Identification confidence evaluation method and device based on fuzzy set
By calculating the identification confidence scores of frequency, repetition rate, and pulse width using fuzzy set theory and then summing them by weight, the problem of not providing identification confidence scores in existing technologies is solved, thereby improving the accuracy of model identification and the decision-making efficiency of the system.
Patent Information
- Application Number
- CN202411182802.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-27
- Publication Date
- 2024-11-29
AI Technical Summary
In existing technologies, the identification results only provide the model number without giving the identification confidence level, and the identification confidence level varies for different parameters, making it difficult to provide accurate identification results.
By using a fuzzy set-based method, the recognition confidence scores in dimensions such as frequency, repetition rate, and pulse width are calculated. The scores are then weighted and summed using given weights to obtain the comprehensive recognition confidence score, which reflects the overall reliability and accuracy of the signal recognition process.
This enables effective quantitative evaluation of model identification results, improving the system's model identification capability and accuracy.
Smart Images

Figure CN119044614A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of signal processing technology, and in particular to a recognition confidence evaluation method and device based on fuzzy sets. Background Art
[0002] The recognition confidence reflects the matching degree between the electromagnetic parameters and the model recognition library, and embodies the accuracy of the model recognition result. The recognition result confidence is obtained according to the matching degree between the signal characteristic parameters and the recognition library parameters, which is of great significance for improving the recognition accuracy and enhancing the signal processing capability of the system.
[0003] However, in the prior art, the traditional recognition results only give the recognition model, but not the recognition confidence. If there is a successful match between electromagnetic parameters and multiple recognition libraries (recognition fuzziness), but there is no recognition confidence, it will be difficult to give an accurate recognition result. Secondly, even if the recognition confidence is given, since different parameters correspond to different recognition confidences, how to refer to different recognition confidences also becomes a problem. Summary of the invention
[0004] The purpose of this application is to overcome the existing technical defects and provide a recognition confidence evaluation method and device based on fuzzy sets, which comprehensively considers the recognition confidence in different dimensions such as frequency, repetition rate, pulse width, and obtains the comprehensive recognition confidence by weighted summation of the given weights of each dimension, intuitively reflecting the overall credibility and accuracy of the entire signal recognition process.
[0005] The purpose of this application is achieved through the following technical solutions:
[0006] In a first aspect, the present application proposes a recognition confidence assessment method based on fuzzy sets, the method comprising:
[0007] Calculate the frequency tolerance, repetition frequency tolerance and pulse width tolerance of each group of signal parameters to be identified;
[0008] Calculate the corresponding frequency recognition confidence, repetition frequency recognition confidence and pulse width recognition confidence respectively according to the frequency tolerance, repetition frequency tolerance and pulse width tolerance;
[0009] According to the frequency recognition confidence, the repetition frequency recognition confidence and the pulse width recognition confidence, the corresponding frequency dimension recognition confidence, the repetition frequency dimension recognition confidence and the pulse width dimension recognition confidence are calculated respectively;
[0010] The comprehensive recognition confidence is calculated based on the recognition confidence in the frequency dimension, the recognition confidence in the repetition frequency dimension and the recognition confidence in the pulse width dimension.
[0011] In one possible implementation, the frequency tolerance σ fThe calculation formula is: Where τ is the pulse width corresponding to the frequency, and SNR is the signal-to-noise ratio;
[0012] Repetition frequency tolerance σ pri The calculation formula is: Where pri is the repetition frequency center value of the signal to be identified;
[0013] Pulse width tolerance σ w The calculation formula is: Where B = min(B s ,B z ), B s is the instantaneous bandwidth of the signal, B z is the bandwidth of the reconnaissance receiver.
[0014] In a possible implementation, the frequency identification confidence IdentConfid f_i The calculation formula is: Where f is the frequency center value of the signal to be identified, f0 is the frequency center value corresponding to the radiation source library, K f is the frequency correlation coefficient;
[0015] IdentConfid p_i The calculation formula is: Where pri is the repetition frequency center value of the signal to be identified, pri0 is the repetition frequency center value corresponding to the radiation source library, and K pri is the repetition frequency correlation coefficient;
[0016] Pulse width identification confidence IdentConfid w_i The calculation formula is Where pw is the pulse width center value of the signal to be identified, pw0 is the pulse width center value corresponding to the radiation source library, K w is the pulse width correlation coefficient.
