CN107682109B - A kind of interference signal classifying identification method suitable for UAV Communication system - Google Patents
A kind of interference signal classifying identification method suitable for UAV Communication system Download PDFInfo
- Publication number
- CN107682109B CN107682109B CN201710943615.1A CN201710943615A CN107682109B CN 107682109 B CN107682109 B CN 107682109B CN 201710943615 A CN201710943615 A CN 201710943615A CN 107682109 B CN107682109 B CN 107682109B
- Authority
- CN
- China
- Prior art keywords
- sample data
- signal
- interference signal
- factor
- classification
- 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.)
- Active
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04K—SECRET COMMUNICATION; JAMMING OF COMMUNICATION
- H04K3/00—Jamming of communication; Counter-measures
- H04K3/20—Countermeasures against jamming
- H04K3/22—Countermeasures against jamming including jamming detection and monitoring
Landscapes
- Engineering & Computer Science (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Noise Elimination (AREA)
Abstract
The invention discloses a kind of interference signal classifying identification methods suitable for UAV Communication system, belong to digital communication signal processing technology field.Interference signal classifying and identifying system is established first against UAV Communication system, then two kinds of characteristic parameters of the improved carrier wave factor and the white noise factor are extracted respectively, for each sample data, two kinds of characteristic parameters are compared respectively, output of the high characteristic parameter of selective discrimination degree as the sample data, classified using interference signal of the support vector machines to the sample data, and calculate the classification accuracy of interference signal, judge whether it is greater than 0.8, if it is, the interference signal classification of the sample data is correct；Classification belonging to the interference signal of all sample datas in the signal kinds statistical module counts period is ultimately interfered with, is laid the foundation for next step signal processing.The present invention is better than traditional threshold classification recognition methods, establishes the characteristic parameter expression of high discrimination, improves the accuracy rate of Classification and Identification.
Description
Technical field
The invention belongs to digital communication signal processing technology fields, and in particular to a kind of suitable for UAV Communication system
Interference signal classifying identification method.
Background technique
In recent years, using unmanned plane execute task flexibly it is convenient and at low cost so that execute various communication tasks at
For the hot spot in unmanned plane research field.Since UAV Communication system is in military and civilian field, it is dry that the moment suffers from nature
It disturbs and human interference, these interference can further influence the other applications such as Unmanned Aerial Vehicle Data Link power control techniques.Therefore, in nothing
Man-machine communication field, carrying out Classification and Identification to acquired interference signal necessarily has high research significance.
For the classifying identification method of interference signal, there are mainly two types of both at home and abroad at present:
First is that the recognition methods based on threshold classification.Traditional threshold classification recognition methods is easy to operate, main according to not
With the difference between signal characteristic parameter, establish the similarity function of robustness, while choosing suitable threshold value, by similarity with
The threshold value of selection is compared, and makes corresponding judgement.And the selection of threshold value depends on specific problem, in different communication
It might have different threshold values in system or different interference signals, therefore such method is generally for the adaptability of Classification and Identification
It is poor.
Second is that the classifying identification method based on machine learning.Existing machine learning recognition methods includes backpropagation (BP)
The statistical learning methods such as neural network, support vector machine, decision tree theory, such method have more excellent self study and adaptive
Ability can preferably solve the problems, such as non-linear and higher-dimension, and structure is simple and speed is fast.However the Classification and Identification of such method
Performance depends greatly on the characterization ability of input characteristic parameter, therefore research should concentrate on how establishing effective spy
Levy parameter.
As the above analysis, currently it is directed to the main problems faced of interference signal Classification and Identification of UAV Communication system
As follows: complicated and changeable due to interfering when UAV Communication system executes task, characteristic parameter may differentiation degree to signal
Deficiency leads to not smoothly classify；Meanwhile traditional threshold classification method applicability is poor, and it is accurate to often lead to lower identification
Rate.
Summary of the invention
The present invention is to solve the above problems, propose a kind of interference signal Classification and Identification suitable for UAV Communication system
Method studies the classification and identification of interference signal using interference signal as research object.
