CN114791593A - Artificial intelligence signal analysis and recognition system - Google Patents
Artificial intelligence signal analysis and recognition system Download PDFInfo
- Publication number
- CN114791593A CN114791593A CN202210317876.3A CN202210317876A CN114791593A CN 114791593 A CN114791593 A CN 114791593A CN 202210317876 A CN202210317876 A CN 202210317876A CN 114791593 A CN114791593 A CN 114791593A
- Authority
- CN
- China
- Prior art keywords
- signal
- module
- signals
- radar
- identification
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
- G01S7/02—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00
- G01S7/41—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00 using analysis of echo signal for target characterisation; Target signature; Target cross-section
Abstract
The invention discloses an artificial intelligence signal analysis and recognition system, which relates to the technical field of signal recognition and comprises a data preprocessing module, a signal recognition module, a signal monitoring module, a controller and a parameter compensation module; the invention utilizes software radio technology to analyze the time domain characteristic and the frequency domain data of the signal, establishes a corresponding mathematical model, restores discrete sampling data to continuous signal characteristics, compares and analyzes the models and the characteristics of the signal, identifies the type and the attribute of the signal, further controls, demodulates and decodes the cell and interprets the information content; then the recognition result is used as a feedback factor, a mathematical model in the recognition and demodulation processes is further verified, the model and algorithm of each link are corrected, and attribute parameters are compensated, so that the mathematical model is continuously corrected and perfected, and the recognition accuracy is improved; and manual guard is not needed at all, so that the work can be widely carried out, and the continuity and effectiveness of the work are guaranteed.
Description
Technical Field
The invention relates to the technical field of signal identification, in particular to a receiving, processing and identifying technology of radio signals, and specifically relates to an artificial intelligent signal analysis and identification system.
Background
With the development of communication technology, wireless communication technology has advanced greatly in military affairs. In the field of military communication, the electronic warfare party can obtain the advantage of communication and get the first opportunity in the war. Electronic countermeasure plays an important role in electronic information reconnaissance, electronic support and threat warning systems, and radar radiation source signal identification is an important link in electronic countermeasure; with the electromagnetic environment becoming very complex in electronic countermeasure, the specific expression is that the number of radar radiation sources is many, the spatial distribution range is wide, and the aliasing of signals in time domain and frequency domain is serious. Radar signals that occur in a short period of time can be tens of thousands or even hundreds of thousands, and a large number of signals can occur at the same time at a certain time.
Present equipment mainly relies on artifically to signal analysis and discernment, needs the people's eye to watch according to the spectrum search signal, and the contrast waveform discovery signal often needs long-time tracking, observation, comparison alone, because time energy is limited, often when discovering a signal, has leaked the signal that many probably appear, to weak signal and burst signal very difficult discernment, intensity of labour is big, and it is tired to last work, often neglects. The identification of signals requires long-term expert knowledge accumulation and experience summarization, so that many jobs are not developed by experts.
Disclosure of Invention
In order to solve the problems existing in the scheme, the invention provides an artificial intelligence signal analysis and identification system. The invention utilizes software radio technology to analyze the time domain characteristic and the frequency domain data of the signal, establishes a corresponding mathematical model, restores discrete sampling data to continuous signal characteristics, compares and analyzes the models and the characteristics of the signal, judges and identifies the type and the attribute of the signal, and then controls, demodulates and decodes the cell and interprets the information content; the results of each subsequent link are used as feedback factors, and then the mathematical model in the previous identification and demodulation process is further verified, the model and algorithm of each link are corrected, and attribute parameters are compensated, so that the mathematical model is continuously corrected and perfected, the identification accuracy is continuously improved, and the whole process is automatically carried out; the invention separates the analysis and identification of the signal from manual work, automatically carries out the analysis and identification by a device program, realizes professional identification and experience judgment by an artificial intelligent algorithm, thereby leading the work to be widely carried out, ensuring the continuity and the effectiveness of the work and leading the accumulation and the summary of the experience to be more automatically carried out.