[0017] In a possible implementation, the frequency dimension identification confidence IdentConfid rf The calculation formula is: N rf is the number of frequency groups loaded by the radiation source library, M rf The number of frequency groups that are successfully matched to the radiation source library;
[0018] Repetition Dimension IdentConfid pri The calculation formula is: N pri is the number of repetition frequency groups loaded into the radiation source library, M pri The number of repetition frequency groups that are successfully matched to the radiation source library;
[0019] Pulse width dimension identification confidence IdentConfid pw The calculation formula is: N pw is the number of pulse width groups loaded into the radiation source library, M pw The number of pulse width groups that successfully match the radiation source library.
[0020] In a possible implementation, the calculation formula of the comprehensive identification confidence SynIdentConfid is: SynIdentConfid=W rf ×IdentConfid rf +W pri ×IdentConfid pri +W pw ×IdentConfid pw , W rf , W pri , W pw are the identification weights of frequency, repetition rate and pulse width, respectively, rf +W pri +W pw =1.
[0021] In a second aspect, the present application proposes a recognition confidence assessment device based on fuzzy sets, the device comprising:
[0022] Tolerance calculation module, used to calculate the frequency tolerance, repetition frequency tolerance and pulse width tolerance of each group of signal parameters to be identified;
[0023] An identification confidence calculation module, used to calculate the corresponding frequency identification confidence, repetition frequency identification confidence and pulse width identification confidence according to the frequency tolerance, repetition frequency tolerance and pulse width tolerance;
[0024] A dimension recognition confidence calculation module, used to calculate the corresponding frequency dimension recognition confidence, repetition frequency dimension recognition confidence and pulse width dimension recognition confidence according to the frequency recognition confidence, repetition frequency recognition confidence and pulse width recognition confidence;
[0025] The comprehensive recognition confidence calculation module is used to calculate the comprehensive recognition confidence based on the frequency dimension recognition confidence, the repetition frequency dimension recognition confidence and the pulse width dimension recognition confidence.
[0026] In one possible implementation, the frequency tolerance σ f The calculation formula is: Where τ is the pulse width corresponding to the frequency, and SNR is the signal-to-noise ratio;
[0027] Repetition frequency tolerance σ pri The calculation formula is: Where pri is the repetition frequency center value of the signal to be identified;
[0028] Pulse width tolerance σ w The calculation formula is: Where B = min(B s ,B z ), B s is the instantaneous bandwidth of the signal, B z is the bandwidth of the reconnaissance receiver.
[0029] In a possible implementation, the frequency identification confidence IdentConfid f_i The calculation formula is: Where f is the frequency center value of the signal to be identified, f0 is the frequency center value corresponding to the radiation source library, K f is the frequency correlation coefficient;
[0030] IdentConfid p_i The calculation formula is: Where pri is the repetition frequency center value of the signal to be identified, pri0 is the repetition frequency center value corresponding to the radiation source library, and K pri is the repetition frequency correlation coefficient;
[0031] Pulse width identification confidence IdentConfid w_i The calculation formula is Where pw is the pulse width center value of the signal to be identified, pw0 is the pulse width center value corresponding to the radiation source library, K w is the pulse width correlation coefficient.
[0032] In a possible implementation, the frequency dimension identification confidence IdentConfid rf The calculation formula is: N rf is the number of frequency groups loaded by the radiation source library, M rf The number of frequency groups that are successfully matched to the radiation source library;
[0033] Repetition Dimension IdentConfid pri The calculation formula is: N pri is the number of repetition frequency groups loaded into the radiation source library, M pri The number of repetition frequency groups that are successfully matched to the radiation source library;
[0034] Pulse width dimension identification confidence IdentConfid pw The calculation formula is: N pw is the number of pulse width groups loaded into the radiation source library, M pw The number of pulse width groups that successfully match the radiation source library.