The interference signal classifying identification method suitable for UAV Communication system is treated journey such as to signal
Under:
Step 1: being directed to UAV Communication system, establishing includes signal generation module, characteristic extracting module, svm classifier mould
The interference signal classifying and identifying system of block and interference signal type statistical module；
Signal generation module generates direct sequence signal, and single tone jamming is added, Multi-tone jamming, partial-band jamming, and frequency sweep is dry
The interference signal for disturbing these four types of data, using this result as sample data；
Signal in the sample data that characteristic extracting module generates signal generation module carries out feature extraction；
Svm classifier module is input to support vector machines to the feature that characteristic extracting module is extracted and classifies.
Interference signal type statistical module to sample data respectively belonging to classification count.
Step 2: extracting change respectively using characteristic extracting module from each sample data that signal generation module generates
Into two kinds of characteristic parameters of the carrier wave factor and the white noise factor；
Extract improved carrier wave factor signal Cg, calculation formula is as follows:
x(λ1) it is the discrete signal that the amplitude maximum obtained after Fast Fourier Transform (FFT) is carried out to time-domain signal s (t)；
Fast Fourier Transform (FFT) is carried out to time-domain signal s (t) and obtains discrete signal x (n), n=1,2 ..., N, according to width
Value size order is successively ordered as x (λ1),x(λ2),...,x(λn)；It defines l=0.95 × n+N (n > 100), n is without spread spectrum
Former sequence length, N be sequence spread spectrum points.
Improved white noise factor signal A is extracted, calculation formula is as follows:
P1For the common mean value of the power spectrum P (n) of time-domain signal s (t)；P2For greater than P1The mean value that calculates again of amplitude.
Step 3: being directed to each sample data, two kinds of characteristic parameters, the high characteristic parameter of selective discrimination degree are compared respectively
Output as the sample data；
Specific comparison process is as follows:
It is poor that the carrier wave factor improved in two sample datas A and B is made, when the carrier wave factor of sample data A is greater than sample
The carrier wave factor of data B, and difference be greater than 20 when, select the value of the carrier wave factor as the output of sample data A；
It is poor that the white noise factor improved in two sample datas A and B is made, when the white noise factor of sample data A is greater than
The white noise factor of sample data B, and difference be greater than 0.6 when, select the value of the white noise factor as the output of sample data A；
When the value of the value of the carrier wave factor of sample data A and the white noise factor could act as the output of the sample data,
It calculates separately the carrier wave factor and the opposite of the white noise factor distinguishes degree g；And select the opposite high value of degree of distinguishing as sample
The output of data A；
Opposite differentiation degree g calculation formula is as follows:
Wherein, when calculating the opposite differentiation degree g of the carrier wave factor using above-mentioned formula, b is that the feature of the carrier wave factor is joined
Number difference, a are the threshold value of the carrier wave factor.When calculating the opposite differentiation degree g of the white noise factor using above-mentioned formula, b is white
The characteristic parameter difference of noise factor, a are the threshold value of the white noise factor.
Step 4: being directed to some sample data, svm classifier module utilizes supporting vector according to the output of the sample data
Machine SVM classifies to the interference signal of the sample data, and calculates the classification accuracy of interference signal.
Firstly, choosing kernel function and parameter training support vector machines；
Then, the sample data duplication of selection is N number of, and selected wherein certain interference type is respectively provided with label 1, remaining
Interference type is respectively provided with label -1, is put into trained support vector machines as training set in conjunction with the output of sample data
Classify, obtains classification results；
Finally, comparison-of-pair sorting's result and manual tag, obtain classification accuracy η；
Calculation formula is as follows:
Step 5: judging whether classification accuracy η is greater than 0.8, if it is, the interference signal classification of the sample data is just
Really, otherwise, the interference signal classification error of the sample data, return step two.
Step 6: it is directed to certain time, all sample datas in interference signal type statistical module counts period
Classification belonging to interference signal.