The purpose of the invention can be realized by the following technical scheme:
an artificial intelligence signal analysis and recognition system comprises a data acquisition module, a data preprocessing module, a signal recognition module, a big data reference module, a signal monitoring module, a controller, a storage module, a signal simulation module, a command input module, an abnormality testing module, an alarm module and a display module;
the data acquisition module comprises MATLAB software and is used for generating a data set of the radar signal in a simulation mode; transmitting a radar signal data set generated by simulation to a data preprocessing module, wherein the data preprocessing module is used for preprocessing a data set signal, analyzing the time domain characteristic and the frequency domain characteristic of the data set signal by using a software radio technology, and establishing a corresponding mathematical model; the specific treatment steps are as follows:
s1: firstly, outputting a data set signal in a sequence form;
s2: carrying out time-frequency transformation on the data set signals, analyzing the time domain characteristics and the frequency domain characteristics of the data set signals by a software radio technology, and outputting the data set signals in a signal time-frequency diagram form;
s3: labeling the category of the signal in the data set signals output in the two forms, randomly extracting 2400 samples from each category of signals as training samples, and taking 600 samples as test samples;
s4: training a RNN-DenseNet-based network to obtain a corresponding mathematical model M;
the data preprocessing module is used for transmitting the corresponding mathematical model M to the signal identification module;
the signal monitoring module is AD acquisition circuit for gather radar signal in real time, obtain the discrete sampling data of radar signal and transmit the discrete sampling data of radar signal to signal identification module, signal identification module receives corresponding mathematical model M and the discrete sampling data of radar signal to carry out the identification analysis to the discrete sampling data of radar signal, the discernment is based on artificial intelligence's intelligent recognition algorithm, and concrete discernment step is:
the method comprises the following steps: acquiring a corresponding mathematical model M, and restoring discrete sampling data of radar signals to continuous signal characteristics according to the mathematical model M, wherein the signal characteristics comprise time domain characteristics and frequency domain characteristics;
step two: automatically comparing and analyzing the signal characteristics obtained by reduction with a big data reference module, wherein the big data reference module stores various radar signal models and signal characteristics, and identifying and obtaining corresponding signal types and attributes based on an artificial intelligence intelligent identification algorithm;
step three: according to the signal type and attribute obtained by identification, performing control, demodulation and decoding on the radar signal to obtain a cell, and finally decoding information content;
the signal identification module is used for transmitting the decoded cells and the decoded information content to the storage module for storage through the controller.
Further, the data set includes seven signals, a regular pulse signal, a chirp signal, a non-chirp signal, a bi-phase encoded signal, a quad-phase encoded signal, a bi-frequency encoded signal, and a quad-frequency encoded signal, wherein each signal generates 3000 samples from-10 dB to 2dB per 2dB signal-to-noise ratio.
The system further comprises an information recording module, an information sorting module and a parameter compensation module, wherein the information recording module is used for recording the identification record of the signal identification module and transmitting the identification record to the information sorting module;
the information sorting module receives and sorts the identification records, constructs a parameter detection training sample, trains based on a machine learning algorithm to obtain a parameter compensation model, and transmits the parameter compensation model to the parameter compensation module, and the parameter compensation module receives the parameter compensation model and then corrects the mathematical model M and the intelligent algorithm.
Further, the identification logic of the intelligent identification algorithm is as follows:
aiming at AD real data, establishing a frequency domain model according to frequency band characteristics and a background environment, wherein input parameters comprise frequency domain resolution fd and frequency domain smoothing parameters fs;
aiming at AD real data, a time domain model is established according to frequency band characteristics and a background environment, and input parameters comprise a time domain resolution td and a time domain smoothing parameter ts;
combining the output results of the time domain model and the frequency domain model according to the weight to form a comprehensive mathematical model;
in the automatic learning stage, the characteristic attribute of the comprehensive mathematical model and the physical attribute of the actual signal are paired
Comparing and analyzing to find out the difference;
automatically adjusting the parameters of fd, fs, td and ts in the comprehensive mathematical model according to the difference, so that the comparison difference is smaller and smaller;
analyzing and identifying frequency domain data and time domain data according to the attributes of the comprehensive mathematical model, and identifying characteristics of large signal components, including a central frequency point and a signal bandwidth;
acquiring data aiming at the identified signal frequency point and bandwidth, and then demodulating, decoding and decoding;
in the machine learning stage, the results of demodulation, decoding and decoding are compared with the correct results of the calculation example, and relevant calculation parameters are adjusted according to the difference; gradually improving the accuracy and the confidence of the algorithm in each step;
in actual use, the program feeds back the deviation of the actual measurement frequency and bandwidth to an upstream recognition algorithm for correcting a mathematical model and algorithm parameters in the recognition algorithm; along with the increase of examples and the accumulation of working time, models and parameters of all links are more and more perfect, and the recognition accuracy is higher and higher; the process is similar to the process of professional knowledge and experience accumulation of artificial recognition and is close to real artificial intelligence;
further, the signal simulation module is used for simulating the radar signal and setting a target value of the radar signal, wherein the target value represents a model and signal characteristics of the radar signal;
before the signal identification module operates, a user inputs a test instruction through the instruction input module, the controller is used for transmitting the test instruction to the abnormity test module, and the abnormity test module is used for detecting the radar signal abnormity detection and abnormity processing capacity of the signal identification module after receiving the test instruction.
