Electromagnetic data matching degree calculation method, signal processing method and signal processing system

Abstract

The application discloses an electromagnetic data matching degree calculation method, a signal processing method and a signal processing system, and belongs to the field of electronic reconnaissance. The method comprises the following steps: S1, calculating the search length of the electromagnetic signals to be matched and setting an initial displacement; S2, calculating the matching search granularity; S3, calculating the matching degrees of different displacements between the electromagnetic signals to be matched; S4, calculating a new initial displacement and a new matching search granularity; S5, repeating steps S3 to S4 until the matching degree search granularity meets the accuracy requirement or reaches a minimum value, and calculating the minimum value matching degree as the optimal matching degree of the electromagnetic signals to be matched, and completing the electromagnetic data matching when the target is detected or identified according to the matching degree. The application improves the accuracy and speed of the matching degree of the non-periodic data of the electromagnetic signals, reduces the communication amount and the calculation amount when the practical equipment is applied, makes the data matching real-time higher, consumes less resources, and guarantees the matching accuracy.

Description

Technical Field

[0001] This invention relates to the field of electronic reconnaissance, and more specifically, to a method for calculating electromagnetic data matching degree, a signal processing method, and a signal processing system. Background Technology

[0002] Data or image matching degree calculation is very important in various fields, especially in signal processing and image processing. The data we deal with is usually large-volume sequential data or matrix data. At the same time, due to the pollution and noise caused by various factors during the sensor or digitization process, the matching degree calculation between different data is computationally intensive and complex, and there is also the possibility of non-convergence.

[0003] Currently, the commonly used method involves calculating the shift relationship of the data to be matched through correlation operations, and then calculating the matching degree after compensation shift as the final matching degree. This method faces resource and computational slowness issues when the data length is large. Furthermore, since the correlation operation in the previous step is entirely based on the data itself, the calculated shift relationship may not be exactly the optimal position for the matching degree, and the final matching degree obtained under this shift condition may not be accurate. This is fatal to subsequent processing.

[0004] Another existing method involves performing feature transformations or extracting feature points from the data, and then calculating the matching degree in the feature domain to reduce computational complexity. This method is highly dependent on feature transformations, and specific feature transformation algorithms need to be developed for different objects and data from different domains. Moreover, with large datasets, the computational complexity of feature transformations is substantial, further increasing the overall computational load.

[0005] Another common existing method is to extract data to reduce its dimensionality and then calculate the matching degree in the lower dimension. This method usually deteriorates the data quality after data extraction, resulting in a large error in the matching degree calculation, which adversely affects subsequent judgment, fusion and other processing.

[0006] Due to the shortcomings of the above methods, the accuracy and speed of electromagnetic data matching, signal processing and image processing scenarios are low, which in turn leads to problems of resource consumption and slow computing speed. The real-time performance of related electronic reconnaissance systems is poor, and existing methods are difficult to meet the application requirements of signal processing and image processing scenarios with high real-time requirements. Summary of the Invention

[0007] The purpose of this invention is to overcome the shortcomings of the prior art and provide an electromagnetic data matching degree calculation method, signal processing method, and signal processing system to improve the accuracy and speed of matching degree of non-periodic electromagnetic signal data, reduce communication and computation load in actual equipment applications, make data matching calculation more real-time, consume less resources, and ensure matching accuracy.

[0008] The objective of this invention is achieved through the following solution:

[0009] An electromagnetic data matching degree calculation method involves receiving and sampling electromagnetic signals using an electromagnetic signal receiving device, then calculating the matching degree when performing target detection or recognition using the electromagnetic signal data; finally, using the matching degree to complete subsequent target detection or recognition processing; the calculation of the matching degree when performing target detection or recognition using electromagnetic signal data specifically includes the following steps:

[0010] Let D1(N) be a uniformly sampled aperiodic electromagnetic signal data sequence of length N, and D2(n) be a uniformly sampled aperiodic electromagnetic signal data sequence of length n, representing two electromagnetic data sequences to be matched, where N and n are both positive integers; and perform the following steps:

[0011] S1, calculate the search length L of the electromagnetic signals D1(N) and D2(n) to be matched using N and n, and set the initial displacement K;

[0012] S2, use N, n and L to calculate the matching search granularity M, where M is a positive integer;

[0013] S3, calculate the matching degree S(i) of D2(N) corresponding to M relative to D1(n) corresponding to M for different displacements, i = 0, 1..L-1, and find the minimum value of L S(i) and its index k. Let S(k) be the minimum value, and k be an integer between 0 and L-1.