The invention has the following advantages that
1) a kind of interference signal classifying identification method suitable for UAV Communication system, improves the ginseng of original feature extraction
Number establishes the characteristic parameter expression of high discrimination, improves the accuracy rate of Classification and Identification；
2) a kind of interference signal classifying identification method suitable for UAV Communication system is tested using support vector machines intersection
The mode of card is better than traditional threshold classification recognition methods in Classification and Identification accuracy rate；
3) a kind of interference signal classifying identification method suitable for UAV Communication system establishes interference letter in a period of time
The statistical model of number type lays the foundation for next step signal processing.
Detailed description of the invention
Fig. 1 is the interference signal classifying identification method schematic diagram that the present invention is suitable for UAV Communication system；
Fig. 2 is the interference signal classifying identification method flow chart that the present invention is suitable for UAV Communication system；
Fig. 3 is the interference signal classifying and identifying system block diagram that the present invention establishes；
Fig. 4 is that the present invention extracts characteristic parameter stream suitable for the interference signal classifying identification method of UAV Communication system
Cheng Tu；
Fig. 5 a is the comparison diagram that the improved carrier wave factor coefficient of the present invention and primary carrier factor coefficient index signaling zone；
Fig. 5 b is the classification results for carrying out interference signal Classification and Identification in the embodiment of the present invention using support vector machines；
Fig. 6 is GUI interactive interface exemplary diagram of the invention.
Specific embodiment
With reference to the accompanying drawing, specific implementation method of the invention is described in detail.
A kind of interference signal classifying identification method suitable for UAV Communication system of the present invention, as shown in Figure 1, logical first
Cross and feature extracted to input signal, that is, two characterization factors are extracted to the sample signal of input: the improved carrier wave factor with
The white noise factor analyzes the inherent characteristic of signal, and the spy high according to characteristic parameter selection standard selective discrimination degree
Parameter is levied, the input as svm classifier identification interference.Wherein input signal, which represents, generates the sample data stage, and wherein signal is raw
Direct sequence signal is generated at module, and single tone jamming is added, Multi-tone jamming, partial-band jamming, these four types of Sweeping nonlinearity
Signal interference.
Secondly, feature extraction is carried out to interference signal by established characteristic parameter, using the side of support vector machines
Method carries out Classification and Identification to interference signal, exports the recognition accuracy to disturbance signal；Svm classifier identification interference is using
Trained good classifier, classifies to the sample signal of input, and recognition accuracy is input to statistics interference type, passes through
Differentiate recognition accuracy, counts the quantity of various types interference signal.
It is compared by the way that thresholding is arranged to recognition accuracy, determines interference signal type, and then interference in a period of time
The statistics of signal kinds establishes the statistical model of interference signal number of species；Finally, on the basis of Feature extraction and recognition,
Interference signal classifying and identifying system is established, disturbance ecology user interaction of the final building one based on matrix labotstory (MATLAB)
System.
As shown in Figure 2, the specific steps are as follows:
Step 1: being directed to UAV Communication system, establishing includes signal generation module, characteristic extracting module, svm classifier mould
The interference signal classifying and identifying system of block and interference signal type statistical module；
The major function of the system has: the extraction of characteristic parameter is carried out to signal, by support vector machines to interference signal
Classified and obtain recognition accuracy, the interference signal type statistics in a period of time is established on the basis of recognition accuracy
Model.
As shown in figure 3, signal generation module generates direct sequence signal first, secondly user selects to generate interference type addition directly
Expand signal, generate signal of communication, by Gaussian white noise channel, Rayleigh channel or Rice channel obtain output input feature vector
Extraction module；
Interference type includes single tone jamming, Multi-tone jamming, partial-band jamming, these four types of Sweeping nonlinearity data it is dry
Disturb signal；
Signal in the sample data that characteristic extracting module generates signal generation module carries out character selection and abstraction, obtains
To characteristic parameter；Specifically: carry out feature extraction using improved characterization factor, by the carrier wave factor extracted and white noise because
Subcharacter is input to svm classifier module.
