CN103812577A - Method for automatically identifying and learning abnormal radio signal type - Google Patents
Method for automatically identifying and learning abnormal radio signal type Download PDFInfo
- Publication number
- CN103812577A CN103812577A CN201210438094.1A CN201210438094A CN103812577A CN 103812577 A CN103812577 A CN 103812577A CN 201210438094 A CN201210438094 A CN 201210438094A CN 103812577 A CN103812577 A CN 103812577A
- Authority
- CN
- China
- Prior art keywords
- signal
- improper
- data
- radio signal
- frequency
- 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.)
- Granted
Links
Images
Landscapes
- Monitoring And Testing Of Transmission In General (AREA)
Abstract
The invention discloses a method for automatically identifying and learning abnormal radio signal types. The method is characterized by completing feature extraction of a radio signal in virtue of analysis and actual measurement of the frequency spectrum data of the radio data in combination with experiential knowledge of radio monitoring experts and a feature extraction method. In the characteristic space of the radio signal, clustering analysis is executed on actually-measured sweep-frequency interference, broadband interference, narrowband interference, and invalid intervention signals by using a clustering analysis method. By means of a result of the clustering analysis, the method may provide a radio monitoring device with a capability of automatically identifying abnormal radio signal types. With accumulation of the frequency spectrum data of the actually-measured radio data and increase in identification errors, the method provides a capability of automatically learning the results of the clustering analysis regularly. According to an identification result in combination with a communication device, the method provides an automatic alarming capability for a monitoring device. The method, used in a radio monitoring device, improves an information processing capability of the radio monitoring device, may achieve unmanned operation of the radio monitoring device, and decreases the workload of monitoring technicians.
Description
Technical field
The present invention relates to radio monitoring field, more specifically relate to the automatic identification of radio signal detection and improper radio signal.
Background technology
Radio monitoring is mainly for the radio signal in radio control region, and the technical information such as its technical parameter, operating characteristic and radiation position are analyzed, identify, monitor and obtained to this radio signal.The discovery of improper radio signal occupies critical role with being identified in radio monitoring, and particularly during occasion, the monitoring to improper radio signal is even more important.
Traditional radio monitoring work is that radio monitoring personnel pass through monitoring equipment, coordinates professional knowledge and monitoring personnel's practical experience manually to complete.And be mainly to judge by spectrogram and the duration of this signal.This RM mainly contains following deficiency: on the one hand, different staff has different Heuristicses, has certain subjectivity in the time doing decision-making, directly affects the correct identification of improper radio signal; In radio monitoring process, discovery and the identification to improper signal depends on technical staff simultaneously, and wherein most of work repeats, and is the waste to human resources, and under case of emergency, has increased greatly its work difficulty and workload; On the other hand, the monitoring flow process of a set of standard is followed in the identification of improper radio signal, due to the information processing capability deficiency of existing radio monitoring equipment, cause the identification of improper radio signal to depend critically upon technical staff, cannot realize radio monitoring equipment nobody hold machine.
Summary of the invention
The object of the invention is to overcome the deficiencies in the prior art, improve automation, the intelligent level of radio monitoring, a kind of automatic recognition system and method thereof of improper radio signal are provided, this recognition system realizes the automatic detection of improper radio signal, and the automatic identification of improper radio signal, automation, the intelligent level of enhancing radio monitoring.Meanwhile, this recognition methods possesses the ability of algorithm parameter being carried out to self study.According to recognition result, in conjunction with communication equipment, this system possesses the automatic alarm ability of monitoring equipment.The method for radio monitoring system, can improve the information processing capability of radio monitoring system, realize radio monitoring system nobody hold machine, effectively alleviate monitoring personnel's work difficulty and workload.
Improper radio signal of the present invention refers to the illegal or interference signal monitoring, and comprises that scan-type interference, broad-band interference, arrowband disturb, illegally intercut signal etc.Can refer to that system is by being tentatively judged as doubtful improper radio signal by suspect signal, but also need further definite signal.
Improper radio signal automatic recognition system of the present invention comprises radio signal monitoring equipment system, radio signal intelligent analysis system, network communicating system and processing controller; Monitoring equipment system receives aerial electromagnetic wave, carries out conversion process, produces the Monitoring Data of signal, comprising: frequency spectrum data, speech data, bearing data, intermediate frequency measurement data etc.; Intelligent analysis system, by Monitoring Data being carried out to a series of intellectual analysis processing, is identified improper signal automatically; Network communicating system is responsible between the each module of system, communicating by letter between system and outside miscellaneous equipment; Processing controller is responsible for coordinating the processing scheduling between modules.Wherein radio signal intelligent analysis system, comprise the improper signal detection module of radio, improper signal characteristic abstraction module, improper signal identification module, system self-learning module, and station database, electromagnetic environment database, improper Signals Data Base; Obtain band scan data by data-interface, adopt segmentation dynamic self-adapting thresholding algorithm to monitor out all signals and the corresponding frequency of this frequency range, signal frequency point is carried out to the detailed data of intermediate frequency measurement or medium-frequency direction finding picked up signal, analyzing and processing is extracted the various features of signal, extracted feature is inputted to improper signal identification module, identify the affiliated classification of this signal; The data that system self-learning module is used improper signal identification module to produce, regular update, improve feature extraction parameter and signal Recognition Algorithm parameter.