[0035] In a possible implementation, the calculation formula of the comprehensive identification confidence SynIdentConfid is: SynIdentConfid=W rf ×IdentConfid rf +W pri ×IdentConfid pri +W pw ×IdentConfid pw , W rf , W pri , W pw are the identification weights of frequency, repetition rate and pulse width, respectively, rf +W pri +W pw =1.
[0036] The above-mentioned main scheme of the present application and its further options can be freely combined to form multiple schemes, all of which are schemes that can be adopted and claimed for protection in the present application; and in the present application, (non-conflicting options) options and other options can also be freely combined. After understanding the scheme of the present application, those skilled in the art can understand that there are multiple combinations based on the prior art and common knowledge, all of which are technical schemes to be protected by the present application, and they are not exhaustively listed here.
[0037] The present application discloses a recognition confidence evaluation method and device based on fuzzy sets, firstly calculating the frequency tolerance, repetition frequency tolerance and pulse width tolerance of each group of signal parameters to be identified, then calculating the corresponding frequency recognition confidence, repetition frequency recognition confidence and pulse width recognition confidence respectively, then calculating the corresponding frequency dimension recognition confidence, repetition frequency dimension recognition confidence and pulse width dimension recognition confidence respectively, and calculating the comprehensive recognition confidence according to the frequency dimension recognition confidence, repetition frequency dimension recognition confidence and pulse width dimension recognition confidence. By comprehensively considering the recognition confidence in different dimensions such as frequency, repetition frequency and pulse width, and obtaining the comprehensive recognition confidence by weighted summation with given weights of each dimension, the overall credibility and accuracy of the whole signal recognition process are intuitively reflected, and the effective quantitative evaluation of the model recognition results is realized, which improves the model recognition capability of the system. BRIEF DESCRIPTION OF THE DRAWINGS
[0038] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.
[0039] Figure 1 A flow chart of the recognition confidence assessment method proposed in an embodiment of the present application is shown.
[0040] Figure 2 Another flowchart of the recognition confidence assessment method proposed in an embodiment of the present application is shown. DETAILED DESCRIPTION
[0041] The following describes the embodiments of the present application through specific examples, and those skilled in the art can easily understand other advantages and effects of the present application from the contents disclosed in this specification. The present application can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present application. It should be noted that the following embodiments and features in the embodiments can be combined with each other without conflict.
[0042] Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making any creative work shall fall within the scope of protection of this application.
[0043] In the prior art, the traditional recognition results only give the recognition model, but not the recognition confidence. If there is a successful match between electromagnetic parameters and multiple recognition libraries (recognition fuzziness), but there is no recognition confidence, it will be difficult to give an accurate recognition result. Secondly, even if the recognition confidence is given, since the recognition confidence corresponding to different parameters is different, how to refer to different recognition confidences also becomes a problem.
[0044] Therefore, in order to solve the above-mentioned technical problems, the embodiment of the present application proposes a recognition confidence assessment method and device based on fuzzy sets. Based on the measurement error theory of electromagnetic parameters and the fuzzy set theory, the recognition confidence of each group of electromagnetic parameters is calculated according to the received electromagnetic parameters and the loaded radiation source library parameters, and then the recognition confidence of each dimension is calculated according to the actual number of matching groups and the number of loaded groups. Finally, the comprehensive recognition confidence is calculated according to the recognition weights of each dimension and the recognition confidence, so as to improve the evaluation ability of model recognition results and improve the perception decision-making level of the system. The recognition confidence assessment method and device are described below.
[0045] Please refer to Figure 1 , Figure 1 The flowchart of the recognition confidence evaluation method proposed in the embodiment of the present application is shown. The method can be applied to electromagnetic signal recognition scenarios. The parameter dimensions involved in the recognition are set as frequency, repetition rate, and pulse width. The recognition confidence of each group of parameters is calculated based on the recognition parameters of frequency, repetition rate, pulse width, and intra-pulse modulation characteristics. The confidence is corrected according to the number of matching parameter groups in each dimension and the number of loaded feature parameter groups. Finally, the recognition confidence of each dimension is weighted and summed according to the weight of each parameter dimension. Figure 1 , Figure 2 Another flow chart of the recognition confidence evaluation method proposed in the embodiment of the present application is shown. The recognition confidence evaluation method includes:
[0046] Step S100, calculating the frequency tolerance, repetition frequency tolerance and pulse width tolerance of each group of signal parameters to be identified.