Further, the anomaly testing module comprises a transmitting antenna, and the specific detection method comprises the following steps:
v1: after the anomaly testing module receives the testing instruction, the signal simulation module transmits a simulated radar signal, and meanwhile, the transmitting antenna is used for transmitting an interference signal to carry out electromagnetic wave interference on the radar signal;
v2: the signal identification module receives discrete sampling data corresponding to radar signals, performs identification analysis, and outputs measured values corresponding to the radar signals, wherein the measured values comprise measured time domain characteristics and measured frequency domain characteristics of the corresponding radar signals;
v3: judging whether the measured value of the radar signal is consistent with the target value; if not, generating an interference signal; the abnormal testing module is used for transmitting the interference signal to the controller, and the controller is used for automatically driving the alarm module to give an alarm when receiving the interference signal and automatically driving the display module to display that the signal interference of the signal identification module is serious and the processing is recommended.
Further, determining whether the measured value of the radar signal is consistent with the target value specifically includes:
acquiring an actual measurement value and a target value of the radar signal, comparing the actual measurement time domain characteristic of the radar signal with the target time domain characteristic to obtain a time domain error, and marking the time domain error as W1; comparing the actually measured frequency domain characteristics of the radar signals with the target frequency domain characteristics to obtain frequency domain errors, and marking the frequency domain errors as W2;
using formulas
Calculating to obtain an error coefficient WX; wherein d1 and d2 are coefficient factors; if the error coefficient WX is less than or equal to the error coefficient threshold value, judging that the measured value is consistent with the target value; generating a normal signal; if the error coefficient WX is larger than the error coefficient threshold value, the measured value is judged to be inconsistent with the target value, and an interference signal is generated.
Compared with the prior art, the invention has the beneficial effects that:
1. in the invention, a data acquisition module generates a data set of a radar signal through MATLAB software simulation, a data preprocessing module is used for preprocessing the data set signal, and the time domain characteristic and the frequency domain characteristic of the data set signal are analyzed by using a software radio technology to establish a corresponding mathematical model; restoring discrete sampling data of radar signals to continuous signal characteristics, comparing and analyzing the due model and characteristics of the radar signals, judging and comparing, identifying the types and attributes of the signals, controlling, demodulating and decoding cells, and interpreting information content; the method has the advantages that manual guard is not needed at all, as long as the normal operation of the equipment during starting is ensured, tasks are automatically executed without being watched by people for several days, discovery signals can be automatically recorded and collected, the burden of personnel is greatly reduced, the working efficiency is improved, and the identification accuracy is also greatly improved;
2. the information sorting module receives and sorts the identification records, constructs a parameter detection training sample, and trains based on a machine learning algorithm to obtain a parameter compensation model; the parameter compensation module receives the parameter compensation model and then corrects the mathematical model M and the intelligent algorithm, the invention takes the result of each link of the signal identification module as a feedback factor, and then further verifies the mathematical model in the previous identification and demodulation processes, corrects the mathematical model and the intelligent algorithm of each link and compensates the attribute parameters, so that the mathematical model is continuously corrected and perfected, the identification accuracy is continuously improved, the whole process is automatically carried out, the work can be widely carried out, and the continuity and the effectiveness of the work are ensured;
3. before the signal identification module runs, the anomaly testing module is used for receiving a testing instruction and then detecting the radar signal anomaly detection and anomaly processing capacity of the signal identification module; after the anomaly testing module receives the testing instruction, the signal simulation module transmits a simulated radar signal, and meanwhile, the transmitting antenna is used for transmitting an interference signal to carry out electromagnetic wave interference on the radar signal; the signal identification module receives discrete sampling data corresponding to radar signals, performs identification analysis, outputs measured values corresponding to the radar signals, and judges whether the measured values of the radar signals are consistent with target values or not; if the measured value is inconsistent with the target value, generating an interference signal; the controller is used for automatically driving the alarm module to give an alarm when receiving the interference signal and automatically driving the display module to display 'the signal recognition module has serious signal interference and is recommended to be processed'; therefore, the identification efficiency and the accuracy of the signal identification module are improved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the embodiments or the prior art descriptions will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and other drawings can be obtained by those skilled in the art without creative efforts.
FIG. 1 is a schematic block diagram of an artificial intelligence signal analysis and recognition system according to the present invention.
Fig. 2 is a system block diagram of embodiment 1 of the present invention.
Fig. 3 is a system block diagram of embodiment 2 of the present invention.
Fig. 4 is a system block diagram of embodiment 3 of the present invention.
Detailed Description
The technical solutions of the present invention will be described below clearly and completely in conjunction with the embodiments, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be obtained by a person skilled in the art without making any creative effort based on the embodiments in the present invention, belong to the protection scope of the present invention.