Radio signal monitoring equipment system of the present invention comprises receiver, spectrum measurement instrument, direction-finding equipment, audiomonitor, control appliance, antenna-feedback system, and the auxiliary system such as communication, power supply, lightning protection, environmental monitoring; Monitoring receiver receives aerial electromagnetic wave by antenna-feedback system, carry out conversion process, produce the frequency spectrum data of signal, speech data, bearing data, intermediate frequency measurement data etc., monitoring receiver obtains latitude and longitude coordinates data by GPS receiver, and monitoring receiver obtains the data message such as geographical environment information, climate temperature at place, monitoring station by environmental control system; Station database of the present invention is that electromagnetic environment database comprises all frequency ranges in this region Monitoring Data under normal circumstances in the data of all radio of declaring, ratifying frequency equipment of this region.
The step of the automatic identifying method of improper radio signal of the present invention is: first, by the frequency range of specifying is carried out to band scan, carry out input; The signal extraction series of features detecting is also selected to have to the feature of better distinguishing ability; Then use the classification of the automatic identification signal of improved FCM algorithm; According to the signal type automatic alarm identifying; To the improper radio signal identifying, system is preserved, after demarcating by manual confirmation, enter improper Signals Data Base, by new database self study feature selecting algorithm parameter and the new cluster centre of FCM, make system have the ability of unceasing study, intelligent level improves constantly.
The present invention adopts a kind of segmented adaptive Threshold to detect measured signal, and monitoring frequency range is carried out after the monitoring of a period of time, obtains the mean value of this actual measurement frequency band signals frequency spectrum data
, wherein
represent
the average energy of this period of time of sampled point,
.The instantaneous value of current this frame frequency segment data
, wherein
represent
the energy of current this frame of sampled point,
,
for the sampling number of frequency range data.The first step: according to the mean value of actual measurement frequency band signals frequency spectrum data
, carry out segment processing, the size of establishing every section is
individual sampled point, calculates the average of data in this section
, wherein
for hop count, by the data in this section successively and
compare, if
, be designated as signaling point, if
, be designated as non-signaling point, wherein
,
for the threshold value of setting, take the method to find out the subscript of this frequency band signals point; Second step: the noise level of this frequency band signals frequency is replaced by the average of non-signal frequency noise level before and after it, and for example the 2nd sampled point is signaling point, and the 1st and the 3rd is non-signaling point, and the noise level of second sampled point is
, adopt the noise level that extracts in this way whole frequency range, be made as
, wherein
represent
the noise level of individual sampled point,
; The 3rd step: use instantaneous value
with
compare one by one, order
represent
the back noise threshold value of individual sampled point,
.If
:
if,
:
, wherein
,
for set point,
, consider noise level when noise level while having signal will be higher than no signal, requirement
.Because also can producing noise and noise, various monitoring equipments self there is certain fluctuation, so this method adds a set point
, so just can better reflect actual conditions; The 4th step: to back noise threshold value
carry out the disposal of gentle filter.Find out the frequency of this frequency band signals according to back noise threshold value, then compare with the station database of setting up, analyze normal signal and can suspect signal, wherein can be divided into two kinds by suspect signal: one is in station database, to have this signal, but energy exceeds the maximum energy value of preserving in storehouse, another kind is in station database, not have this signal.
By the feature extraction for radio signal after the filtering of back noise threshold value of measured signal frequency spectrum data.The present invention has proposed 21 features according to the physical features of radio signal and has been respectively: signal estimation bandwidth, signal average, signal variance, signal peak-peak, the second largest peak value of signal, the third-largest peak value of signal, signal back noise level, signal ceiling capacity, be less than the number of the continuity point of back noise, be less than the equispaced of the continuity point of back noise, be greater than the number of the continuity point of back noise, be greater than the equispaced of the continuity point of back noise, the ratio of signal peak number, signal is greater than the ratio of the number of back noise, signal zero-crossing rate, level value is greater than the variance of the frequency of back noise, the mean square deviation of normalization instantaneous amplitude absolute value, the kurtosis of normalization instantaneous amplitude, the standard deviation of instantaneous amplitude absolute value, the ratio of normalization instantaneous amplitude mean square deviation and average, power spectrum signal symmetry.Wherein signal estimation bandwidth is the shared bandwidth of this signal, signal average represents the average level of this frame signal, signal variance represents the fluctuating level of this frame signal, signal peak-peak represents the peak-peak of this frame signal, the second largest peak value of signal represents the second largest peak value of this frame signal, the third-largest peak value of signal represents the third-largest peak value of this frame signal, signal back noise level represents this frame signal noise level, signal ceiling capacity represents the energy value of this frame signal maximum, the number representation signal that is less than the continuity point of back noise leaches after noise the number at zero point continuously, the equispaced representation signal that is less than the continuity point of back noise leaches after noise the equispaced between zero point continuously, the number representation signal that is greater than the continuity point of back noise leaches the number of the continuity point that is greater than zero after noise, the equispaced representation signal that is greater than the continuity point of back noise leaches the equispaced of the continuity point that is greater than zero after noise, the ratio of signal peak number represents the ratio between number and the sampled point of this frame signal peak value, the ratio that is greater than the number of back noise represents that this frame signal is greater than the ratio between counting of back noise threshold value and sampled point, signal zero-crossing rate represents that this frame signal converts the ratio of zero passage after time-domain signal to, the variance that level value is greater than the frequency of back noise represents that this frame signal is greater than the variance of the sampled point respective frequencies of back noise threshold value, the mean square deviation of normalization instantaneous amplitude absolute value represents that this frame signal converts the mean square deviation of normalization amplitude absolute value after time-domain signal to, the kurtosis of normalization instantaneous amplitude represents that this frame signal converts the kurtosis of normalization instantaneous amplitude after time-domain signal to, the standard deviation of instantaneous amplitude absolute value represents that this frame signal converts the standard deviation of instantaneous amplitude absolute value after time-domain signal to, normalization instantaneous amplitude mean square deviation and the ratio of average represent that this frame signal converts the mean square deviation of normalization instantaneous amplitude and the ratio of average after time-domain signal to, power spectrum signal symmetry represents the symmetry of this frame signal spectrogram.In actual applications, be not all features be all necessary to the cluster analysis of radio signal, the present invention adopts optimization method to extract the essential feature of radio signal cluster analysis, as genetic algorithm, rough set attribute reduction, neural net etc.