Svm classifier module is input to support vector machines to the feature that characteristic extracting module is extracted and classifies.It is raw first
At training set to classifier training, then the classification of interference signal is carried out, exports disturbance ecology accuracy rate；
Disturbance ecology accuracy rate is carried out discriminating whether to be greater than threshold value 0.8 by interference signal type statistical module, greater than threshold value
When, it counts in a period of time obtain the number of various interference types respectively, its statistical result is exported；Otherwise feature is returned to
Extraction module carries out feature extraction.
The above module composition one is suitable for the interference signal classifying and identifying system of UAV Communication system, can classify
It identifies interference type and counts interference signal type in a period of time, lay the foundation for next step signal processing.
Step 2: extracting change respectively using characteristic extracting module from each sample data that signal generation module generates
Into two kinds of characteristic parameters of the carrier wave factor and the white noise factor；
Sample data has many redundancies and substantial amounts, dimension are high, if being directly used in classification and identification algorithm, not only
Operation is complicated, time-consuming big, and is easy to cause not restraining for classification and identification algorithm, influences system identification performance.To solve above deposit
The problem of, need to be using the characteristic parameter of low sample low dimensional as the input of classification and identification algorithm.
The characteristic parameter for extracting low sample low dimensional includes extracting the improved carrier wave factor and two kinds of features of the white noise factor
Parameter；
1. into the carrier wave factor
Original carrier wave factor coefficient C carries out Fast Fourier Transform (FFT) to time-domain signal s (t) first and obtains discrete signal x
(n), then n=1,2 ..., N successively sort x (λ according to amplitude size order1),x(λ2),...,x(λn), finally calculate letter
Number spectrum maximum value and consecutive value ratio；
Original carrier wave factor coefficient describes the projecting degree of signal spectrum, is expressed as follows:
λ1,λ2,...,λnAfter indicating the sequence of amplitude size order, the discrete frequency point sequence of discrete signal x (n).
Sample of the present invention data are because sequence is longer than 1000 frequency points, and after spread spectrum, the original statistics of sequence is advised
Rule variation is unobvious, therefore improves to this factor.
Define improved carrier wave factor sequence CgAre as follows:
It wherein defines l=0.95 × n+N (n > 100), n is the former sequence length without spread spectrum, and N is sequence spread spectrum points.
2. improving the white noise factor
For additive white gaussian interference signal, the envelope of frequency spectrum is flat；And for other normal signal, envelope
Distribution often have different degrees of fluctuations.Therefore can be found out by following several steps white Gaussian interference signal because
Son, and then to distinguish different types of signal:
Improved white noise factor signal A is extracted, calculation formula is as follows:
P1For the common mean value of the power spectrum P (n) of time-domain signal s (t)；P2For greater than P1The mean value that calculates again of amplitude.
Step 3: being directed to each sample data, two kinds of characteristic parameters, the high characteristic parameter of selective discrimination degree are compared respectively
Output as the sample data；
After establishing both the above characteristic parameter, for the signal of communication of input, first to improving the carrier wave factor
Characteristic parameter extraction is carried out with the signal of the white noise factor；Secondly two kinds of characteristic parameters of signal extraction are compared, is chosen
Suitable characterization factor distinguishes both signals, and control methods is as follows:
Because characterization factor is the signal in order to distinguish two kinds and its above quantity, therefore illustrates feature by taking two kinds of signals as an example
Parameter is between the differentiation degree signal.