[0047] Frequency tolerance σ f The calculation formula is: Where τ is the pulse width corresponding to the frequency, and SNR is the signal-to-noise ratio;
[0048] Repetition frequency tolerance σ pri The calculation formula is: Where pri is the repetition frequency center value of the signal to be identified;
[0049] Pulse width tolerance σ w The calculation formula is: Where B = min(B s ,B z ), B s is the instantaneous bandwidth of the signal, B z is the bandwidth of the reconnaissance receiver.
[0050] The frequency tolerance, repetition rate tolerance and pulse width tolerance of each set of signal parameters to be identified are calculated to help quantify the difference between the signal characteristics and the preset parameters in the database.
[0051] Frequency Tolerance refers to the allowable deviation between the center frequency of the signal to be identified and the preset frequency of the corresponding signal in the reference radiation source library. When calculating, the frequency drift caused by measurement errors and environmental factors is usually taken into account, and the frequency matching degree is evaluated by comparing whether the difference between the two frequencies falls within the preset tolerance range.
[0052] Pulse Repetition Interval Tolerance: also known as pulse repetition period tolerance, refers to the deviation range between the time intervals between adjacent pulses in the pulse sequence of the signal to be identified and the corresponding parameters in the reference library. By analyzing the time characteristics of the pulse sequence, the signal identity can be further confirmed.
[0053] Pulse Width Tolerance: Pulse width refers to the duration of a single pulse. Pulse width tolerance measures whether the difference between the pulse width of the signal to be identified and the pulse width of the corresponding signal in the standard library is within the allowable range, which helps to distinguish signals from different devices or modes.
[0054] In actual operation, the above tolerance needs to be determined according to specific system requirements, signal types and application scenarios, and a recognition confidence value is assigned to each parameter in combination with fuzzy set theory. By combining the recognition confidence of these parameters and the recognition weights of each dimension, the final comprehensive recognition confidence can be obtained, thereby improving the accuracy of model recognition and the decision-making efficiency of the system.
[0055] Step S200: Calculate the corresponding frequency recognition confidence, repetition frequency recognition confidence and pulse width recognition confidence respectively according to the frequency tolerance, repetition frequency tolerance and pulse width tolerance.
[0056] Frequency identification confidence IdentConfid f_i The calculation formula is: Where f is the frequency center value of the signal to be identified, f0 is the frequency center value corresponding to the radiation source library, K f is the frequency correlation coefficient.
[0057] If the frequency type is fixed, group change or jump change, the frequency correlation coefficient K f =1. If the frequency type is agile, the frequency agility range is [f min ,f max ], the frequency variation range of the radiation source library is [f lib_min ,f lib_max ], and there is overlap between the two, then the calculation formula of the correlation coefficient Kf is
[0058] IdentConfid p_i The calculation formula is: Where pri is the repetition frequency center value of the signal to be identified, pri0 is the repetition frequency center value corresponding to the radiation source library, and K pri is the repetition frequency correlation coefficient.
[0059] If the repetition rate type is fixed, group variable or staggered, then K pri =1. If the repetition type is jitter, the jitter range is [pri min ,pri max ], the frequency range of the radiation source library is [pri lib_min ,pri lib_max ], and there is overlap between the two, then the calculation formula of the correlation coefficient Kf is:
[0060] Pulse width identification confidence IdentConfid w_i The calculation formula is Where pw is the pulse width center value of the signal to be identified, pw0 is the pulse width center value corresponding to the radiation source library, K w is the pulse width correlation coefficient.
[0061] Step S300: Calculate the corresponding frequency dimension recognition confidence, repetition frequency dimension recognition confidence and pulse width dimension recognition confidence respectively according to the frequency recognition confidence, repetition frequency recognition confidence and pulse width recognition confidence.