The essential feature of radio signal cluster analysis has formed the feature space of identification radio signal.If the essential feature of selecting is
, wherein
representative select the
individual essential feature,
.The present invention adopts a kind of improved FCM clustering method, in the feature space of radio signal, improper radio signal is carried out to cluster analysis, obtain respectively sweep-frequency Békésy audiometer interference, broad-band interference, arrowband by monitor signal database and disturb, illegally intercut the improper radio signals such as signal
individual cluster centre
, wherein
,
represent
the improper radio signal of class,
represent
the improper radio signal of class
individual cluster centre,
,
represent
the improper radio signal of class
of individual cluster centre
individual characteristic value,
.Make the essential feature of suspicious signal extraction be
, wherein
represent of suspicious signal extraction
individual essential feature,
.Pass through distance measure
, can obtain the degree of uncertainty that radio signal belongs to each cluster centre of each classification, wherein
representative can suspect signal and
the improper radio signal of class
the distance of individual cluster centre,
for distance measure, as Euclid distance, Minkowski distance, Hamming distance from etc.According to uncertain ranking criteria, obtain radio signal type and correlation properties again.
For a certain definite regional electromagnetic environment temporal evolution, measured signal energy is relevant to monitoring equipment simultaneously, the same signal energy difference that different monitoring equipments monitor.The present invention has considered electromagnetic environment and status of equipment, the ability of cluster analysis result being carried out to self study is provided, recognition methods can be consistent with current electromagnetic environment, improve the accuracy of identification of improper radio signal, concrete grammar is, the system automatically improper radio signal of identification is kept in volatile data base automatically, testing staff enters the improper database of final radio after these signals are made to final confirmation, system regularly adopts the improper signal library of the radio of renewal and local monitoring of environmental to carry out feature selecting Parameter Self-learning and the self study of FCM cluster result, make the electromagnetic environment that systems approach can adaptive change, accuracy of detection and the discrimination of improper radio signal are improved.
The present invention mainly has the following advantages: this measured signal detection method has the noise level for different business frequency range, automatically adjusts the function of threshold value; This measured signal detection method has universality, is applicable to all monitoring equipments; This signal recognition method adopt the method for fuzzy cluster analysis obtain after multiple cluster centres of the improper radio signal of every class according to distance measure judge can suspect signal classification, so more meet reality; This self-learning capability makes the present invention can adapt to various monitoring of environmental, and improves the accuracy of identification of improper radio signal; According to recognition result, combining wireless communication equipment, the method provides the auto-alarm function of monitoring equipment, can realize radio monitoring equipment nobody hold machine, effectively alleviate monitoring personnel's workload, improve the service efficiency of monitoring equipment.
Accompanying drawing explanation
The improper radio signal automatic recognition system of Fig. 1
Fig. 2 radio signal intelligent analysis system
The improper radio signal of Fig. 3 is identified overall flow figure automatically
The extraction flow chart of Fig. 4 segmented adaptive threshold value
Fig. 5 can suspect signal overhaul flow chart
Fig. 6 signal identification process figure
Fig. 7 self study flow chart
Fig. 8 automatic alarm flow chart
Embodiment
As shown in Figure 1, improper radio signal automatic recognition system has been expanded radio intelligent analytical system on the basis of existing radio monitoring equipment, as shown in Figure 2, radio intelligent analytical system is called the service of monitoring equipment system by communication interface, obtain frequency spectrum data, intermediate frequency data, bearing data, IQ data of radio signal etc., by these intelligent data analysis system automatic analysis data, accurately detection signal, extract signal characteristic, automatically identification signal, report to the police, by network system, analysis result reported to superior system by warning system.Monitoring equipment system comprises antenna-feedback system, environmental control system, monitoring receiver, GPS receiver, control processor and communication interface etc.Whole system is carried out interconnected communication by network system and other system or superior system.