As shown in figure 4, specific comparison process is as follows:
It is poor that the carrier wave factor improved in two sample datas A and B is made, when the carrier wave factor g1 of sample data A is greater than sample
The carrier wave factor g2 of notebook data B, and difference be greater than 20 when, select output of the value of carrier wave factor g1 as sample data A；
It is poor that the white noise factor improved in two sample datas A and B is made, when the white noise factor g1 of sample data A is big
In the white noise factor g2 of sample data B, and when difference is greater than 0.6, select the value of white noise factor g2 as sample data A's
Output；
When the value of the carrier wave factor g1 of sample data A and the value of white noise factor g2 could act as the output of the sample data
When, it calculates separately the carrier wave factor and the opposite of the white noise factor distinguishes degree g；And select the opposite high value of degree of distinguishing as sample
The output of notebook data A；
Opposite differentiation degree g calculation formula is as follows:
Wherein, when calculating the opposite differentiation degree g of the carrier wave factor using above-mentioned formula, b is that the feature of the carrier wave factor is joined
Number difference, a are the threshold value of the carrier wave factor；The present invention chooses improved carrier wave factor a=20.When utilization above-mentioned formula calculates white noise
When the opposite differentiation degree g of the sound factor, b is the characteristic parameter difference of the white noise factor, and a is the threshold value of the white noise factor, this hair
It is bright to choose improved white noise factor a=0.6.
Step 4: being directed to some sample data, svm classifier module utilizes supporting vector according to the output of the sample data
Machine SVM classifies to the interference signal of the sample data, and calculates the classification accuracy of interference signal.
Classified by the high interference signal of discrimination of the support vector machines to extraction, and exported to interference signal
Recognition accuracy.
It is specifically divided into two parts: 1) support vector machines is carried out with the selection of kernel function and parameter；
The categorised decision rule of SVM is as follows:
Wherein, xi,yiFor training sample；φ(xj), φ (xi) it is mapping from the input space to some feature space；αiFor
Classification factor is set as 1.<φ(xi),φ(xj) > indicate to φ (xi),φ(xj) carry out inner product.In the present invention, training sample
This is that the training set formed after two kinds of characterization factors is extracted in four kinds of interference.
According to above-mentioned decision rule, kernel function k (x, y)=< φ (x is definedi),φ(yi)>；
By kernel function, low-dimensional nonlinear characteristic is mapped as High-dimensional Linear feature, so that the selection of hyperplane becomes to hold
Easily, radial kernel function of the invention:
Wherein, x, y are training sample；σ is the width parameter of function, controls the radial effect range of function.
2) Classification and Identification is carried out using support vector machines.
Specific assorting process is as follows:
By the sample data of selection, duplication obtains 100 identical sample datas, generates 100 single tone jamming data,
100 Multi-tone jamming data, 100 partial-band jamming data, 100 Sweeping nonlinearity data, for these data, in extraction
State the selected characteristic parameter of sample data.Certain interference type label that 100 selected needs are classified be 1, remaining 300
Data tagged -1 obtain classification results as being trained in training set input support vector machines；To test set into
The classification of row such as formula (4)；Finally, comparison-of-pair sorting's result and manual tag, obtain classification accuracy, defining classification accuracy rate η
Are as follows:
Step 5: judging whether classification accuracy η is greater than thresholding 0.8, if it is, the interference signal of the sample data point
Class is correct, otherwise, the interference signal classification error of the sample data, return step two.
By to recognition accuracy be arranged thresholding be compared, determine interference signal type, if by support vector machines into
The recognition accuracy of row classification is greater than 0.8, then it is assumed that classification is correct.The correct interference type number of classification in statistics a period of time
Amount, to the statistical analysis of noise type in a period of time, establishes the statistical model of noise type quantity, helps to obtain one section
The changing rule of noise type quantity in time finds out main noise present in signal propagation, lays for follow-up signal processing
Solid foundation.
It can determine interference noise type when recognition accuracy is greater than 80%, count the noise type of a period of time
It was found that rule existing for main noise type, result is then inputted into Unmanned Aerial Vehicle Data Link power control module, using identifying
The noise type come carries out the selection and power control of communication channel.
Step 6: it is directed to certain time, all sample datas in interference signal type statistical module counts period
Classification belonging to interference signal.
The present invention can carry out the extraction of characteristic parameter by the interference signal classifying and identifying system established to signal, and
Classification and Identification is carried out to interference signal by support vector machines, finally establishes the interference signal type statistics mould in a period of time
Type.
By the setting of the above communication system, the available accuracy rate in such communication environment disturbance ecology, thus right
Signal of communication in true environment is estimated.