[0014] S4, calculate the new initial displacement K = (k-1)*M, and calculate the new matching search granularity M;

[0015] S5. Repeat steps S3 to S4 until the matching degree search granularity M meets the accuracy requirements or reaches the minimum value, and calculate the minimum matching degree S(k) as the matching degree of the two signals to be matched. Subsequent processing will use this matching degree to complete the electromagnetic data matching during target detection or recognition.

[0016] Furthermore, if the electromagnetic signal data involved in the matching is two-dimensional aperiodic data, then steps S1 to S5 are completed from the two dimensions of the two-dimensional data matrix respectively.

[0017] Furthermore, in step S1, L = b*(N+n) / n; b in the formula is adjusted according to the actual data, and is generally set to between 3 and 5.

[0018] Furthermore, in step S2, M = (Nn) / L.

[0019] Further, in step S3, a total of L matching degrees are calculated: S(i)=∑(D1) M*i+j+K -D2 j )*Wj, where Wj is the weighting function, j = 0, 1, 2…n-1; where D1 and D2 are two electromagnetic data sequences to be matched, and K in the subscript is the initial displacement.

[0020] Furthermore, in step S4, the new search granularity is a multiple of the original search granularity M: c*M / L, and the new search granularity is a positive integer; c in the formula is adjusted according to the actual data and set to between 3 and 5, keeping it consistent with the value of b in S1.

[0021] Furthermore, in step S1, the initial displacement K = 0 is set.

[0022] Furthermore, if the electromagnetic signal data to be matched is two-dimensional data, the initial displacement and matching search granularity of the two dimensions are calculated in steps S1 and S2 respectively, and are set as L1, L2, M1, M2 respectively.

[0023] In step S3, the matching degree S(i) will be the matching degree of the two-dimensional data. Let the obtained one-dimensional data matching degree arrays be S1 and S2, then the matching degree S(i) is a function of S1 and S2, that is, let: S(i) = f(S1, S2), and the minimum value of S(i) obtained in step S3 will be determined by f(S1, S2), corresponding to two integers, let k1 and k2 respectively;

[0024] The initial displacement and matching search granularity updated in step S4 are updated separately for the data in two dimensions: the first dimension: j = (k1-1)*L1 and the second dimension: j = (k2-1)*L2.

[0025] A signal processing method includes a data acquisition step and a data processing step;

[0026] In the data acquisition step, electronic devices are used to detect electromagnetic signals, and D1(N) is a uniformly discrete aperiodic electromagnetic signal sequence of length N, and D2(n) is a uniformly discrete aperiodic electromagnetic signal sequence of length n, where N and n are both positive integers.

[0027] In the data processing step, the electromagnetic data matching degree calculation method described above is executed.

[0028] A signal processing system includes an electromagnetic signal receiving device and a computer system. The electromagnetic signal receiving device is equipped with a sensor to collect electromagnetic signal data. The computer system's memory stores a program. When the program is loaded by the computer system's processor, it executes the electromagnetic data matching degree calculation method described above.

[0029] The beneficial effects of this invention include:

[0030] (1) This invention solves the problem of slow calculation speed of electromagnetic signal data in electronic reconnaissance system, relies on fewer resources, has high real-time performance, can improve the accuracy and speed of matching degree of electromagnetic signal non-periodic data, and at the same time, by using the non-periodic data itself, adopting the method of gradually approximating and gradually adjusting the matching accuracy, it avoids the extra amount of computation of feature transformation, greatly reduces the number of iterations of the calculation, and at the same time ensures the accuracy of matching.