[0062] Frequency dimension identification confidence IdentConfid rf The calculation formula is: N rf is the number of frequency groups loaded by the radiation source library, M rf IdentConfid is the number of frequency groups that successfully match the radiation source library. f_i ,i=1,2,…,M rf .
[0063] Repetition Dimension IdentConfid pri The calculation formula is: N pri is the number of repetition frequency groups loaded into the radiation source library, M pri IdentConfid is the number of repetition frequency groups that successfully match the radiation source library. p_i ,i=1,2,…,M pri .
[0064] Pulse width dimension identification confidence IdentConfid pw The calculation formula is: N pw is the number of pulse width groups loaded into the radiation source library, M pw The number of pulse width groups that successfully match the radiation source library, IdentConfid w_i ,i=1,2,…,M pw .
[0065] Step S400: Calculate the comprehensive recognition confidence according to the recognition confidence in the frequency dimension, the recognition confidence in the repetition frequency dimension and the recognition confidence in the pulse width dimension.
[0066] The calculation formula of comprehensive identification confidence SynIdentConfid is: SynIdentConfid=W rf ×IdentConfid rf +W pri ×IdentConfid pri +W pw ×IdentConfid pw , W rf , W pri , W pw are the identification weights of frequency, repetition rate and pulse width, respectively, rf +W pri +W pw=1.
[0067] In a possible implementation, the frequency, repetition rate and pulse width of the simulation signal parameters and the corresponding radiation source library parameters are set. First, the respective tolerance values are calculated according to the signal-to-noise ratio, and then the frequency recognition confidence, repetition rate recognition confidence and pulse width recognition confidence of each group are calculated, and finally the comprehensive recognition confidence is calculated according to the weight of each dimension. Table 1 shows the recognition confidence of each dimension parameter and the comprehensive recognition confidence:
[0068] Table 1
[0069]
[0070]
[0071] Compared with the prior art, the embodiments of the present application have the following beneficial effects:
[0072] Based on fuzzy set theory and combined with the measurement error theory of the frequency, repetition rate and pulse width of electromagnetic targets, the tolerance of each group of frequency, repetition rate and pulse width is first calculated, and then the recognition confidence of frequency, repetition rate and pulse width is calculated according to the center value and range of the radiation source identification library. Finally, the comprehensive confidence of model identification is obtained according to the identification dimension weights of frequency, repetition rate and pulse width. This realizes the effective quantitative evaluation of model identification results and improves the model identification capability of the system.
[0073] A possible implementation of a recognition confidence assessment device based on fuzzy sets is given below, which is used to execute the various execution steps and corresponding technical effects of the recognition confidence assessment method shown in the above embodiment and possible implementation. The device includes:
[0074] Tolerance calculation module, used to calculate the frequency tolerance, repetition frequency tolerance and pulse width tolerance of each group of signal parameters to be identified;
[0075] An identification confidence calculation module, used to calculate the corresponding frequency identification confidence, repetition frequency identification confidence and pulse width identification confidence according to the frequency tolerance, repetition frequency tolerance and pulse width tolerance;
[0076] A dimension recognition confidence calculation module, used to calculate the corresponding frequency dimension recognition confidence, repetition frequency dimension recognition confidence and pulse width dimension recognition confidence according to the frequency recognition confidence, repetition frequency recognition confidence and pulse width recognition confidence;
[0077] The comprehensive recognition confidence calculation module is used to calculate the comprehensive recognition confidence based on the frequency dimension recognition confidence, the repetition frequency dimension recognition confidence and the pulse width dimension recognition confidence.
[0078] In one possible implementation, the frequency tolerance σ fThe calculation formula is: Where τ is the pulse width corresponding to the frequency, and SNR is the signal-to-noise ratio;
[0079] Repetition frequency tolerance σ pri The calculation formula is: Where pri is the repetition frequency center value of the signal to be identified;
[0080] Pulse width tolerance σ w The calculation formula is: Where B = min(B s ,B z ), B s is the instantaneous bandwidth of the signal, B z is the bandwidth of the reconnaissance receiver.