Improper radio signal automatically identification comprises input, signal characteristic abstraction, signal identification, the several committed steps of automatic alarm, its overall flow as shown in Figure 3, first a certain frequency range is carried out the scanning of a period of time, obtain the related data of this frequency range, determine that by segmented adaptive threshold value provided by the invention method extracts the back noise threshold value of this frequency range, detailed process is shown in Fig. 4.Passing threshold determines that method extracts this frequency range and have the frequency of signal, in conjunction with the station database of having built up find out this frequency range all can suspect signal, detailed process is shown in Fig. 5.Then to carrying out intermediate frequency measurement by suspect signal, extract after corresponding essential feature in conjunction with monitor signal database, adopt signal recognition method provided by the invention analyze can suspect signal type, detailed process is shown in Fig. 6.Finally by self-learning method provided by the invention, improper radio signal is learnt, obtain multiple cluster centres that such improper radio signal is new, detailed process is shown in Fig. 7, the invention provides auto-alarm function simultaneously, automatically generate warning prompt information and send to associated terminal system, alleviated like this monitoring personnel's workload, detailed process is shown in Fig. 8.Treat that all suspicious signal analysis are complete, proceed to next band scan or finish this subtask.
Fig. 4 is the extraction flow chart of segmented adaptive threshold value, basically identical, as follows described in its processing procedure and summary of the invention: the average spectral line that records band number certificate after the band scan of the first step by a period of time
instantaneous value with present frame frequency range data
(as described in summary of the invention), thereby second step carries out segment processing according to the average spectral line of these frequency range data to be extracted this frequency range and has the frequency of signal, the 3rd step is considered to continuous according to the noise level of radio signal adjacent channel, its difference is little, thereby replace its noise level by the average that has the noise of non-signal frequency before and after the frequency of signal, extract the back noise level of whole frequency range
, the 4th step is compared with the instantaneous value of each sampled point in frequency range data and corresponding back noise, calculates the back noise threshold value of whole frequency range according to the method described in summary of the invention
, the 5th step: back noise threshold value is carried out to smothing filtering.
Fig. 5 be can suspect signal identification process figure, known station database of this process need, its processing procedure is as follows: first step passing threshold determines that method detects monitoring frequency range and has the frequency of signal, if being greater than back noise threshold value, the energy value of this channel thinks that this frequency has signal, otherwise think that it is back noise, second step is compared the frequency and the station database that there are signal, occur not recording this frequency in this frequency or station database in station database, but when exceeding the value of preserving in station database, energy all thinks that this signal is for can suspect signal.
Fig. 6 is signal identification process figure, and its processing procedure is as follows: the first step is to carrying out intermediate frequency measurement by suspect signal, extracts signal characteristic, and select essential feature with optimization method, and second step calculates the improper radio signal of every class with method of fuzzy cluster analysis
individual cluster centre,
calculate the distance of this signal to each cluster centre by distance measure again, select the minimum classification of distance, if can suspect signal be such improper radio signal when the 3rd step minimum range is less than a certain threshold value, otherwise enter the 4th step: to being further analyzed by suspect signal, judge in conjunction with expertise whether this can suspect signal be new improper radio signal, if confirm its relevant information, if not save data.
Fig. 7 is automatic learning flow chart, along with the accumulation of Monitoring Data and monitoring time, signal is also changing in time, the invention provides self-learning capability, its processing procedure is: the first step is by judging that to identification that can suspect signal whether it be without unusual radio signal, if not finish, if improper radio signal enters second step: judge whether it is new improper radio signal, if be added into monitor signal database and extract the essential feature of this type of improper radio signal, obtain multiple cluster centres of this type of improper radio signal, if can suspect signal be existing certain improper radio signal in storehouse, be added into equally monitor signal database and extract the essential feature of this improper radio signal, obtain multiple cluster centres that this type of improper radio signal is new.
Fig. 8 is automatic alarm flow chart, in the time that monitoring equipment is found improper radio signal, combining wireless communication equipment, the invention provides auto-alarm function, its processing procedure is: the first step receives the relevant information of improper radio signal, this recognition system of second step generates corresponding warning prompt information automatically, and the 3rd step sends to information relevant monitoring technology personnel automatically by Wireless Telecom Equipment.
Claims (9)
1. improper radio signal automatic recognition system comprises: radio signal monitoring equipment system, radio signal intelligent analysis system, network communicating system and a processing controller, and in described automatic recognition system
Monitoring equipment system receives aerial electromagnetic wave, carries out conversion process, produces the Monitoring Data of signal, and described Monitoring Data comprises: frequency spectrum data, speech data, bearing data, intermediate frequency measurement data etc.;
Intelligent analysis system, by Monitoring Data being carried out to a series of intellectual analysis processing, is identified improper signal automatically;
Network communicating system is responsible between the each module of system, communicating by letter between system and outside miscellaneous equipment;
Processing controller is responsible for coordinating the processing scheduling between modules; It is characterized in that:
Described radio signal intelligent analysis system, comprise: the improper signal detection module of radio, improper signal characteristic abstraction module, improper signal identification module, system self-learning module, and station database, electromagnetic environment database, improper Signals Data Base;
Wherein obtain band scan data by data-interface, adopt segmentation dynamic self-adapting thresholding algorithm to monitor out all signals and the corresponding frequency of this frequency range, signal frequency point is carried out to the detailed data of intermediate frequency measurement or medium-frequency direction finding picked up signal, analyzing and processing is extracted the various features of signal, extracted feature is inputted to improper signal identification module, identify the affiliated classification of this signal; The data that system self-learning module is used improper signal identification module to produce, regular update, improve feature extraction parameter and signal Recognition Algorithm parameter.