As shown in fig. 1 to 4, an artificial intelligence signal analysis and identification system includes a data acquisition module, a data preprocessing module, a signal identification module, a signal monitoring module, a controller, a storage module, a signal simulation module, a command input module, an anomaly testing module, an alarm module, and a display module;
example 1
As shown in fig. 2; the data acquisition module comprises MATLAB software, the MATLAB software is used for generating a data set of radar signals in a simulation mode, the data set comprises seven signals of conventional pulse signals, linear frequency modulation signals, nonlinear frequency modulation signals, two-phase coded signals, four-phase coded signals, two-frequency coded signals and four-frequency coded signals, and 3000 samples are generated every 2dB signal-to-noise ratio from-10 dB to 2dB of each signal;
the data acquisition module is used for transmitting a radar signal data set generated by simulation to the data preprocessing module, the data preprocessing module is used for preprocessing a data set signal, analyzing the time domain characteristic and the frequency domain characteristic of the data set signal by using a software radio technology, and establishing a corresponding mathematical model; the specific treatment steps are as follows:
s1: firstly, outputting a data set signal in a sequence form;
s2: carrying out time-frequency transformation on the data set signals, analyzing the time domain characteristics and the frequency domain characteristics of the signals by a software radio technology, and outputting the data set signals in the form of a signal time-frequency diagram;
s3: labeling the category of the signal in the data set signals output in the two forms, randomly extracting 2400 samples from each category of signals as training samples, and taking 600 samples as test samples;
s4: training a RNN-DenseNet-based network to obtain a corresponding mathematical model M;
the data preprocessing module is used for transmitting the corresponding mathematical model M to the signal identification module;
the signal monitoring module is AD acquisition circuit for gather radar signal in real time, obtain the discrete sampling data of radar signal and transmit the discrete sampling data of radar signal to signal identification module, signal identification module receives corresponding mathematical model M and the discrete sampling data of radar signal to carry out the identification analysis to the discrete sampling data of radar signal, the discernment is based on artificial intelligence's intelligent recognition algorithm, and concrete discernment step is:
the method comprises the following steps: acquiring a corresponding mathematical model M, and restoring discrete sampling data of radar signals to continuous signal characteristics according to the mathematical model M, wherein the signal characteristics comprise time domain characteristics and frequency domain characteristics;
step two: automatically comparing and analyzing the signal characteristics obtained by reduction with a big data reference module, wherein the big data reference module stores various radar signal models and signal characteristics, and identifying and obtaining corresponding signal types and attributes based on an artificial intelligence intelligent identification algorithm;
step three: according to the identified signal type and attribute, performing control, demodulation and decoding on the radar signal to obtain a cell, and finally decoding information content;
the signal identification module is used for transmitting the decoded cell and the decoded information content to the storage module for storage through the controller;
the invention utilizes software radio technology to analyze the time domain characteristic and the frequency domain data of the signal, establishes a corresponding mathematical model, restores discrete sampling data to continuous signal characteristics, compares and analyzes the due model and the characteristics of the radar signal, judges the model and the characteristics, identifies the type and the attributes of the signal based on the artificial intelligent identification algorithm, and then controls, demodulates and decodes the cell and interprets the information content; the method does not need manual guard at all, as long as the normal operation of the equipment during starting is ensured, the task is automatically executed without being watched for several days, and the discovery signal can be automatically recorded and collected, so that the burden of personnel is greatly reduced, the working efficiency is improved, and the identification accuracy is also greatly improved;
the identification logic of the intelligent identification algorithm is as follows:
aiming at AD real data, a frequency domain model is established according to frequency band characteristics and a background environment, and input parameters comprise frequency domain resolution fd and frequency domain smoothing parameter fs;
aiming at AD real data, a time domain model is established according to frequency band characteristics and a background environment, and input parameters comprise time domain resolution td and a time domain smoothing parameter ts;
combining the output results of the time domain model and the frequency domain model according to the weight to form a comprehensive mathematical model;
in the automatic learning stage, the characteristic attribute of the comprehensive mathematical model and the physical attribute of the actual signal are paired
Comparing and analyzing to find out the difference;
automatically adjusting the parameters of fd, fs, td and ts in the comprehensive mathematical model according to the difference to ensure that the contrast difference is smaller and smaller;
analyzing and identifying frequency domain data and time domain data according to the attributes of the comprehensive mathematical model, and identifying characteristics of large signal components, including a central frequency point and a signal bandwidth;
acquiring data aiming at the identified signal frequency point and bandwidth, and then demodulating, decoding and decoding;
in the machine learning stage, the results of demodulation, decoding and decoding are compared with the correct results of the calculation example, and relevant calculation parameters are adjusted according to the difference; gradually improving the accuracy and the confidence of the algorithm in each step;
in actual use, the program feeds back the deviation of the actual measurement frequency and bandwidth to an upstream recognition algorithm for correcting a mathematical model and algorithm parameters in the recognition algorithm; along with the increase of the calculation examples and the accumulation of the working time, the model and the parameters of each link become more and more perfect, and the recognition accuracy rate becomes higher and higher; this is much like the process of professional knowledge and experience accumulation of human recognition, approaching true artificial intelligence.