[0081] In a possible implementation, the frequency identification confidence IdentConfid f_i The calculation formula is: Where f is the frequency center value of the signal to be identified, f0 is the frequency center value corresponding to the radiation source library, K f is the frequency correlation coefficient;
[0082] IdentConfid p_i The calculation formula is: Where pri is the repetition frequency center value of the signal to be identified, pri0 is the repetition frequency center value corresponding to the radiation source library, and K pri is the repetition frequency correlation coefficient;
[0083] Pulse width identification confidence IdentConfid w_i The calculation formula is Where pw is the pulse width center value of the signal to be identified, pw0 is the pulse width center value corresponding to the radiation source library, K w is the pulse width correlation coefficient.
[0084] In a possible implementation, the frequency dimension identification confidence IdentConfid rf The calculation formula is: N rf is the number of frequency groups loaded by the radiation source library, M rf The number of frequency groups that are successfully matched to the radiation source library;
[0085] Repetition Dimension IdentConfid pri The calculation formula is: N pri is the number of repetition frequency groups loaded into the radiation source library, M pri The number of repetition frequency groups that are successfully matched to the radiation source library;
[0086] Pulse width dimension identification confidence IdentConfid pw The calculation formula is: N pw is the number of pulse width groups loaded into the radiation source library, M pw The number of pulse width groups that successfully match the radiation source library.
[0087] In a possible implementation, the calculation formula of the comprehensive identification confidence SynIdentConfid is: SynIdentConfid=W rf ×IdentConfid rf +W pri ×IdentConfid pri +W pw ×IdentConfid pw , W rf , W pri , W pw are the identification weights of frequency, repetition rate and pulse width, respectively, rf +W pri +W pw =1.
[0088] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application should be included in the protection scope of the present application.
Claims
1. A recognition confidence evaluation method based on fuzzy sets, characterized in that: The method comprises: Calculate the frequency tolerance, repetition frequency tolerance and pulse width tolerance of each group of signal parameters to be identified; Calculate the corresponding frequency recognition confidence, repetition frequency recognition confidence and pulse width recognition confidence respectively according to the frequency tolerance, repetition frequency tolerance and pulse width tolerance; According to the frequency recognition confidence, the repetition frequency recognition confidence and the pulse width recognition confidence, the corresponding frequency dimension recognition confidence, the repetition frequency dimension recognition confidence and the pulse width dimension recognition confidence are calculated respectively; The comprehensive recognition confidence is calculated based on the recognition confidence in the frequency dimension, the recognition confidence in the repetition frequency dimension and the recognition confidence in the pulse width dimension.
2. The recognition confidence assessment method according to claim 1, characterized in that: Frequency tolerance σ f The calculation formula is: Where τ is the pulse width corresponding to the frequency, and SNR is the signal-to-noise ratio; Repetition frequency tolerance σ pri The calculation formula is: Where pri is the repetition frequency center value of the signal to be identified; Pulse width tolerance σ w The calculation formula is: Where B = min(B s ,B z ), B s is the instantaneous bandwidth of the signal, B z is the bandwidth of the reconnaissance receiver.
3. The recognition confidence assessment method according to claim 2, characterized in that: Frequency identification confidence IdentConfid f_i The calculation formula is: Where f is the frequency center value of the signal to be identified, f0 is the frequency center value corresponding to the radiation source library, K f is the frequency correlation coefficient; IdentConfid p_i The calculation formula is: Where pri is the repetition frequency center value of the signal to be identified, pri0 is the repetition frequency center value corresponding to the radiation source library, and K pri is the repetition frequency correlation coefficient; Pulse width identification confidence IdentConfid w_i The calculation formula is Where pw is the pulse width center value of the signal to be identified, pw0 is the pulse width center value corresponding to the radiation source library, K w is the pulse width correlation coefficient.