2. improper radio signal automatic recognition system as claimed in claim 1, wherein said radio signal monitoring equipment system comprises receiver, spectrum measurement instrument, direction-finding equipment, audiomonitor, control appliance, antenna-feedback system, and the auxiliary system such as communication, power supply, lightning protection, environmental monitoring.
3. improper radio signal automatic recognition system as claimed in claim 2, wherein said monitoring receiver receives aerial electromagnetic wave by antenna-feedback system, carries out conversion process, produce the frequency spectrum data of signal, speech data, bearing data, intermediate frequency measurement data etc.; Monitoring receiver obtains latitude and longitude coordinates data by GPS receiver; Monitoring receiver obtains the data message such as geographical environment information, climate temperature at place, monitoring station by environmental control system.
4. improper radio signal automatic recognition system as claimed in claim 1, wherein said station database is that electromagnetic environment database comprises all frequency ranges in this region Monitoring Data under normal circumstances in the data of all radio of declaring, ratifying frequency equipment of this region.
5. an automatic identifying method for improper radio signal, comprises the steps:
First by the frequency range of specifying is carried out to band scan, carry out input;
The signal extraction series of features detecting is also selected to have to the feature of better distinguishing ability;
Then use the classification of the automatic identification signal of improved FCM clustering method;
According to the signal type automatic alarm identifying;
To the improper radio signal identifying, system is preserved, and after demarcating, enters improper Signals Data Base by manual confirmation, by new database self study feature selecting algorithm parameter and the new cluster centre of FCM, make system there is the ability of unceasing study.
6. the automatic identifying method of the improper radio signal as described in claim 1-5, wherein adopt segmented adaptive Threshold to carry out described input, described segmented adaptive Threshold specifically comprises: monitoring frequency range is carried out after the monitoring of a period of time, obtain the mean value of this actual measurement frequency band signals frequency spectrum data
, wherein
represent
the average energy of this period of time of sampled point,
; And the instantaneous value of current this frame frequency segment data
, wherein
represent
the energy of current this frame of sampled point,
,
for the sampling number of frequency range data;
The first step: according to the mean value of actual measurement frequency band signals frequency spectrum data
, carry out segment processing, the size of establishing every section is
individual sampled point, calculates the average of data in this section
, wherein
for hop count, by the data in this section successively and
compare, if
, be designated as signaling point, if
, be designated as non-signaling point, wherein
,
for the threshold value of setting, take the method to find out the subscript of this frequency band signals point;
Second step: the noise level of this frequency band signals frequency is replaced by the average of non-signal frequency noise level before and after it, and for example the 2nd sampled point is signaling point, and the 1st and the 3rd is non-signaling point, and the noise level of second sampled point is
, adopt the noise level that extracts in this way whole frequency range, be made as
, wherein
represent
the noise level of individual sampled point,
;
The 3rd step: use instantaneous value
with
compare one by one, order
represent
the back noise threshold value of individual sampled point,
; If
:
if,
:
, wherein
,
for set point,
, consider noise level when noise level while having signal will be higher than no signal, requirement
;
for set point, its fluctuation that represents various monitoring equipments self generation noise and noise is better to reflect actual conditions;
The 4th step: to back noise threshold value
carry out the disposal of gentle filter; Find out the frequency of this frequency band signals according to back noise threshold value, then compare with the station database of setting up, analyze normal signal and can suspect signal, wherein can be divided into two kinds by suspect signal: one is in station database, to have this signal, but energy exceeds the maximum energy value of preserving in storehouse, another kind is in station database, not have this signal.
7. the automatic identifying method of the improper radio signal as described in claim 1-5, wherein by the feature extraction for radio signal after the filtering of back noise threshold value of measured signal frequency spectrum data, the feature of extracting is respectively: signal estimation bandwidth, signal average, signal variance, signal peak-peak, the second largest peak value of signal, the third-largest peak value of signal, signal back noise level, signal ceiling capacity, be less than the number of the continuity point of back noise, be less than the equispaced of the continuity point of back noise, be greater than the number of the continuity point of back noise, be greater than the equispaced of the continuity point of back noise, the ratio of signal peak number, signal is greater than the ratio of the number of back noise, signal zero-crossing rate, level value is greater than the variance of the frequency of back noise, the mean square deviation of normalization instantaneous amplitude absolute value, the kurtosis of normalization instantaneous amplitude, the standard deviation of instantaneous amplitude absolute value, the ratio of normalization instantaneous amplitude mean square deviation and average, power spectrum signal symmetry, adopt the methods such as genetic algorithm, rough set attribute reduction, neural net to carry out described feature extraction.
