a corresponding first interval is formed based on the maximum pulse repetition interval and a predetermined target parameter; Determine, from the plurality of pulse repetition interval transformation values, each pulse repetition interval transformation value that is equal to any one of the first parameters in the first interval, and construct a corresponding pulse position set based on the pulse position of each determined pulse repetition interval transformation value; Counting the number of pulse positions in the pulse position set to obtain a target number; Based on the target quantity, determining a reference quantity; Based on the target parameter and the reference quantity, forming a corresponding second interval; For each second parameter in the second interval, based on the second parameter, determine a two-dimensional transformation result of a pulse repetition interval corresponding to the second parameter; Based on the two-dimensional transformation result of the pulse repetition interval corresponding to each second parameter in the second interval, corresponding sample pulse two-dimensional data is formed.

5. The pulse data type analysis method according to claim 4, characterized in that: The step of determining, for each second parameter in the second interval, a two-dimensional transformation result of a pulse repetition interval corresponding to the second parameter based on the second parameter, comprises: For each second parameter in the second interval, a sum of the second parameter and the target parameter is calculated to obtain a target sum corresponding to the second parameter, and a corresponding third interval is formed based on the target sum and the target quantity, and, in the pulse position set, a pulse position corresponding to the second parameter is determined, and a pulse repetition interval change value corresponding to the pulse position is determined as a first pulse repetition interval change value, and, in the pulse position set, a pulse position corresponding to each third parameter in the third interval is determined respectively, and a pulse repetition interval change value corresponding to the pulse position is determined as a second pulse repetition interval change value; For each second parameter in the second interval, based on the pulse position set, the pulse position corresponding to the second parameter is determined, and the pulse repetition interval corresponding to the pulse position is calculated, and the difference between each second pulse repetition interval transformation value corresponding to the second parameter and the first pulse repetition interval transformation value is calculated respectively, and, based on the pulse repetition interval and the difference, a two-dimensional transformation result of each pulse repetition interval corresponding to the second parameter is obtained.

6. The pulse data type analysis method according to claim 4, characterized in that: The step of forming corresponding sample pulse two-dimensional data based on the pulse repetition interval two-dimensional transformation result corresponding to each second parameter in the second interval comprises: Determine the pulse repetition interval two-dimensional transformation result of the target row from the pulse repetition interval two-dimensional transformation results corresponding to each second parameter in the second interval, wherein the plurality of pulse repetition interval two-dimensional transformation results form a plurality of rows and columns; Determine the filtering parameters corresponding to the target row, and perform filtering processing on the pulse repetition interval two-dimensional transformation result of the target row based on the filtering parameters; After obtaining the filtering processing results of the pulse repetition interval two-dimensional transformation results of each row, corresponding sample pulse two-dimensional data are formed based on the filtering processing results of the pulse repetition interval two-dimensional transformation results of each row, wherein the sample pulse two-dimensional data includes the filtering processing results of the pulse repetition interval two-dimensional transformation results of each row.

7. The pulse data type analysis method according to claim 6, characterized in that: The step of determining the filtering parameters corresponding to the target row and filtering the pulse repetition interval two-dimensional transformation result of the target row based on the filtering parameters includes: Calculating a target row error based on a filtering result of a two-dimensional transformation result of a pulse repetition interval of a row preceding the target row and a two-dimensional transformation result of a pulse repetition interval of a row preceding the target row; Determine a corresponding filter offset parameter based on the target row error, a pulse repetition interval two-dimensional transformation result of a row before the target row, and a pre-configured weight parameter; Determining the filtering parameters corresponding to the target row based on the filtering offset parameters and the filtering parameters corresponding to the previous row of the target row; Based on the filtering parameters, filtering processing is performed on the pulse repetition interval two-dimensional transformation result of the target row, wherein the filtering processing includes multiplying the filtering parameters and the pulse repetition interval two-dimensional transformation result of the target row.

8. A pulse data type analysis device, characterized in that: include: A pulse data conversion module, used to convert the acquired sample pulse data into sample pulse two-dimensional data, wherein the coordinate system of the sample pulse two-dimensional data is a pulse repetition interval-pulse signal arrival time two-dimensional coordinate system; A pulse type labeling module, used for labeling the sample pulse two-dimensional data to form pulse type label data corresponding to the sample pulse two-dimensional data, wherein the pulse type label data is used to reflect the data type of the sample pulse two-dimensional data, and the data type at least includes a low repetition rate pulse type; A neural network training module, used for training a candidate neural network based on the sample pulse two-dimensional data and the pulse type label data to form a target neural network corresponding to the candidate neural network; The pulse type analysis module is used to convert the acquired target pulse data into target pulse two-dimensional data, and use the target neural network to analyze the target pulse two-dimensional data, and output the target data type corresponding to the target pulse two-dimensional data, wherein the coordinate system of the target pulse two-dimensional data is the pulse repetition interval-pulse signal arrival time two-dimensional coordinate system.

9. An electronic device, characterized in that: include: Memory for storing computer programs; A processor connected to the memory is used to execute a computer program stored in the memory to implement the pulse data type analysis method described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is run, the pulse data type analysis method according to any one of claims 1 to 7 is executed.

Images