Example 2
As shown in fig. 3; the system also comprises an information recording module, an information sorting module and a parameter compensation module, wherein the information recording module is used for recording the identification record of the signal identification module and transmitting the identification record to the information sorting module, and the identification record comprises identification result information, actual result information and corresponding various environment parameter values during each identification of the signal identification module; the identification result information is represented by cells decoded by the signal identification module and information content decoded by the signal identification module, the actual result information is represented by actual cells and actual information content corresponding to radar signal discrete sampling data, and various environmental parameter values comprise temperature, humidity, wind pressure, wind speed, interference signals and the like;
the information sorting module receives and sorts the identification records, constructs a parameter detection training sample, trains based on a machine learning algorithm to obtain a parameter compensation model and transmits the parameter compensation model to the parameter compensation module, and the parameter compensation module receives the parameter compensation model and then corrects a mathematical model M and an intelligent algorithm, wherein the specific compensation step is as follows:
SS 1: acquiring an identification record of a signal identification module at the current moment of the system, and inputting current identification result information, actual result information and corresponding various environmental parameter values into a parameter compensation model to obtain an attribute parameter compensation coefficient;
SS 2: correcting the mathematical model M and the intelligent algorithm according to the attribute parameter compensation coefficient, so that the mathematical model M is continuously corrected and perfected, and the identification accuracy is improved;
according to the invention, the results of each link of the signal identification module are used as feedback factors, and then the mathematical models in the previous identification and demodulation processes are further verified, the mathematical models and intelligent algorithms of each link are corrected, attribute parameters are compensated, so that the mathematical models are continuously corrected and perfected, the identification accuracy is continuously improved, and the whole process is automatically carried out, so that the work can be widely carried out, and the continuity and effectiveness of the work are guaranteed;
example 3
As shown in fig. 4; the signal simulation module is used for simulating radar signals and setting a target value of the radar signals, and the target value represents a model and signal characteristics of the radar signals;
before the signal identification module runs, a user inputs a test instruction through the instruction input module and transmits the test instruction to the controller, the controller is used for transmitting the test instruction to the abnormity test module, and the abnormity test module is used for detecting the radar signal abnormity detection and abnormity processing capability of the signal identification module after receiving the test instruction; the anomaly testing module comprises a transmitting antenna, and the specific detection method comprises the following steps:
v1: after the anomaly testing module receives the testing instruction, the signal simulation module transmits a simulated radar signal, and meanwhile, the transmitting antenna is used for transmitting an interference signal to perform electromagnetic wave interference on the radar signal;
v2: the signal identification module receives discrete sampling data corresponding to radar signals, performs identification analysis, and outputs measured values corresponding to the radar signals, wherein the measured values comprise measured time domain characteristics and measured frequency domain characteristics corresponding to the radar signals;
v3: judging whether the measured value of the radar signal is consistent with the target value; the method comprises the following specific steps:
acquiring an actual measurement value and a target value of a radar signal, wherein the target value comprises a target time domain characteristic and a target frequency domain characteristic of the radar signal;
comparing the actually measured time domain characteristic of the radar signal with the target time domain characteristic to obtain a time domain error, and marking the time domain error as W1; comparing the measured frequency domain characteristic of the radar signal with the target frequency domain characteristic to obtain a frequency domain error, and marking the frequency domain error as W2;
using a formula
Calculating to obtain an error coefficient WX; wherein d1 and d2 are coefficient factors;
comparing the error coefficient WX to an error coefficient threshold;
if the error coefficient WX is less than or equal to the error coefficient threshold value, judging that the measured value is consistent with the target value; generating a normal signal;
if the error coefficient WX is larger than the error coefficient threshold value, judging that the measured value is inconsistent with the target value, and generating an interference signal;
the abnormal testing module is used for transmitting a normal signal and an interference signal to the controller, and the controller is used for automatically driving the alarm module to give an alarm when the interference signal is received and automatically driving the display module to display 'the signal identification module has serious signal interference and suggests processing'; therefore, the identification efficiency and the accuracy of the signal identification module are improved.