8. the automatic identifying method of the improper radio signal as described in claim 1-5, the feature of extracting of described radio signal has formed the feature space of identification radio signal, establishes the essential feature of selecting and is
, wherein
representative select the
individual essential feature,
; Wherein, by described FCM clustering method, in the feature space of radio signal, improper radio signal is carried out to cluster analysis, specifically comprises step:
Obtain respectively sweep-frequency Békésy audiometer interference, broad-band interference, arrowband by monitor signal database and disturb, illegally intercut the improper radio signals such as signal
individual cluster centre
, wherein
,
represent
the improper radio signal of class,
represent
the improper radio signal of class
individual cluster centre,
,
represent
the improper radio signal of class
of individual cluster centre
individual characteristic value,
;
Make the essential feature of suspicious signal extraction be
, wherein
represent of suspicious signal extraction
individual essential feature,
;
Pass through distance measure
, obtain the degree of uncertainty that radio signal belongs to each cluster centre of each classification, wherein
representative can suspect signal and
the improper radio signal of class
the distance of individual cluster centre,
for distance measure;
According to uncertain ranking criteria, obtain radio signal type and correlation properties.
9. the automatic identifying method of the improper radio signal as described in claim 1-5, the improper radio signal that wherein system is identified is automatically kept in volatile data base automatically, testing staff enters the improper database of final radio after these signals are made to final confirmation, system regularly adopts the improper signal library of the radio of renewal and local monitoring of environmental to carry out feature selecting Parameter Self-learning and the self study of FCM cluster result, make the electromagnetic environment that systems approach can adaptive change, improved accuracy of detection and the discrimination of improper radio signal.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210438094.1A CN103812577B (en) | 2012-11-06 | 2012-11-06 | The automatic recognition system and its method of improper radio signal |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210438094.1A CN103812577B (en) | 2012-11-06 | 2012-11-06 | The automatic recognition system and its method of improper radio signal |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103812577A true CN103812577A (en) | 2014-05-21 |
| CN103812577B CN103812577B (en) | 2017-10-13 |
Family
ID=50708838
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210438094.1A Expired - Fee Related CN103812577B (en) | 2012-11-06 | 2012-11-06 | The automatic recognition system and its method of improper radio signal |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103812577B (en) |
Cited By (24)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105897488A (en) * | 2016-06-13 | 2016-08-24 | 中南大学 | Visualization method of radio signal data |
| CN106027174A (en) * | 2014-07-09 | 2016-10-12 | 西华大学 | Radio monitoring unattended system and signal analysis and identification method thereof |
| CN106023499A (en) * | 2016-04-28 | 2016-10-12 | 北京北邮国安技术股份有限公司 | Fiber security signal dual identification method and system |
| CN106301621A (en) * | 2016-08-31 | 2017-01-04 | 成都九华圆通科技发展有限公司 | A kind of method using K line chart to find wireless station and analysis air interference |
| CN106936517A (en) * | 2015-12-28 | 2017-07-07 | 镇江高科信息科技有限公司 | A kind of automatic recognition system and its method of abnormal radio signal |
| CN107122738A (en) * | 2017-04-26 | 2017-09-01 | 成都蓝色起源科技有限公司 | Automatic Communication Signals Recognition based on deep learning model and its realize system |
| CN107710805A (en) * | 2015-07-03 | 2018-02-16 | 夏普株式会社 | Terminal installation, base station apparatus, communication means and integrated circuit |
| CN108243129A (en) * | 2017-12-15 | 2018-07-03 | 西安电子科技大学 | Radio signal characteristics extracting method based on attention depth network |
| CN108414832A (en) * | 2018-02-09 | 2018-08-17 | 中国人民解放军31007部队 | A kind of abnormal signal frequency spectrum parameter detection method based on image procossing |
| CN108631896A (en) * | 2018-05-02 | 2018-10-09 | 陕西广电网络传媒（集团）股份有限公司 | A kind of frequency spectrum discerning method for radio broadcasting filtering |
| CN109002848A (en) * | 2018-07-05 | 2018-12-14 | 西华大学 | A kind of detection method of small target based on Feature Mapping neural network |
| CN109274455A (en) * | 2018-09-05 | 2019-01-25 | 北京众谱达科技有限公司 | Applied to the radio tube control equipment under complex electromagnetic environment |
| CN109474355A (en) * | 2018-01-17 | 2019-03-15 | 国家无线电频谱管理研究所有限公司 | Adaptive noise THRESHOLD ESTIMATION and method for extracting signal based on spectrum monitoring data |
| CN109752600A (en) * | 2019-01-14 | 2019-05-14 | 电子科技大学 | Abnormal electromagnetic signal analysis system and method based on software defined radio |
| CN109973076A (en) * | 2019-04-10 | 2019-07-05 | 中煤科工集团西安研究院有限公司 | Visual detection device and method in coal mine down-hole drilling |