[0031] (2) This invention greatly reduces the number of iterations of electromagnetic data in scenarios with huge data volume and high real-time requirements, while ensuring the accuracy of matching.

[0032] (3) This invention reduces the amount of communication and computation for non-periodic electromagnetic data collected by sensors, making the data matching calculation more real-time, consuming less resources, and ensuring matching accuracy. Attached Figure Description

[0033] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0034] Figure 1 The first set of actual electromagnetic signal data collected and transformed by the electronic reconnaissance system from the actual radar in this embodiment of the invention: the left and right figures are D1 and D2 respectively, N = 2048, and n is about 92;

[0035] Figure 2 This is the effect after matching the first set of actual electromagnetic signal data collected and transformed by the electronic reconnaissance system with the actual radar in an embodiment of the present invention;

[0036] Figure 3 The second set of actual electromagnetic signal data collected by the sensor in this embodiment of the invention: the left and right figures are D1 and D2 respectively, N = 2048, and n is about 143;

[0037] Figure 4This is the effect after matching the second set of actual electromagnetic signal data collected by the sensor in this embodiment of the invention;

[0038] Figure 5 This is a flowchart of the method steps in an embodiment of the present invention. Detailed Implementation

[0039] The invention will be further described below with reference to the accompanying drawings and embodiments. All features disclosed in all embodiments of this specification, or steps implied in all disclosed methods or processes, may be combined and / or extended or substituted in any way, except for mutually exclusive features and / or steps. The invention will be further described below with reference to the accompanying drawings and embodiments.

[0040] In order to improve the accuracy and speed of matching of non-periodic electromagnetic signal data in electronic reconnaissance systems, the present invention provides the following processing method.

[0041] In the electronic reconnaissance system embodiment, to address the issues of high computational load and difficulty in convergence during the matching degree calculation process for aperiodic data, a gradual approximation and adjustment of matching accuracy approach is adopted, thereby achieving the effect of rapid convergence with a smaller computational load. For two electromagnetic signal objects D1(N) and D2(n) to be matched, let them be two uniformly discrete aperiodic digital sequences of length N and n, respectively, where N and n are both positive integers. > In practice, N is much larger than n. In some cases where weighting is required, let the weighting function be W(i), i = 0, 1, 2, ..., n. If there is no weighting function, then let W(i) = 1. This embodiment includes the following steps:

[0042] Step (1): Calculate the search length and initial displacement of the data to be matched: Search length L = b*(N+n) / n, where N and n are the lengths of the objects to be matched; Initial displacement K = 0; b in the formula is adjusted according to the actual data, and is generally set to between 3 and 5.

[0043] Step (2): Calculate the matching degree search granularity: M = (Nn) / L; where M is a positive integer.

[0044] Step (3): Calculate the matching degree of D2 relative to D1 for different displacements according to the predetermined granularity. For i = 0, 1, ..., L-1, calculate a total of L matching degrees: S(i) = ∑(D1 M*i+j+K -D2 j )*Wj. Where j=0,1,2…n-1. Calculate a total of L S(i) values, and find the minimum value of the L S(i) and its index k. Assume that S(k) is the minimum value (k is an integer between 0 and L-1).

[0045] Step (4): Calculate the new initial displacement K = (k-1)*M, and calculate the new search granularity, which is a multiple of the original granularity M: c*M / L, take a positive integer, and c is adjusted according to the actual data, generally set to between 3 and 5.

[0046] Step (5): Repeat steps (3) to (4) until the matching granularity M meets the accuracy requirements or reaches the minimum value of 1, and calculate the minimum value S(k) as the optimal matching degree. The above steps are the method for calculating the matching degree of one-dimensional data.

[0047] If the matched aperiodic data is two-dimensional, then in steps (1), (2), and (3), the initial displacement, search granularity, and matching degree of the two dimensions need to be calculated respectively. Let them be L1, L2, M1, and M2 respectively.