The above formulas are all calculated by removing dimensions and taking numerical values thereof, the formula is a formula which is obtained by acquiring a large amount of data and performing software simulation to obtain the closest real situation, and the preset parameters and the preset threshold value in the formula are set by the technical personnel in the field according to the actual situation or obtained by simulating a large amount of data.
The working principle of the invention is as follows:
during working, firstly, a data acquisition module generates a data set of radar signals through MATLAB software simulation, a data preprocessing module is used for preprocessing the data set signals, time domain characteristics and frequency domain characteristics of the data set signals are analyzed by utilizing a software radio technology, and a corresponding mathematical model is established; the signal identification module receives the corresponding mathematical model M and the sampling data of the radar signal dispersion, carries out identification analysis on the sampling data of the radar signal dispersion, identifies an intelligent identification algorithm based on artificial intelligence, firstly obtains the corresponding mathematical model M, restores the sampling data of the radar signal dispersion to continuous signal characteristics according to the mathematical model M, automatically carries out comparison analysis with a big data reference module according to the restored signal characteristics, and identifies and obtains the corresponding signal type and attribute based on the intelligent identification algorithm of the artificial intelligence; then controlling, demodulating and decoding the radar signal to obtain a cell, and finally decoding information content; the method has the advantages that manual guard is not needed at all, as long as the normal operation of the equipment during starting is ensured, tasks are automatically executed without being watched by people for several days, discovery signals can be automatically recorded and collected, the burden of personnel is greatly reduced, the working efficiency is improved, and the identification accuracy is also greatly improved;
the information sorting module receives the identification records and sorts the identification records, a parameter detection training sample is constructed, and training is based on a machine learning algorithm to obtain a parameter compensation model; the parameter compensation module receives the parameter compensation model and then corrects the mathematical model M and the intelligent algorithm, the invention takes the results of each link of the signal identification module as a feedback factor, and then further verifies the mathematical model in the previous identification and demodulation process, corrects the mathematical model and the intelligent algorithm of each link, compensates the attribute parameters, so that the mathematical model is continuously corrected and perfected, the identification accuracy is continuously improved, the whole process is automatically carried out, the work can be widely carried out, and the continuity and the effectiveness of the work are ensured;
before the signal identification module runs, the abnormity testing module is used for receiving the testing instruction and then detecting the radar signal abnormity detection and abnormity processing capability of the signal identification module; after the anomaly testing module receives the testing instruction, the signal simulation module transmits a simulated radar signal, and meanwhile, the transmitting antenna is used for transmitting an interference signal to carry out electromagnetic wave interference on the radar signal; the signal identification module receives discrete sampling data corresponding to radar signals, performs identification analysis, outputs measured values corresponding to the radar signals, and judges whether the measured values of the radar signals are consistent with target values or not; if the measured value is not consistent with the target value, generating an interference signal; the controller is used for automatically driving the alarm module to give an alarm when receiving the interference signal and automatically driving the display module to display 'the signal recognition module has serious signal interference and is recommended to be processed'; therefore, the identification efficiency and the accuracy of the signal identification module are improved.
In the description herein, references to the description of "one embodiment," "an example," "a specific example" or the like are intended to mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, the schematic representations of the terms used above do not necessarily refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
The preferred embodiments of the invention disclosed above are intended to be illustrative only. The preferred embodiments are not intended to be exhaustive or to limit the invention to the precise forms disclosed. Obviously, many modifications and variations are possible in light of the above teaching. The embodiments were chosen and described in order to best explain the principles of the invention and the practical application, to thereby enable others skilled in the art to best utilize the invention. The invention is limited only by the claims and their full scope and equivalents.
Claims (7)
1. An artificial intelligence signal analysis and identification system is characterized by comprising a data acquisition module, a signal identification module, a big data reference module, a signal monitoring module, a signal simulation module, an instruction input module, an abnormality testing module, an alarm module and a display module;
the data acquisition module comprises MATLAB software and is used for generating a data set of the radar signal in a simulation mode; transmitting a radar signal data set generated by simulation to a data preprocessing module, wherein the data preprocessing module is used for preprocessing a data set signal, analyzing the time domain characteristic and the frequency domain characteristic of the data set signal by using a software radio technology, and establishing a corresponding mathematical model; the specific treatment steps are as follows:
s1: firstly, outputting data set signals in a sequence form;
s2: carrying out time-frequency transformation on the data set signals, analyzing the time domain characteristics and the frequency domain characteristics of the data set signals by a software radio technology, and outputting the data set signals in a signal time-frequency diagram form;
s3: labeling the class of the signal in the data set signals output in the two forms, randomly extracting 2400 samples from each class of signals as training samples, and taking 600 samples as test samples;
s4: training a RNN-DenseNet-based network to obtain a corresponding mathematical model M;
the data preprocessing module is used for transmitting the corresponding mathematical model M to the signal identification module;
the signal monitoring module is an AD acquisition circuit and is used for acquiring radar signals in real time, obtaining discrete sampling data of the radar signals and transmitting the discrete sampling data of the radar signals to the signal identification module, the signal identification module receives the corresponding mathematical model M and the discrete sampling data of the radar signals and carries out identification analysis on the discrete sampling data of the radar signals, and the identification is based on an intelligent identification algorithm of artificial intelligence, and the specific identification analysis process is as follows:
the method comprises the following steps: acquiring a corresponding mathematical model M, and restoring discrete sampling data of radar signals to continuous signal characteristics according to the mathematical model M, wherein the signal characteristics comprise time domain characteristics and frequency domain characteristics;
step two: automatically comparing and analyzing the signal characteristics obtained by reduction with a big data reference module, wherein the big data reference module stores various radar signal models and signal characteristics, and identifying and obtaining corresponding signal types and attributes based on an artificial intelligence intelligent identification algorithm;
step three: according to the signal type and attribute obtained by identification, performing control, demodulation and decoding on the radar signal to obtain a cell, and finally decoding information content; the signal identification module is used for transmitting the decoded cells and the decoded information content to the storage module for storage through the controller.