| CN110213001A (en) * | 2019-04-26 | 2019-09-06 | 天维讯达（北京）科技有限公司 | A kind of practical frequency point of questionable signal finds method and device accurately |
| CN111585671A (en) * | 2020-04-15 | 2020-08-25 | 国网河南省电力公司郑州供电公司 | Electric power LTE wireless private network electromagnetic interference monitoring and identifying method |
| CN112769506A (en) * | 2021-01-20 | 2021-05-07 | 成都理工大学 | Rapid radio detection method based on frequency spectrum |
| CN112867053A (en) * | 2021-01-07 | 2021-05-28 | 杭州恒生数字设备科技有限公司 | Detection server with function of frequency spectrograph |
| CN112929103A (en) * | 2021-01-26 | 2021-06-08 | 赵炳健 | Illegal broadcasting station detection method and system |
| CN113283157A (en) * | 2021-04-02 | 2021-08-20 | 殷强 | System, method, terminal and medium for predicting life cycle of intelligent stamping press part |
| CN116318457A (en) * | 2023-05-17 | 2023-06-23 | 成都中星世通电子科技有限公司 | Radio signal monitoring method and system |
| CN117147966A (en) * | 2023-08-30 | 2023-12-01 | 中国人民解放军军事科学院系统工程研究院 | Electromagnetic spectrum signal energy anomaly detection method |
| CN117176292A (en) * | 2023-11-02 | 2023-12-05 | 物空科技(四川)集团有限公司 | Wireless signal positioning detection method, device, equipment and storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1481088A (en) * | 2002-09-04 | 2004-03-10 | 北京中星微电子有限公司 | Warning typed multimedia automatic monitored control system based on ethernet |
| WO2005062531A1 (en) * | 2003-12-22 | 2005-07-07 | Huiming Zhang | A mobile monitor-collecting terminal for realizing audio/video monitoring service and the related communication system and method |
| CN201716377U (en) * | 2010-06-03 | 2011-01-19 | 四川省无线电监测站 | Microwave frequency band radio monitoring system |
-
2012
- 2012-11-06 CN CN201210438094.1A patent/CN103812577B/en not_active Expired - Fee Related
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1481088A (en) * | 2002-09-04 | 2004-03-10 | 北京中星微电子有限公司 | Warning typed multimedia automatic monitored control system based on ethernet |
| WO2005062531A1 (en) * | 2003-12-22 | 2005-07-07 | Huiming Zhang | A mobile monitor-collecting terminal for realizing audio/video monitoring service and the related communication system and method |
| CN201716377U (en) * | 2010-06-03 | 2011-01-19 | 四川省无线电监测站 | Microwave frequency band radio monitoring system |
Non-Patent Citations (1)
| Title |
|---|
| 肖乐，李兵等: "基于RMTP的电子信号智能处理系统的基础信息传输涉及", 《工业控制计算机》 * |
Cited By (36)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106027174A (en) * | 2014-07-09 | 2016-10-12 | 西华大学 | Radio monitoring unattended system and signal analysis and identification method thereof |
| CN106027174B (en) * | 2014-07-09 | 2018-10-23 | 西华大学 | Radio monitoring unattended system and signal analysis and identification method thereof |
| CN107710805B (en) * | 2015-07-03 | 2021-05-18 | 夏普株式会社 | Terminal device, base station device, and communication method |
| CN107710805A (en) * | 2015-07-03 | 2018-02-16 | 夏普株式会社 | Terminal installation, base station apparatus, communication means and integrated circuit |
| CN106936517A (en) * | 2015-12-28 | 2017-07-07 | 镇江高科信息科技有限公司 | A kind of automatic recognition system and its method of abnormal radio signal |
| CN106023499B (en) * | 2016-04-28 | 2018-05-22 | 北京北邮国安技术股份有限公司 | A kind of dual recognition methods of optical fiber security signal and system |
| CN106023499A (en) * | 2016-04-28 | 2016-10-12 | 北京北邮国安技术股份有限公司 | Fiber security signal dual identification method and system |
| CN105897488A (en) * | 2016-06-13 | 2016-08-24 | 中南大学 | Visualization method of radio signal data |
| CN105897488B (en) * | 2016-06-13 | 2019-04-09 | 中南大学 | A kind of method for visualizing of radio-signal data |
| CN106301621A (en) * | 2016-08-31 | 2017-01-04 | 成都九华圆通科技发展有限公司 | A kind of method using K line chart to find wireless station and analysis air interference |
| CN107122738A (en) * | 2017-04-26 | 2017-09-01 | 成都蓝色起源科技有限公司 | Automatic Communication Signals Recognition based on deep learning model and its realize system |
| CN108243129B (en) * | 2017-12-15 | 2019-12-24 | 西安电子科技大学 | Attention depth network-based radio signal feature extraction method |
| CN108243129A (en) * | 2017-12-15 | 2018-07-03 | 西安电子科技大学 | Radio signal characteristics extracting method based on attention depth network |
| CN109474355A (en) * | 2018-01-17 | 2019-03-15 | 国家无线电频谱管理研究所有限公司 | Adaptive noise THRESHOLD ESTIMATION and method for extracting signal based on spectrum monitoring data |
| CN108414832A (en) * | 2018-02-09 | 2018-08-17 | 中国人民解放军31007部队 | A kind of abnormal signal frequency spectrum parameter detection method based on image procossing |
| CN108414832B (en) * | 2018-02-09 | 2020-11-03 | 中国人民解放军31007部队 | Abnormal signal spectrum parameter detection method based on image processing |
| CN108631896A (en) * | 2018-05-02 | 2018-10-09 | 陕西广电网络传媒（集团）股份有限公司 | A kind of frequency spectrum discerning method for radio broadcasting filtering |
| CN109002848A (en) * | 2018-07-05 | 2018-12-14 | 西华大学 | A kind of detection method of small target based on Feature Mapping neural network |
| CN109002848B (en) * | 2018-07-05 | 2021-11-05 | 西华大学 | Weak and small target detection method based on feature mapping neural network |