[0048] Meanwhile, the matching degree in step (3) will be the matching degree of the two-dimensional data (let the obtained one-dimensional data matching degree arrays be S1 and S2 respectively) S is a function of S1 and S2.

[0049] Let S = f(S1, S2). Therefore, the minimum value of S obtained in step (3) will be determined by f(S1, S2), which will correspond to two integers, assuming they are k1 and k2 respectively.

[0050] Correspondingly, the initial displacement and search granularity updated in step (4) also need to be updated separately for the data in both dimensions. That is:

[0051] j=(k1-1)*L1

[0052] And j = (k² - 1) * L²

[0053] For two objects D1(N) and D2(n) with lengths N and n respectively, let the computational cost of calculating the matching degree once be X. According to the smallest granularity case, the computational cost of point-by-point matching is approximately X*(Nn).

[0054] The computational workload according to the method of this invention is approximately:

[0055] X*3*[(N+n) / n]*(int)(lg3[(Nn)*n / (N+n) / 6]))

[0056] In practical electronic reconnaissance systems, taking a typical example of N=4096 and n=256, the computational cost of the point-to-point matching scheme is approximately 3840*X, while the time consumption of the scheme of this invention is approximately 153*X, a reduction of 25 times in computational cost. If N=2048 and n=128, the computational cost of the system using the point-to-point matching scheme is approximately 1920*X, while the time consumption of the electronic reconnaissance system using the scheme of this invention is approximately 102*X, a reduction of 18 times in computational cost. Therefore, this invention can significantly reduce the computational cost of electromagnetic signal data detected by electronic reconnaissance systems. In the case of two-dimensional data, this invention will bring even more significant advantages compared to one-dimensional data, with the efficiency improvement multiplying.

[0057] In other embodiments of the invention, matching degree calculations are performed on 50 pairs of aperiodic electromagnetic data from two sets of actual radar data acquired by an actual electronic reconnaissance system. Assume the data length of the first set of data D1 acquired in the actual system is 2048, and the length of the second set of data D2 acquired in the actual system is 92. + 5. Data pairs, such as Figure 1 The second set of data, D2, has a data degree of 2048 and a length of approximately 143. + 5. Data pairs, such as Figure 3 There are approximately 100 pairs of data in two categories. The method of this invention is used to visualize the matched data, and the visualization effect is as follows: Figure 2 , Figure 4 As shown.

[0058] For the first set of data, the conventional matching method requires approximately 1950 calculations; for the second set of data, it requires approximately 1900 calculations. For two typical sets of actual electronic reconnaissance system data, this invention reduces the computational load by about 20 times compared to traditional methods. Detailed data records are as follows:

[0059]

[0060] In summary, the method proposed in this invention can be applied to the calculation of matching degree of aperiodic electromagnetic data with a large amount of data. Compared with traditional methods, it greatly improves the convergence speed and computational efficiency, making the electronic reconnaissance system more real-time and less resource-intensive when performing electromagnetic data matching, while maintaining more accurate and efficient matching degree.

[0061] Example 1

[0062] like Figure 5 As shown, an electromagnetic data matching degree calculation method utilizes an electromagnetic signal receiving device to receive and sample electromagnetic signals, and then calculates the matching degree when performing target detection or recognition using the electromagnetic signal data; subsequently, the matching degree is used to complete the target detection or recognition processing; the calculation of the matching degree when performing target detection or recognition using electromagnetic signal data specifically includes the following steps:

[0063] Let D1(N) be a uniformly sampled aperiodic electromagnetic signal data sequence of length N, and D2(n) be a uniformly sampled aperiodic electromagnetic signal data sequence of length n, representing two electromagnetic data sequences to be matched, where N and n are both positive integers; and perform the following steps:

[0064] S1, calculate the search length L of the electromagnetic signals D1(N) and D2(n) to be matched using N and n, and set the initial displacement K;

[0065] S2, use N, n and L to calculate the matching search granularity M, where M is a positive integer;

[0066] S3, calculate the matching degree S(i) of D2(N) corresponding to M relative to D1(n) corresponding to M for different displacements, i = 0, 1..L-1, and find the minimum value of L S(i) and its index k. Let S(k) be the minimum value, and k be an integer between 0 and L-1.