Mainly illustrate effectiveness of the invention below with two groups of sequences:
Improved carrier wave factor coefficient CgThe comparison diagram indexed with primary carrier factor coefficient C for signaling zone, such as Fig. 5 a institute
Show, it can be seen that the improved carrier wave factor is higher for signal distinguishing degree；Interference signal classification is carried out using support vector machines
Identification as a result, as shown in Figure 5 b, it is known that with the increase of signal-to-noise ratio, the accuracy rate of classification is higher and higher.
At the same time, the present invention establishes a Matlab GUI user interface, as shown in fig. 6, can pass through GUI
Interface selects to need the interference type that generates, by channel type, the characteristic parameter of extraction and the SVM training set of generation.
Gui interface is made of five modules, respectively interference type, channel type, training set generate, interference signal type statistics with it is defeated
Classification accuracy module out.First by the dry value input interface made an uproar than (JRN) that oneself sets, select classified interference type and
Its corresponding characteristic parameter；Secondly, its corresponding training set of selection channel type generates type；Finally, click recognition is accurate
Rate button, exportable classification accuracy.Interference signal type statistical module is for counting each of system identification in a period of time
Kind of interference signal number of species, will be on the right side of its statistical result showed in text box.
Interference signal classifying identification method in the present invention is more able to satisfy unmanned plane for the demand of communication quality, is classifying
While identification interference, selects multiple channels to enrich simulated environment, use the increase of the technique study signal-to-noise ratio of support vector machines
With the relationship of disturbance ecology accuracy rate, and it is made that the graphical interface of user of classification of disturbance system.Experiment shows the present invention couple
There is certain validity in the classification of raising with the interference of UAV Communication quality.
Claims (1)
1. a kind of interference signal classifying identification method suitable for UAV Communication system, which is characterized in that specific step is as follows:
Step 1: be directed to UAV Communication system, establish include signal generation module, characteristic extracting module, svm classifier module and
The interference signal classifying and identifying system of interference signal type statistical module；
Signal generation module generates direct sequence signal, and single tone jamming is added, Multi-tone jamming, partial-band jamming, Sweeping nonlinearity number
According to the interference signal of these four types, using this result as sample data；
Signal in the sample data that characteristic extracting module generates signal generation module carries out feature extraction；
Svm classifier module is input to support vector machines to the feature that characteristic extracting module is extracted and classifies；
Interference signal type statistical module to sample data respectively belonging to classification count；
Step 2: being extracted respectively improved from each sample data that signal generation module generates using characteristic extracting module
Two kinds of characteristic parameters of the carrier wave factor and the white noise factor；
Extract improved carrier wave factor signal Cg, calculation formula is as follows:
x(λ1) it is the discrete signal that the amplitude maximum obtained after Fast Fourier Transform (FFT) is carried out to time-domain signal s (t)；
Fast Fourier Transform (FFT) is carried out to time-domain signal s (t) and obtains discrete signal x (n), n=1,2 ..., N, it is big according to amplitude
Small sequence is successively ordered as x (λ1),x(λ2),...,x(λn)；It defines l=0.95 × n+N (n > 100), n is the original without spread spectrum
Sequence length, N are sequence spread spectrum points；
Improved white noise factor signal A' is extracted, calculation formula is as follows:
P1For the common mean value of the power spectrum P (n) of time-domain signal s (t)；P2For greater than P1The mean value that calculates again of amplitude；
Step 3: being directed to each sample data, two kinds of characteristic parameters, the high characteristic parameter conduct of selective discrimination degree are compared respectively
The output of the sample data；
Specific comparison process is as follows:
It is poor that the carrier wave factor improved in two sample datas A and B is made, when the carrier wave factor of sample data A is greater than sample data B