2. An artificial intelligence signal analysis and recognition system as claimed in claim 1, wherein the recognition logic of said intelligent recognition algorithm is as follows:
aiming at AD real data, establishing a frequency domain model according to frequency band characteristics and a background environment, wherein input parameters comprise frequency domain resolution fd and frequency domain smoothing parameters fs;
aiming at AD real data, a time domain model is established according to frequency band characteristics and a background environment, and input parameters comprise a time domain resolution td and a time domain smoothing parameter ts;
combining the output results of the time domain model and the frequency domain model according to the weight to form a comprehensive mathematical model;
in the automatic learning stage, the characteristic attribute of the comprehensive mathematical model and the physical attribute of the actual signal are paired
Comparing and analyzing to find out the difference;
automatically adjusting the parameters of fd, fs, td and ts in the comprehensive mathematical model according to the difference to ensure that the contrast difference is smaller and smaller;
analyzing and identifying frequency domain data and time domain data according to the attribute of the comprehensive mathematical model, and identifying the characteristics of large signal components, including a central frequency point and a signal bandwidth;
acquiring data aiming at the identified signal frequency point and bandwidth, and then demodulating, decoding and decoding;
in the machine learning stage, the results of demodulation, decoding and decoding will be compared with the correct result of the calculation example, and the related calculation parameters are adjusted according to the difference.
3. An artificial intelligence signal analysis and recognition system as claimed in claim 1, wherein said data set includes seven signals selected from the group consisting of regular pulse signals, chirp signals, non-chirp signals, biphase code signals, quadriphase code signals, biphase code signals and quadriphase code signals, wherein each signal generates 3000 samples every 2dB signal-to-noise ratio from-10 dB to 2 dB.
4. The artificial intelligence signal analysis and recognition system of claim 1, further comprising an information recording module, an information sorting module and a parameter compensation module, wherein the information recording module is configured to record the recognition records of the signal recognition module and transmit the recognition records to the information sorting module;
the information sorting module receives and sorts the identification records, constructs a parameter detection training sample, trains based on a machine learning algorithm to obtain a parameter compensation model and transmits the parameter compensation model to the parameter compensation module, and the parameter compensation module receives the parameter compensation model and then corrects the mathematical model M and the intelligent algorithm.
5. An artificial intelligence signal analysis and identification system as claimed in claim 1, wherein said signal simulation module is adapted to simulate radar signals and to set a target value for said radar signals, said target value representing a model and signal characteristics of the radar signals;
before the signal identification module operates, a user inputs a test instruction through the instruction input module, the controller is used for transmitting the test instruction to the abnormity test module, and the abnormity test module is used for detecting the radar signal abnormity detection and abnormity processing capacity of the signal identification module after receiving the test instruction.
6. The system of claim 5, wherein the anomaly testing module comprises a transmitting antenna, and the detecting steps are as follows:
v1: after the anomaly testing module receives the testing instruction, the signal simulation module transmits a simulated radar signal, and meanwhile, the transmitting antenna is used for transmitting an interference signal to carry out electromagnetic wave interference on the radar signal;
v2: the signal identification module receives discrete sampling data corresponding to radar signals, performs identification analysis, and outputs measured values corresponding to the radar signals, wherein the measured values comprise measured time domain characteristics and measured frequency domain characteristics of the corresponding radar signals;
v3: judging whether the measured value of the radar signal is consistent with the target value; if not, generating an interference signal; the abnormal testing module is used for transmitting the interference signal to the controller, and the controller is used for automatically driving the alarm module to give an alarm when receiving the interference signal and automatically driving the display module to display that the signal recognition module has serious signal interference and suggests processing.