4. The recognition confidence assessment method according to claim 3, characterized in that: Frequency dimension identification confidence IdentConfid rf The calculation formula is: N rf is the number of frequency groups loaded by the radiation source library, M rf The number of frequency groups that are successfully matched to the radiation source library; Repetition Dimension IdentConfid pri The calculation formula is: N pri is the number of repetition frequency groups loaded into the radiation source library, M pri The number of repetition frequency groups that are successfully matched to the radiation source library; Pulse width dimension identification confidence IdentConfid pw The calculation formula is: N pw is the number of pulse width groups loaded into the radiation source library, M pw The number of pulse width groups that successfully match the radiation source library.
5. The recognition confidence assessment method according to claim 4, characterized in that: The calculation formula of comprehensive identification confidence SynIdentConfid is: SynIdentConfid=W rf ×IdentConfid rf +W pri ×IdentConfid pri +W pw ×IdentConfid pw , W rf , W pri , W pw are the identification weights of frequency, repetition rate and pulse width, respectively, rf +W pri +W pw =1.
6. A recognition confidence evaluation device based on fuzzy sets, characterized in that: The device comprises: Tolerance calculation module, used to calculate the frequency tolerance, repetition frequency tolerance and pulse width tolerance of each group of signal parameters to be identified; An identification confidence calculation module, used to calculate the corresponding frequency identification confidence, repetition frequency identification confidence and pulse width identification confidence according to the frequency tolerance, repetition frequency tolerance and pulse width tolerance; A dimension recognition confidence calculation module, used to calculate the corresponding frequency dimension recognition confidence, repetition frequency dimension recognition confidence and pulse width dimension recognition confidence according to the frequency recognition confidence, repetition frequency recognition confidence and pulse width recognition confidence; The comprehensive recognition confidence calculation module is used to calculate the comprehensive recognition confidence based on the frequency dimension recognition confidence, the repetition frequency dimension recognition confidence and the pulse width dimension recognition confidence.
7. The recognition confidence evaluation device according to claim 6, characterized in that: Frequency tolerance σ f The calculation formula is: Where τ is the pulse width corresponding to the frequency, and SNR is the signal-to-noise ratio; Repetition frequency tolerance σ pri The calculation formula is: Where pri is the repetition frequency center value of the signal to be identified; Pulse width tolerance σ w The calculation formula is: Where B = min(B s ,B z ), B s is the instantaneous bandwidth of the signal, B z is the bandwidth of the reconnaissance receiver.
8. The recognition confidence evaluation device according to claim 7, characterized in that: Frequency identification confidence IdentConfid f_i The calculation formula is: Where f is the frequency center value of the signal to be identified, f0 is the frequency center value corresponding to the radiation source library, K f is the frequency correlation coefficient; IdentConfid p_i The calculation formula is: Where pri is the repetition frequency center value of the signal to be identified, pri0 is the repetition frequency center value corresponding to the radiation source library, and K pri is the repetition frequency correlation coefficient; Pulse width identification confidence IdentConfid w_i The calculation formula is Where pw is the pulse width center value of the signal to be identified, pw0 is the pulse width center value corresponding to the radiation source library, K w is the pulse width correlation coefficient.
9. The recognition confidence evaluation device according to claim 8, characterized in that: Frequency dimension identification confidence IdentConfid rf The calculation formula is: N rf is the number of frequency groups loaded by the radiation source library, M rf The number of frequency groups that are successfully matched to the radiation source library; Repetition Dimension IdentConfid pri The calculation formula is: N pri is the number of repetition frequency groups loaded into the radiation source library, M pri The number of repetition frequency groups that are successfully matched to the radiation source library; Pulse width dimension identification confidence IdentConfid pw The calculation formula is: N pw is the number of pulse width groups loaded into the radiation source library, M pw The number of pulse width groups that successfully match the radiation source library.
10. The recognition confidence evaluation device according to claim 9, characterized in that: The calculation formula of comprehensive identification confidence SynIdentConfid is: SynIdentConfid=W rf ×IdentConfid rf +W pri ×IdentConfid pri +W pw ×IdentConfid pw , W rf , W pri , W pw are the identification weights of frequency, repetition rate and pulse width, respectively, rf +W pri +W pw =1.