[0067] S4, calculate the new initial displacement K = (k-1)*M, and calculate the new matching search granularity M;

[0068] S5. Repeat steps S3 to S4 until the matching degree search granularity M meets the accuracy requirements or reaches the minimum value, and calculate the minimum matching degree S(k) as the matching degree of the two signals to be matched. Subsequent processing will use this matching degree to complete the electromagnetic data matching during target detection or recognition.

[0069] Example 2

[0070] Based on Example 1, this method is applicable to cases where the data to be matched is two-dimensional data. If the electromagnetic signal data involved in the matching is two-dimensional non-periodic data, then steps S1 to S5 are completed from the two dimensions of the two-dimensional data matrix respectively.

[0071] Example 3

[0072] Based on Example 1, in step S1, L = b*(N+n) / n; b in the formula is adjusted according to the actual data, and is generally set to between 3 and 5.

[0073] Example 4

[0074] Based on Example 1, in step S2, M = (Nn) / L.

[0075] Example 5

[0076] Based on Example 1, in step S3, a total of L matching degrees are calculated: S(i)=∑(D1) M*i+j+K -D2 j)*Wj, where Wj is the weighting function, j = 0, 1, 2…n-1; where D1 and D2 are two electromagnetic data sequences to be matched, and K in the subscript is the initial displacement.

[0077] Example 6

[0078] Based on Example 1, in step S4, the new search granularity is a multiple of the original search granularity M: c*M / L, and the new search granularity is a positive integer; c in the formula is adjusted according to the actual data and set to between 3 and 5, keeping it consistent with the value of b in S1.

[0079] Example 7

[0080] Based on Example 1, in step S1, the initial displacement K = 0 is set.

[0081] Example 8

[0082] Based on Example 2, if the electromagnetic signal data to be matched is two-dimensional data, the initial displacement and matching search granularity of the two dimensions are calculated in steps S1 and S2 respectively, and are set as L1, L2, M1, M2 respectively.

[0083] In step S3, the matching degree S(i) will be the matching degree of the two-dimensional data. Let the obtained one-dimensional data matching degree arrays be S1 and S2, then the matching degree S(i) is a function of S1 and S2, that is, let: S(i) = f(S1, S2), and the minimum value of S(i) obtained in step S3 will be determined by f(S1, S2), corresponding to two integers, let k1 and k2 respectively;

[0084] The initial displacement and matching search granularity updated in step S4 are updated separately for the data in two dimensions: the first dimension: j = (k1-1)*L1 and the second dimension: j = (k2-1)*L2.

[0085] Example 9

[0086] A signal processing method includes a data acquisition step and a data processing step;

[0087] In the data acquisition step, electronic devices are used to detect electromagnetic signals, and D1(N) is a uniformly discrete aperiodic electromagnetic signal sequence of length N, and D2(n) is a uniformly discrete aperiodic electromagnetic signal sequence of length n, where N and n are both positive integers.

[0088] In the data processing step, the electromagnetic data matching degree calculation method described in any one of Examples 1 to 8 is executed.

[0089] Example 10

[0090] A signal processing system includes an electromagnetic signal receiving device and a computer system. The electromagnetic signal receiving device is equipped with a sensor to collect electromagnetic signal data. The computer system's memory stores a program. When the program is loaded by the computer system's processor, it executes the electromagnetic data matching degree calculation method as described in any of Examples 1 to 8.

[0091] The units described in the embodiments of the present invention can be implemented in software or hardware, and the described units can also be located in a processor. The names of these units do not necessarily limit the specific unit itself.

[0092] According to one aspect of this application, a computer program product or computer program is provided, comprising computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the methods provided in the various alternative implementations described above.