The carrier wave factor, and difference be greater than 20 when, select the value of the carrier wave factor as the output of sample data A；
It is poor that the white noise factor improved in two sample datas A and B is made, when the white noise factor of sample data A is greater than sample
The white noise factor of data B, and difference be greater than 0.6 when, select the value of the white noise factor as the output of sample data A；
When the value of the value of the carrier wave factor of sample data A and the white noise factor could act as the output of the sample data, respectively
It calculates the carrier wave factor and the opposite of the white noise factor distinguishes degree g；And select the opposite high value of degree of distinguishing as sample data A
Output；
Opposite differentiation degree g calculation formula is as follows:
Wherein, when calculating the opposite differentiation degree g of the carrier wave factor using above-mentioned formula, b is that the characteristic parameter of the carrier wave factor is poor
Value, a are the threshold value of the carrier wave factor；When calculating the opposite differentiation degree g of the white noise factor using above-mentioned formula, b is white noise
The characteristic parameter difference of the factor, a are the threshold value of the white noise factor；
Step 4: being directed to some sample data, svm classifier module utilizes support vector machines according to the output of the sample data
Classify to the interference signal of the sample data, and calculates the classification accuracy of interference signal；
Firstly, choosing kernel function and parameter training support vector machines；
Then, the sample data duplication of selection is N number of, and selected wherein certain interference type is respectively provided with label 1, remaining interference
Type is respectively provided with label -1, is put into trained support vector machines as training set in conjunction with the output of sample data and carries out
Classification, obtains classification results；
Finally, comparison-of-pair sorting's result and manual tag, obtain classification accuracy η；
Calculation formula is as follows:
Step 5: judge whether classification accuracy η is greater than 0.8, if it is, the interference signal classification of the sample data is correct,
Otherwise, the interference signal classification error of the sample data, return step two；
Step 6: it is directed to certain time, the interference of all sample datas in interference signal type statistical module counts period
Classification belonging to signal.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710943615.1A CN107682109B (en) | 2017-10-11 | 2017-10-11 | A kind of interference signal classifying identification method suitable for UAV Communication system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710943615.1A CN107682109B (en) | 2017-10-11 | 2017-10-11 | A kind of interference signal classifying identification method suitable for UAV Communication system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107682109A CN107682109A (en) | 2018-02-09 |
| CN107682109B true CN107682109B (en) | 2019-07-30 |
Family
ID=61140016
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710943615.1A Active CN107682109B (en) | 2017-10-11 | 2017-10-11 | A kind of interference signal classifying identification method suitable for UAV Communication system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107682109B (en) |
Families Citing this family (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109034088A (en) * | 2018-08-06 | 2018-12-18 | 北京邮电大学 | A kind of unmanned plane signal detection method and device |
| CN109359523B (en) * | 2018-09-06 | 2021-08-24 | 东南大学 | Satellite navigation interference type identification method based on SVM multi-classification algorithm |
| CN110336637A (en) * | 2019-07-15 | 2019-10-15 | 北京航空航天大学 | A kind of unmanned plane interference signal feature selection approach |
| CN113057655A (en) * | 2020-12-29 | 2021-07-02 | 深圳迈瑞生物医疗电子股份有限公司 | Recognition method, recognition system and detection system for electroencephalogram signal interference |
| CN113422668B (en) * | 2021-06-15 | 2022-05-27 | 河南科技大学 | Cooperative communication transmission method based on crisp-resistant relay |
| CN115809422B (en) * | 2021-09-13 | 2023-11-21 | 国家电网有限公司 | Unmanned aerial vehicle RF signal identification method and system based on SVM |
| CN113904659A (en) * | 2021-10-11 | 2022-01-07 | 山东万里红信息技术有限公司 | Multi-target self-adaptive anti-interference method based on space vector |
| CN117675085A (en) * | 2023-11-27 | 2024-03-08 | 国网电力空间技术有限公司 | Unmanned aerial vehicle autonomous flight monitoring method and system for power grid inspection |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101853531B (en) * | 2010-05-25 | 2012-09-05 | 北京航空航天大学 | Helicopter flight state identification method based on presort technology and RBF (Radial Basis Function) neural network |