| CN109274455B (en) * | 2018-09-05 | 2024-03-22 | 北京众谱达科技有限公司 | Radio monitoring equipment applied to complex electromagnetic environment |
| CN109274455A (en) * | 2018-09-05 | 2019-01-25 | 北京众谱达科技有限公司 | Applied to the radio tube control equipment under complex electromagnetic environment |
| CN109752600B (en) * | 2019-01-14 | 2021-06-08 | 电子科技大学 | Abnormal electromagnetic signal analysis system and method based on software defined radio technology |
| CN109752600A (en) * | 2019-01-14 | 2019-05-14 | 电子科技大学 | Abnormal electromagnetic signal analysis system and method based on software defined radio |
| CN109973076A (en) * | 2019-04-10 | 2019-07-05 | 中煤科工集团西安研究院有限公司 | Visual detection device and method in coal mine down-hole drilling |
| CN110213001A (en) * | 2019-04-26 | 2019-09-06 | 天维讯达（北京）科技有限公司 | A kind of practical frequency point of questionable signal finds method and device accurately |
| CN110213001B (en) * | 2019-04-26 | 2021-07-27 | 天维讯达（北京）科技有限公司 | Actual frequency point alignment method and device for suspicious signals |
| CN111585671A (en) * | 2020-04-15 | 2020-08-25 | 国网河南省电力公司郑州供电公司 | Electric power LTE wireless private network electromagnetic interference monitoring and identifying method |
| CN112867053A (en) * | 2021-01-07 | 2021-05-28 | 杭州恒生数字设备科技有限公司 | Detection server with function of frequency spectrograph |
| CN112769506A (en) * | 2021-01-20 | 2021-05-07 | 成都理工大学 | Rapid radio detection method based on frequency spectrum |
| CN112929103A (en) * | 2021-01-26 | 2021-06-08 | 赵炳健 | Illegal broadcasting station detection method and system |
| CN113283157A (en) * | 2021-04-02 | 2021-08-20 | 殷强 | System, method, terminal and medium for predicting life cycle of intelligent stamping press part |
| CN116318457A (en) * | 2023-05-17 | 2023-06-23 | 成都中星世通电子科技有限公司 | Radio signal monitoring method and system |
| CN117147966A (en) * | 2023-08-30 | 2023-12-01 | 中国人民解放军军事科学院系统工程研究院 | Electromagnetic spectrum signal energy anomaly detection method |
| CN117147966B (en) * | 2023-08-30 | 2024-05-07 | 中国人民解放军军事科学院系统工程研究院 | Electromagnetic spectrum signal energy anomaly detection method |
| CN117176292A (en) * | 2023-11-02 | 2023-12-05 | 物空科技(四川)集团有限公司 | Wireless signal positioning detection method, device, equipment and storage medium |
| CN117176292B (en) * | 2023-11-02 | 2024-01-05 | 物空科技(四川)集团有限公司 | Wireless signal positioning detection method, device, equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103812577B (en) | 2017-10-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103812577A (en) | Method for automatically identifying and learning abnormal radio signal type | |
| CN106936517A (en) | A kind of automatic recognition system and its method of abnormal radio signal | |
| CN110287552B (en) | Motor bearing fault diagnosis method and system based on improved random forest algorithm | |
| CN116304766B (en) | Multi-sensor-based quick assessment method for state of switch cabinet | |
| CN105424395A (en) | Method and device for determining equipment fault | |
| CN107294213B (en) | Intelligent monitoring system for power grid equipment | |
| CN103323749A (en) | Multi-classifier information fusion partial discharge diagnostic method | |
| Honghao et al. | Spectrum anomalies autonomous detection in cognitive radio using hidden markov models | |
| CN104994334A (en) | Automatic substation monitoring method based on real-time video | |
| CN106358212B (en) | The detection method and device of indoor distributed system | |
| CN111080105A (en) | Transformer area user-to-user relationship identification method and system based on voltage time sequence data | |
| CN109874148B (en) | Antenna feeder anomaly detection method, device and system and computer equipment | |
| CN105488260B (en) | A kind of aircraft noise identification and non-aircraft noise filter method | |
| CN104751285A (en) | Accuracy judgment and warning system for front-end collected data in automation of electric network management | |
| CN108337645B (en) | Gridding radio signal monitoring system architecture and interference signal joint detection method | |
| CN107257351A (en) | One kind is based on grey LOF Traffic anomaly detections system and its detection method | |
| CN103019172A (en) | Self-organizing radio monitoring system and method | |
| CN116224925B (en) | Intelligent processing management system | |
| CN106875613A (en) | A kind of fire alarm Situation analysis method | |
| CN114371353A (en) | Power equipment abnormity monitoring method and system based on voiceprint recognition | |
| CN105974223B (en) | A kind of method and system for on-line checking electrical equipment working condition | |
| CN109948738A (en) | Energy consumption method for detecting abnormality, the apparatus and system of coating drying room | |
| CN116845863A (en) | Terminal power grid abnormal equipment access alarm method and device based on space-time track | |
| CN114662058B (en) | Wireless station monitoring method and device | |
| CN101923605B (en) | Wind pre-warning method for railway disaster prevention |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20171013 Termination date: 20191106 |