[0093] In another aspect, this application also provides a computer-readable medium, which may be included in the electronic device described in the above embodiments; or it may exist independently and not assembled into the electronic device. The computer-readable medium carries one or more programs, which, when executed by the electronic device, cause the electronic device to perform the methods described in the above embodiments.

[0094] All parts not covered in this invention are the same as or can be implemented using existing technologies.

[0095] The above technical solution is only one embodiment of the present invention. For those skilled in the art, based on the application methods and principles disclosed in the present invention, it is easy to make various types of improvements or modifications, and not limited to the methods described in the above specific embodiments of the present invention. Therefore, the methods described above are only preferred and are not restrictive.

[0096] In addition to the examples above, other embodiments may be obtained by those skilled in the art based on the above disclosure or by making modifications using knowledge or technology in related fields. The features of each embodiment may be interchanged or replaced. Modifications and changes made by those skilled in the art that do not depart from the spirit and scope of the present invention should be within the protection scope of the appended claims.

Claims

1. A method of calculating electromagnetic data match-up, characterized by, Electromagnetic signals are received and sampled using electromagnetic signal receiving equipment. A matching degree is then calculated when using the electromagnetic signal data for target detection or recognition. This matching degree is then used to complete subsequent target detection or recognition processing. The calculation of the matching degree when using electromagnetic signal data for target detection or recognition specifically includes the following steps: Let D1(N) be a uniformly sampled aperiodic electromagnetic signal data sequence of length N, and D2(n) be a uniformly sampled aperiodic electromagnetic signal data sequence of length n, representing two electromagnetic data sequences to be matched, where N and n are both positive integers; and perform the following steps: S1, calculate the search length L of the electromagnetic signals D1(N) and D2(n) to be matched using N and n, and set the initial displacement K; S2, use N, n and L to calculate the matching search granularity M, where M is a positive integer; S3, calculate the matching degree S(i) of the displacement of D1(N) corresponding to M relative to D2(n) corresponding to M, i=0,1..L-1, and find the minimum value of L S(i) and its index k. Let S(k) be the minimum value, and k be an integer between 0 and L-1. In step S3, a total of L matching degrees are calculated: S(i) = ∑(D1) M*i+j+K -D2 j )*Wj, where Wj is the weighting function, j=0,1,2…n-1; where D1 and D2 are two electromagnetic data sequences to be matched, and K in the subscript is the initial displacement; S4, calculate the new initial displacement K = (k-1) * M, and calculate the new matching search granularity M; S5. Repeat steps S3 to S4 until the matching degree search granularity M meets the accuracy requirements or reaches the minimum value, and calculate the minimum matching degree S(k) as the matching degree of the two signals to be matched. Subsequent processing will use this matching degree to complete the electromagnetic data matching during target detection or recognition. Where L = b * (N + n) / n; b in the formula is adjusted according to the actual data and is set to be between 3 and 5, and the initial displacement K is set to 0; Where M = (Nn) / L; The new search granularity is a multiple of the original search granularity M: c * M / L, and the new search granularity is a positive integer; c in the formula is adjusted according to the actual data and is set to between 3 and 5, which keeps it consistent with the value of b in S1.

2. A signal processing method, characterized by, This includes data acquisition steps and data processing steps; In the data acquisition step, electronic devices are used to detect electromagnetic signals, and D1(N) is a uniformly discrete aperiodic electromagnetic signal sequence of length N, and D2(n) is a uniformly discrete aperiodic electromagnetic signal sequence of length n, where N and n are both positive integers. In the data processing step, the electromagnetic data matching degree calculation method as described in claim 1 is executed.

3. A signal processing system, characterized by It includes an electromagnetic signal receiving device and a computer system. The electromagnetic signal receiving device is equipped with a sensor to collect electromagnetic signal data. The computer system's memory stores a program. When the program is loaded by the computer system's processor, it executes the electromagnetic data matching degree calculation method as described in claim 1.

Images