| CN102760444B (en) * | 2012-04-25 | 2014-06-11 | 清华大学 | Support vector machine based classification method of base-band time-domain voice-frequency signal |
| CN102820034B (en) * | 2012-07-16 | 2014-05-21 | 中国民航大学 | Noise sensing and identifying device and method for civil aircraft |
| CN103885043B (en) * | 2014-03-31 | 2016-08-24 | 西安电子科技大学 | The miscellaneous Robust classification method of making an uproar of Aircraft Targets based on broad match filtering |
| CN106330385B (en) * | 2016-08-29 | 2018-11-09 | 电子科技大学 | A kind of interference type recognition methods |
| CN106908774B (en) * | 2017-01-06 | 2020-01-10 | 南京航空航天大学 | One-dimensional range profile identification method based on multi-scale nuclear sparse preserving projection |
-
2017
- 2017-10-11 CN CN201710943615.1A patent/CN107682109B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN107682109A (en) | 2018-02-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107682109B (en) | A kind of interference signal classifying identification method suitable for UAV Communication system | |
| CN106330385B (en) | A kind of interference type recognition methods | |
| CN106789788B (en) | A kind of wireless digital signal Modulation Mode Recognition method and device | |
| CN108171136A (en) | A kind of multitask bayonet vehicle is to scheme to search the system and method for figure | |
| CN105975518B (en) | Expectation cross entropy feature selecting Text Classification System and method based on comentropy | |
| CN103605990B (en) | Integrated multi-classifier fusion classification method and integrated multi-classifier fusion classification system based on graph clustering label propagation | |
| CN107830996B (en) | Fault diagnosis method for aircraft control surface system | |
| CN109934269A (en) | A kind of opener recognition methods of electromagnetic signal and device | |
| CN110120926A (en) | Modulation mode of communication signal recognition methods based on evolution BP neural network | |
| CN110515096A (en) | Satellite navigation interference signal identification device and its method based on convolutional neural networks | |
| CN105447502A (en) | Transient power disturbance identification method based on S conversion and improved SVM algorithm | |
| CN101763502A (en) | High-efficiency method and system for sensitive image detection | |
| CN109886284B (en) | Fraud detection method and system based on hierarchical clustering | |
| CN109948726B (en) | Power quality disturbance classification method based on deep forest | |
| CN112560596B (en) | Radar interference category identification method and system | |
| CN110134719A (en) | A kind of identification of structural data Sensitive Attributes and stage division of classifying | |
| CN110377605A (en) | A kind of Sensitive Attributes identification of structural data and classification stage division | |
| CN110046593A (en) | The complex electric energy quality disturbance recognition methods of S-transformation and random forest is improved based on segmentation | |
| CN105046323A (en) | Regularization-based RBF network multi-label classification method | |
| CN112528774A (en) | Intelligent sorting system and method for unknown radar signals in complex electromagnetic environment | |
| CN115577357A (en) | Android malicious software detection method based on stacking integration technology | |
| CN115932898A (en) | Efficient interference identification method for satellite navigation system | |
| CN108809874B (en) | Radar and communication multi-signal classification method based on circulation support vector machine | |
| CN106372618A (en) | Road extraction method and system based on SVM and genetic algorithm | |
| CN113128584B (en) | Mode-level unsupervised sorting method of multifunctional radar pulse sequence |
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 | ||
| CB03 | Change of inventor or designer information |
Inventor after: Liu Chunhui Inventor after: Ding Wenrui Inventor after: Hu Yuan Inventor after: Liu Chunlei Inventor before: Ding Wenrui Inventor before: Hu Yuan Inventor before: Liu Chunlei |
|
| CB03 | Change of inventor or designer information | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| EE01 | Entry into force of recordation of patent licensing contract |
Application publication date: 20180209 Assignee: Beijing northern sky long hawk UAV Technology Co.,Ltd. Assignor: BEIHANG University Contract record no.: X2021990000039 Denomination of invention: A method of jamming signal classification and recognition for UAV Communication System Granted publication date: 20190730 License type: Exclusive License Record date: 20210119 |
|
| EE01 | Entry into force of recordation of patent licensing contract |