7. The system of claim 6, wherein the step of determining whether the measured value of the radar signal is consistent with the target value comprises:
acquiring an actually measured value and a target value of a radar signal, comparing the actually measured time domain characteristic of the radar signal with the target time domain characteristic to obtain a time domain error, and marking the time domain error as W1; comparing the measured frequency domain characteristic of the radar signal with the target frequency domain characteristic to obtain a frequency domain error, and marking the frequency domain error as W2;
using a formula
Calculating to obtain an error coefficient WX; wherein d1 and d2 are coefficient factors; if the error coefficient WX is less than or equal to the error coefficient threshold value, judging that the measured value is consistent with the target value; generating a normal signal; if the error coefficient WX is larger than the error coefficient threshold value, the measured value is judged to be inconsistent with the target value, and an interference signal is generated.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210317876.3A CN114791593A (en) | 2022-03-29 | 2022-03-29 | Artificial intelligence signal analysis and recognition system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210317876.3A CN114791593A (en) | 2022-03-29 | 2022-03-29 | Artificial intelligence signal analysis and recognition system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114791593A true CN114791593A (en) | 2022-07-26 |
Family
ID=82462131
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210317876.3A Pending CN114791593A (en) | 2022-03-29 | 2022-03-29 | Artificial intelligence signal analysis and recognition system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114791593A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115840875A (en) * | 2022-11-10 | 2023-03-24 | 北京擎天信安科技有限公司 | Millimeter wave radar abnormal signal detection method and system based on analog transducer |
| CN116963156A (en) * | 2023-07-12 | 2023-10-27 | 深圳市华检检测技术有限公司 | Wireless signal transmission capability detection method and device |
-
2022
- 2022-03-29 CN CN202210317876.3A patent/CN114791593A/en active Pending
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115840875A (en) * | 2022-11-10 | 2023-03-24 | 北京擎天信安科技有限公司 | Millimeter wave radar abnormal signal detection method and system based on analog transducer |
| CN116963156A (en) * | 2023-07-12 | 2023-10-27 | 深圳市华检检测技术有限公司 | Wireless signal transmission capability detection method and device |
| CN116963156B (en) * | 2023-07-12 | 2024-04-16 | 深圳市华检检测技术有限公司 | Wireless signal transmission capability detection method and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114791593A (en) | Artificial intelligence signal analysis and recognition system | |
| CN108055094B (en) | Unmanned aerial vehicle manipulator frequency spectrum feature identification and positioning method | |
| CN108280395B (en) | Efficient identification method for flight control signals of low-small-slow unmanned aerial vehicle | |
| CN110133599B (en) | Intelligent radar radiation source signal classification method based on long-time and short-time memory model | |
| CN107277443B (en) | Large-range peripheral safety monitoring method and system | |
| CN110969194B (en) | Cable early fault positioning method based on improved convolutional neural network | |
| CN103679753A (en) | Track identifying method of probability hypothesis density filter and track identifying system | |
| CN112734617A (en) | Geological disaster early warning monitoring method based on 5G positioning technology | |
| CN116684878B (en) | 5G information transmission data safety monitoring system | |
| CN114154545A (en) | Intelligent unmanned aerial vehicle measurement and control signal identification method under strong mutual interference condition | |
| CN112257500A (en) | Intelligent image recognition system and method for power equipment based on cloud edge cooperation technology | |
| CN115546608A (en) | Unmanned aerial vehicle data link electromagnetic interference classification and threat assessment method | |
| CN102346948B (en) | Circumference invasion detection method and system | |
| CN102670182A (en) | Pulse wave acquisition quality real-time analysis device | |
| CN116560946A (en) | Soil pollution data pushing system based on cloud computing | |
| CN116032602A (en) | Method, device, equipment and storage medium for automatically identifying threat data | |
| CN108599880A (en) | The air-ground intercom system of civil aviaton based on convolutional neural networks interferes method for early warning | |
| CN115327942A (en) | Intelligent environment monitoring system | |
| CN114580468A (en) | Interference signal identification method based on time-frequency waterfall graph and convolutional neural network | |
| CN115865610A (en) | Network alarm information processing method, device and equipment | |
| CN117119508B (en) | System and method for analyzing error code number of wireless communication channel | |
| CN114441463B (en) | Full spectrum water quality data analysis method | |
| CN117278073B (en) | Automatic adjustment method for ultra-wideband antenna signals | |
| CN115859056B (en) | Unmanned aerial vehicle target detection method based on neural network | |
| US11551061B2 (en) | System for generating synthetic digital data of multiple sources |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |