CN102456158A - Security assessment method for air traffic management (ATM) information system based on ANNBP (Artificial Neural Network Blood Pressure) model - Google Patents
Security assessment method for air traffic management (ATM) information system based on ANNBP (Artificial Neural Network Blood Pressure) model Download PDFInfo
- Publication number
- CN102456158A CN102456158A CN2010105198364A CN201010519836A CN102456158A CN 102456158 A CN102456158 A CN 102456158A CN 2010105198364 A CN2010105198364 A CN 2010105198364A CN 201010519836 A CN201010519836 A CN 201010519836A CN 102456158 A CN102456158 A CN 102456158A
- Authority
- CN
- China
- Prior art keywords
- network
- blank pipe
- error
- infosystem
- training
- 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
- 238000000034 method Methods 0.000 title claims abstract description 46
- 238000013528 artificial neural network Methods 0.000 title claims description 25
- 230000036772 blood pressure Effects 0.000 title 1
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 17
- 238000004891 communication Methods 0.000 claims abstract description 12
- 238000011156 evaluation Methods 0.000 claims description 52
- 238000012549 training Methods 0.000 claims description 43
- 230000006870 function Effects 0.000 claims description 29
- 238000012360 testing method Methods 0.000 claims description 18
- 238000004458 analytical method Methods 0.000 claims description 17
- 230000008569 process Effects 0.000 claims description 12
- 238000003062 neural network model Methods 0.000 claims description 6
- 238000013439 planning Methods 0.000 claims description 6
- 210000002569 neuron Anatomy 0.000 claims description 5
- 238000004088 simulation Methods 0.000 claims description 5
- 230000001419 dependent effect Effects 0.000 claims description 4
- 238000007726 management method Methods 0.000 claims description 4
- 230000008859 change Effects 0.000 claims description 3
- 230000009466 transformation Effects 0.000 claims description 3
- 239000000203 mixture Substances 0.000 claims description 2
- 230000001131 transforming effect Effects 0.000 claims description 2
- 238000006243 chemical reaction Methods 0.000 claims 4
- 230000001537 neural effect Effects 0.000 claims 2
- 238000012546 transfer Methods 0.000 claims 2
- 230000003292 diminished effect Effects 0.000 claims 1
- 238000009472 formulation Methods 0.000 claims 1
- 230000008676 import Effects 0.000 claims 1
- 238000010606 normalization Methods 0.000 claims 1
- 229920006395 saturated elastomer Polymers 0.000 claims 1
- 238000013341 scale-up Methods 0.000 claims 1
- 238000005516 engineering process Methods 0.000 abstract description 5
- 210000005036 nerve Anatomy 0.000 abstract 4
- 238000012544 monitoring process Methods 0.000 abstract 1
- 238000011160 research Methods 0.000 description 5
- 238000012545 processing Methods 0.000 description 4
- 230000008901 benefit Effects 0.000 description 3
- 238000013278 delphi method Methods 0.000 description 3
- 238000011161 development Methods 0.000 description 3
- 230000018109 developmental process Effects 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000013210 evaluation model Methods 0.000 description 3
- 238000011002 quantification Methods 0.000 description 3
- 238000013461 design Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000007429 general method Methods 0.000 description 2
- 230000036541 health Effects 0.000 description 2
- 230000009545 invasion Effects 0.000 description 2
- 238000012417 linear regression Methods 0.000 description 2
- 238000013178 mathematical model Methods 0.000 description 2
- 230000003449 preventive effect Effects 0.000 description 2
- 238000000513 principal component analysis Methods 0.000 description 2
- 238000012847 principal component analysis method Methods 0.000 description 2
- VUFNLQXQSDUXKB-DOFZRALJSA-N 2-[4-[4-[bis(2-chloroethyl)amino]phenyl]butanoyloxy]ethyl (5z,8z,11z,14z)-icosa-5,8,11,14-tetraenoate Chemical compound CCCCC\C=C/C\C=C/C\C=C/C\C=C/CCCC(=O)OCCOC(=O)CCCC1=CC=C(N(CCCl)CCCl)C=C1 VUFNLQXQSDUXKB-DOFZRALJSA-N 0.000 description 1
- 241000700605 Viruses Species 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000001186 cumulative effect Effects 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000009795 derivation Methods 0.000 description 1
- 230000001066 destructive effect Effects 0.000 description 1
- 230000009977 dual effect Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000012854 evaluation process Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000013277 forecasting method Methods 0.000 description 1
- 238000010230 functional analysis Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 230000000644 propagated effect Effects 0.000 description 1
- 230000001902 propagating effect Effects 0.000 description 1
- 238000004451 qualitative analysis Methods 0.000 description 1
- 238000000611 regression analysis Methods 0.000 description 1
- 238000012827 research and development Methods 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 230000033772 system development Effects 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
- 238000005303 weighing Methods 0.000 description 1
Images
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a security assessment method for an air traffic management (ATM) information system based on an ANNBP model. The security assessment method adopts a nerve network technology principle to carry out security assessment to an air management information system, and is a novel method for assessing the air management system in an information security domain. The air management information system comprises four parts, namely four systems such as a communication system, a navigation system, a monitoring system and an automatic system. The nerve network starts from the indexes which affect the air management information system, so that specific indexes which affect the security of the four systems in the air management information system are used as the input of the nerve network. The whole network adopts an error back-propagation algorithm to compare network output with target output; if the difference of the output value of the network and the target value is out of the error range, a gradient descent algorithm is adopted to adjust the weight value until the difference satisfies the range which is allowed by the error. The network final output is the security level of the whole air management information system. And the integral security of the air management information system can be precisely estimated by a nerve network model.
Description
Technical field
The present invention a kind ofly utilizes nerual network technique (be called for short: blank pipe) ATM (Air Traffic Management) infosystem is carried out the method for safety assessment (Security Evaluation) to the empty traffic administration of aviation.It relates to blank pipe infotech and safety technique, belongs to safety assessment field in the information security field.
Background technology
The blank pipe information resources are to guarantee to realize the basic resource of air traffic safety management.The blank pipe informatization has got into the important period of pushing forward comprehensively and accelerating development.Through building for many years, in each important service department of air traffic control system, built up multiple infosystem, Computer information network has become the groundwork means of blank pipe service operation.In the face of blank pipe infosystem in large scale, that application is various, the user is huge and professional degree of dependence is high, the extensive information-based and networked outstanding feature that has become the air traffic control system development.
At present, the network security situation allows of no optimist both at home and abroad, and the blank pipe infosystem faces huge threat (invasion, attack and virus etc.).Local and overseas hostile forces are becoming increasingly rampant to satellite, the wireless and attack destructive activity of ground network and the invasion breaking-up activity that utilizes information network to carry out.In case the blank pipe infosystem is attacked, radio communications system is interfered, the system failure etc.; Can make flight safety receive serious threat; The orchestration of blank pipe will be had a strong impact on, and the lighter can cause flight to operate interruption normally, and weight person understands critical flight safety; Serious harm public interest and national security.
Blank pipe is not only the important support point of whole aviation development, and is the important duty of any one sovereign state.So, the relation of conscientious correct handling blank pipe and air defence, strengthen the integrated construction of blank pipe air defence, make between system to organically combine, to interconnect, information sharing and operation are efficiently.Any information security risk is all with the directly safe and stable operation of the whole Civil Aviation System of influence; Influence the operate as normal running of civil aviaton; Even possibly injure country, people life property safety, so the blank pipe information security is the major issue that is directly connected to national economy.
Security evaluation to air traffic control system is an important technology of Prevention-Security, also is the important component part of information security engineering science.Through safety assessment, scientifically analyze the security status of blank pipe infosystem, the general safety situation of system is made evaluation, take preventive measures.Setting up sound Civil Aviation Air guard system index system of safety is to implement the important assurance of civil aviaton's information security strategy; By the security assurance information assessment indicator system China's aviation infosystem and core business system are carried out unified Analysis and aspect ratio; To help China's aviation information Prevention-Security situation is made the conclusion of quantification; For CAAC provides decision support; The planning that China aviation information security is built, the input that information security is built, so the research and development of information security management policy making, information security technology is all significant.
Assessment to security risk needs certain foundation and MOI, and the current evaluation method that obtains the comparison widespread use has: Delphi method, FTA, analytical hierarchy process, PCA, DEA method, fuzzy analysis and gray theory etc.
Delphi method (Delrhi method) is not referred to as the expert consulting method yet, results from sciemtifec and technical sphere at first, is applied to the prediction in any field afterwards gradually, like military forecasting, population forecast, health care prediction, operation and demand forecast, educational forecasting etc.In addition, also be used for estimating, decision-making, managing communication and planning.Delphi method is a kind of qualitative forecasting method; Method through back-to-back group decision-making consulting; Group member works alone separately, then with system, independently comprehensive their judgement of mode, overcome for some authority about shortcoming; Reduce respondent's psychological pressure, forecasting reliability is increased.
Fault tree analysis (Fault Tree Analysis) model is proposed in 1961 by the Waston.H.A of Bell phone testing laboratory, as the mathematical model of analytic system reliability, has become fairly perfect systems reliability analysis technology at present.The fault tree analysis method can be divided into qualitative and quantitative dual mode.The qualitative analysis of fault tree is exactly through asking the minimal cut set of fault tree, obtain whole fault modes of top event, to find the weakest environment or the most critical position on the system architecture, concentrating strength on the key position that minimal cut set is found is strengthened.
Analytical hierarchy process (Analytic Hierarchy Process) is to be put forward in 1970's by the famous operational research expert Satie of the U.S., is the Multi-objective Decision Analysis method that a kind of qualitative and quantitative combines.At present, the AHP educational circles that planned strategies for is regarded as simple and effective Multiobjective Decision Making Method.The range of application of AHP is enlarging gradually; Up to now; AHP has been applied to the analysis and the decision-making in a lot of fields, as: economic analysis and plan, human behavior science, health care, accounting, sociology, education, the talent, military commanding, geography, OR Methods evaluation, architecture, scientific technological advance, environment, law etc.AHP is as a kind of policy-making thought mode, to the important effect of the various decision process play more and more of people.
Principal component analysis (Principal Component Analysis) is a branch of multivariate statistical analysis, and before this by Karl, Pearson is applied to nonrandom vector, and then Hotelling has been generalized to random vector with it.Principal Component Analysis Method is to use mathematical statistics and linear algebra knowledge; The former random vector that its component is relevant by means of an orthogonal transformation, changes into the incoherent new random vector of its component; And with variance estimating as quantity of information; New random vector is carried out dimension-reduction treatment use decision analysis and functional analysis knowledge again,, further change into unidimensional system to low-dimensional system through constructing suitable cost function.Principal Component Analysis Method is widely used in the evaluation of numerous objects in the fields such as economy, society, science and education, environmental protection with the characteristics such as objectivity of its theoretical terseness, tax power.
DEA (Data Envelopment Analysis) is to be the basis with the relative efficiency notion, the unit (department or enterprise) of same type is carried out a kind of new method of relative effectiveness or benefit evaluation according to many indexs input and many indexs output.Since 1978; Since at first proposing the C2R model and be used to estimate interdepartmental relative effectiveness by the famous scholar of planning strategies for Cha Ensi, Ku Bai and Lodz; The DEA method constantly improves; The evaluation that often is applied to enterprise operation overall efficiency, industry production situation, community service department facility etc. is particularly at the community service department to non-simple profit, and like school, hospital, the evaluation aspect of some cultural facility etc. is considered to a kind of effective method.
Nineteen sixty-five; U.S. kybernetics scholar L.A.Zadeh professor has delivered the famous paper of " fuzzy set " (Fuzzy Set) this piece on " Information and Control " magazine; Propose to describe the middle transition of phenomenon difference, thereby broken through the absolute relation that belongs in the classical set theory or do not belong to this notion of subordinate function.Zadeh teaches this initiative work, indicates a new branch---the birth of fuzzy mathematics of mathematics.Fuzzy theory also provides a kind of new method for Comprehensive Evaluation Problem.Application of Fuzzy theory is set up and is estimated mathematical model; Can make evaluation index quantification qualitatively; Quantitative fuzzy evaluation mark sense accuracy is approached; Make evaluation method have more science, practicality, be widely used in various fields such as society, economy, military affairs, engineering at present, obtained a large amount of achievements in research.
Gray system theory is that Chinese scholar Deng Julong is taught in nineteen eighty-two and at first proposes, and through 20 years of development, sets up the structural system of a new branch of science basically.So-called gray system is meant in the system not only adularescent parameter (known parameters) but also the system of black parameter (unknown parameter) arranged, and its research contents comprises the quantification, modeling, prediction, decision-making, control of objective things etc.Gray system theory is from the non-completeness research of information and the theory of dealing with complicated system; It is not to leave for the research system from the internal system special regularity; But, reach on higher level and understand mechanism such as internal system variation tendency, mutual relationship through to the observational data of a certain level of system mathematics manipulation in addition.Under the drive of gray system theory, a collection of emerging cross disciplines such as grey geology, grey thremmatology, gray control theory, grey chaology, the analysis of regional economy gray system, grey axiology, comprehensive anti-of grey have also been produced in succession.
In evaluation problem, the relation overwhelming majority between objective attribute target attribute is a nonlinear relationship, and general method is difficult to this relation of reflection.The information source of many problems is imperfect, and evaluation rule is usually conflicting, and irrationality can follow sometimes.People are difficult to each mutuality of objectives of description scheme exactly usually; More can't express the weight allocation between them with quantitative relation formula; If can utilize existing scheme and evaluation result thereof, just can directly make evaluation according to the characteristic of giving new departure to scheme, then not only can reduce artificial uncertain factor; Improve the accuracy of evaluation result, can also alleviate estimator's burden greatly.
Nerual network technique can address the above problem effectively.The Nonlinear Processing ability of neural network has broken through the limitation based on the existing evaluation method of linear process; General evaluation method is ambiguous, imperfect in information, exist in the complex environment such as contradiction and often be difficult to use, and nerual network technique then can be crossed over this obstacle.
Summary of the invention
The present invention at first analyzes the demand for security of blank pipe, proposes the 3 layers of neural network assessment models of blank pipe infosystem based on the BP neural network.Then according to the BP neural network model of setting up; With the main safety indexes of blank pipe infosystem as training sample; The data that provide are to find out the inner link between input and the output through study and training; BP network with training is assessed the blank pipe infosystem, and compares with traditional assessment method.Experimental result shows that this network has very strong adaptivity and fault-tolerant ability, and this model is used for the safety assessment of blank pipe infosystem, be consistent with actual result, and tool has great advantage and potentiality.
The objective of the invention is, overcome the deficiency of prior art and the influence of artificial subjective factor.Because in evaluation problem, the relation overwhelming majority between objective attribute target attribute is a nonlinear relationship, and general method is difficult to this relation of reflection; The information source of many problems is imperfect, and evaluation rule is usually conflicting, and irrationality can follow sometimes; People are difficult to each mutuality of objectives of description scheme exactly usually; More can't express the weight allocation between them with quantitative relation formula; If can utilize existing scheme and evaluation result thereof, just can directly make evaluation according to the characteristic of giving new departure to scheme, then not only can reduce artificial uncertain factor; Improve the accuracy of evaluation result, can also alleviate estimator's burden greatly.Nerual network technique can address the above problem effectively.The Nonlinear Processing ability of neural network has broken through the limitation based on the existing evaluation method of linear process; General evaluation method is ambiguous, imperfect in information, exist in the complex environment such as contradiction and often be difficult to use, and nerual network technique then can be crossed over this obstacle.Utilize the method for neural network assessment, can in time find potential safety hazard and risk in the blank pipe infosystem, the safe input of making rational planning for is taked corresponding safe precaution measure, increases economic efficiency.
Realize that technical solution of the present invention is:, set up the neural network assessment models according to the characteristics of blank pipe infosystem.Characteristics according to network are analyzed the security of system, set up the network model that is applicable to the blank pipe infosystem, and its concrete performing step is:
The safety assessment neural network model of the blank pipe infosystem of 1) building is as shown in Figure 1, and this model is the system of input more than, single output.The input of this model is 12 evaluation index values of blank pipe infosystem, and evaluation index is as shown in Figure 2.Replace traditional assessment method with the BP network that has trained, provide assessment result through neural network.
2) evaluation index chooses
The Civil Aviation Air guard system is a complicacy and huge system; The data information that relates to aspects such as the communication relevant, navigation, supervision, meteorology, information, air traffic control with the blank pipe operation; And related hardware, like communication facilities, communication media, radar navigation set, meteorological equipment, navigational intelligence equipment etc.With system's orderliness layering, be the prerequisite of carrying out safety assessment.
Facts have proved that a good air traffic control system safe evaluation method should satisfy following requirement: evaluation index can reflect the situation and the technical quality characteristic of air traffic control system all-sidedly and accurately; Evaluation model is simple and clear, and is workable, is easy to grasp; Evaluation conclusion can reflect rationality, integrity and the safe reliability of air traffic control system; The data that adopted in the evaluation are easy to obtain, and the data processing work amount is little; Each evaluation index has clear and definite evaluation criterion.Based on above condition, from communication, navigation is kept watch on, and 12 main influence factors have been chosen as evaluation index in 4 aspects of robotization, have constituted a scientific and reasonable air traffic control system assessment indicator system.
The index that influences blank pipe safety mainly comprises communication, and four aspects of robotization are kept watch in navigation.Specific targets are respectively (1) satellite system, (2) VHF (Very High Frequency; Very high frequency(VHF)) system, (3) interior telephone system, (4) DVOR (Doppler VHF Omnidirectional Range; Doppler VHF omnirange) system, (5) ILS (Instrument Landing System; The instrument landing system) system, (6) DME (Distance Measuring Equipment; Distnace determination device) system, (7) SSR (Secondary Surveillance Radar; Secondary surveillance radar) system, (8) ADS (Automatic Dependent Surveillance, automatic dependent surveillance) system, (9) GPS (Global Position System, GPS) system, (10) flight planning, (11) flight information, (12) supervisory system.The comprehensive assessment index system of blank pipe infosystem is as shown in Figure 2.
3) division of safe class
Usually assessment result is divided into four grades, representative is very safe respectively, compares safety, danger, danger close.Shown in the table 1:
Table 1 safe class is divided table
| Very safe | Compare safety | Dangerous | Danger close |
| 0.85-1 | 0.75-0.85 | 0.6-0.75 | 0-0.6 |
| A | B | C | D |
4) confirm the neuron number of BP network
The input layer number of BP network depends on the number of the safety indexes that influences the blank pipe infosystem.Security Evaluation Model by the blank pipe infosystem can get, and its number is 12.And its output layer is exactly a safety indexes number of weighing the blank pipe infosystem, and its number is 1, and net result is safe class.Utilize the function newrb among the Matlab to create a BP neural network.
5) BP network hidden neuron is chosen
For Network Design, wherein neuronic number influences the test performance of network to a great extent in the hidden layer.Defeated people's layer of this network and hidden layer, and the transport function between hidden layer and the output layer adopts logarithmic function; Consider the specification and the learning time of network, select for use the Trainlm function that function is trained, maximum training step number epochs is 1000; The least error goal that sets is 0.01; Show is 20.Other parameters are default value.
6) training sample is chosen
Training sample is chosen 12 groups of blank pipe stations and is adopted data in fact, and training sample is as shown in table 2, and through repetition test, the performance of finally choosing network when neuronic number is 25 in the hidden layer is best.The service check of BP neural network.Utilize Matlab that network is carried out emulation.Network utilizes function T rainlm that network is trained through initialization, and after training for 3 steps, network error has reached the error requirements of setting, and training finishes, and is as shown in Figure 5.
Table 2 training sample
7) network performance check
For the performance of network after further check is trained, make further simulation analysis in the face of training result down.Utilize the postreg function to do nonlinear regression analysis, and obtain both related coefficients, thereby can be used as the good and bad distinguishing rule of network training result the output result and the target output of network simulation.We utilize the emulation output vector of network and the linear regression analysis between the target vector, and the target vector that obtains to the related coefficient of network output important evaluation sign as network performance.If network performance is good, the network analog value that obtains so should equate with the network real output value, promptly be on the diagonal line of coordinate axis first quartile; Intercept equals 0; Slope equals 1, and degree of fitting equals 1, and it is just passable greater than 0.80 to get degree of fitting R in the practical application usually.The Nonlinear regression equation that obtains at last between BP neuron network simulation value and the actual output is: A=0.998T+ (0.00418) (R=0.99), the result is satisfied, network performance is fine.As shown in Figure 6.
8) network test
The network that trains more than utilizing is below tested the performance of network, gets four groups of backs as test sample book, and test sample book is as shown in table 3.
Table 3 test sample book
Can find out from the test result of network, not only prove the network feasibility of design, and the stable performance of network, accuracy is high, and error is very little, and test result matches with actual result.
The self-learning capability that this network had greatly facilitates the memory and the extraction of knowledge again; Network can extract the rule, the association that are comprised and handle particular problem, and can carry out completion to imperfect information through study from typical example.From the angle of estimating; Neural network is through to the study of existing program and evaluation result thereof, can obtain implicit people's wherein experience, knowledge and to the intuitive thoughts such as view of each target importance, in case when being used for estimating; Network just can reproduce these experiences, knowledge and intuitive thought; Rational judgement made in challenge, both embodied people's subjective judgement thus, significantly reduced disadvantageous artificial factor in the evaluation procedure again.It is thus clear that neural network is the effective way of multiattribute comprehensive evaluation.
9) assessment result explanation
The BP neural network has very strong approaching property.Each item content that is comprised with " safe and reliable, the economical rationality " of air traffic control system and its definition is a foundation; Communicate by letter from blank pipe system main composition; Navigation is kept watch on and main 12 typical evaluation indexes that the automated system four levels has been established air traffic control system, adopts the assessment indicator system of a cover system; Foundation is based on the air traffic control system safety evaluation model of BP neuroid, and finds the solution.The test result of this model shows that the air traffic control system assessment indicator system that is adopted has reflected the situation of air traffic control system well, and the BP network algorithm that is adopted meets the nonlinear characteristic of air traffic control system, can be used for the safety evaluatio of complicated air traffic control systems such as civil aviaton.
Description of drawings
Fig. 1 blank pipe safety assessment neural network model figure
Fig. 2 blank pipe infosystem evaluation index system figure
Fig. 3 BP network model figure
Fig. 4 BP flow through a network figure
The convergence of Fig. 5 network training, error curve diagram
The non-linear regression figure of Fig. 6 network
Embodiment
Shown in Figure 2; The Civil Aviation Air guard system is a complicacy and huge system; The data information that relates to aspects such as the communication relevant, navigation, supervision, meteorology, information, air traffic control with the blank pipe operation; And related hardware, like communication facilities, communication media, radar navigation set, meteorological equipment, navigational intelligence equipment etc.As by evaluation object, air traffic control system is made up of a plurality of core systems, belongs to typical many inputs, multiple-target system.With system's orderliness layering, be the prerequisite of carrying out safety evaluation.
After the evaluation index system of blank pipe infosystem has been arranged, will assess the security of whole blank pipe infosystem, in conjunction with Artificial Neural Network, Fig. 1 is the illustraton of model to the total system evaluation process.
Formulate the evaluation grade standard:, safe class is divided into A, B, C, the D fourth class according to the characteristics of blank pipe infosystem.Very safe, relatively safety, dangerous, danger close of correspondence respectively.The corresponding corresponding number interval that divides of each grade of model, the comprehensive assessment score value of gained is high more, explains that the security of whole blank pipe infosystem is just high more.
BP network model, i.e. error backpropagation algorithm are mainly adopted in network evaluation.The BP network model is that a kind of forward is found the solution, reverse propagated error and reach and revise the network model that connects flexible strategy between the network layer, and it is divided into input layer, hidden layer and output layer usually, and wherein hidden layer also maybe more than one.In forward-propagating, input signal passes through transforming function transformation function to the forward-propagating successively of latent layer from input layer; Then, the size of network based training error is compared the result who obtains output layer with desired output; If deviation exceeds allowed band, then regulate weights and threshold value automatically, promptly the contrary of error transmits; Output error signal is reduced, thereby make network output force into desired output, reach usually till the error mean square difference minimum.The study of network is broadcast contrary alternately the realizing repeatedly of propagating with error through the saequential transmission of pattern.Generally speaking, the BP network model adopts the strategy of momentum method and learning rate adjustment, thereby has improved the performance of network, has reduced it and has been absorbed in local minimum and improves speed of convergence.The BP network model is as shown in Figure 3, and derivation slightly in detail.
Algorithm realization flow for whole network is as shown in Figure 4.The performing step of BP back-propagation algorithm is following:
Step 6. is calculated the hidden layer elemental error;
Step 7. is asked error gradient;
Step 8. pair weights and threshold values are made amendment, and error signal is returned along original connecting path, through revising the neuronic weights of each layer, propagate to input layer one by one and go to calculate.Pass through the forward-propagating process again, these two processes make error signal minimum repeatedly;
Step 9. is got back to step 3;
Whether step 10. is calculated whole errors and is met the demands, and learns as satisfying then, otherwise gets back to step 6;
Two kinds of weights methods of adjustment are arranged in practical application at present.Can find out from above step, in standard BP algorithm, sample of every input. all will return error and adjust weights, this weights method of adjustment to each sample training in rotation is called single sample training again.Since single sample training is followed be live in the present " selfish departmentalism " and principle. the error that only produces to each sample is adjusted, and turns round and look at this unavoidably and loses quilt, and the number of times of whole training is increased, and causes speed of convergence short excessively.Another kind method is after all sample inputs. the total error E of computational grid
Always:
Calculate the error signal of each layer and adjust weights according to total error then, the batch processing mode of this cumulative errors is called batch (Batch) training or cycle (Epoch) training.Because batch training has been followed reducing " collectivism " principle that global error is a target, thereby can guarantee that total error is to reducing power to variation.More for a long time, criticize the fast convergence rate when training at sample number than single sample training.The training process flow diagram is like figure.The program of this problem adopts batch BP algorithm flow of training.
After having set up the neural network assessment models of blank pipe infosystem; Utilize the BP error backpropagation algorithm that network is carried out training and testing, when network training arrived the scope of error permission, network training finished; With test data the network that trains is tested; When test result meets the demands, show that network performance is good, can be applicable in the assessment of blank pipe infosystem.Training and testing for network adopts the MATLAB instrument to carry out emulation and analysis.Emulation and analysis result such as Fig. 5 and Fig. 6.Concrete implementation procedure is following, wherein introduces the main function that relates to.
Utilize the matlab instrument,, introduce the main function that this program is used in conjunction with above algorithm.Utilize newff () function creation network, function net.iw{1,1} and net.b{1} are initial weight and threshold function table.Function net.trainParam.show function is the graphic presentation interval; Function net.trainParam.lr is a learning rate; Function net.trainParam.ePochs is the maximum step number that allows; Function net.trainParam.goal is the permissible error scope; Train () is the network training function; Function sim () is the network simulation function; Function postreg () realizes the nonlinear fitting to network.
Good blank pipe infosystem neural network model carries out comprehensive assessment to the blank pipe information system security to utilize training and testing at last, exports evaluation grade at last and provides corresponding assessment report.Make corresponding strick precaution and innovative approach according to assessment report, can risk is reduced to minimum level even avoid dangerous generation, take preventive measures.And improve the air traffic control system overall economic benefits, national product is played the effect of promotion.
Claims (6)
1. (be called for short: blank pipe) infosystem comprises four major parts, communication system, navigational system, surveillance and automated system four big systems for ATM, Air Traffic Management in air traffic control.Neural network is based on from influencing the index of blank pipe infosystem, with the input of 12 specific targets that influence four big securities of system in the blank pipe infosystem as neural network.Be respectively (1) satellite system, (2) VHF (Very High Frequency, very high frequency(VHF)) system, (3) interior telephone system; (4) DVOR (Doppler VHF Omnidirectional Range, Doppler VHF omnirange) system, (5) ILS (Instrument Landing System; The instrument landing system) system, (6) DME (Distance Measuring Equipment, Distnace determination device) system; (7) SSR (Secondary Surveillance Radar, secondary surveillance radar) system, (8) ADS (Automatic Dependent Surveillance; Automatic dependent surveillance) system, (9) GPS (Global Position System, GPS) system; (10) flight planning, (11) flight information, (12) supervisory system.Adopt neural net method to carry out safety assessment to above system; Wherein neural network algorithm adopts error repercussion propagation algorithm; Set up the neural network assessment models of blank pipe infosystem, total system is assessed, provide assessment result and assessment report at last.
2. according to the blank pipe infosystem neural network assessment models described in the claim 1; It is characterized in that said neural network model assessment may further comprise the steps: the safe comprehensive assessment index system of 1) confirming the blank pipe infosystem; The blank pipe infosystem is made up of four big systems, wherein influences the evaluation index system of 12 indexs of this four big security of system as the blank pipe infosystem; 2) the evaluation grade standard of formulation blank pipe infosystem; 3) set up the neural network model that is suitable for the blank pipe infosystem; 4) choose the blank pipe achievement data and import data, adopt the error inverse-transmitting method that network is trained as network; 5) through repetition training, and through behind the simulation analysis, the optimal network structure in the selection training process is as the network training model; 6) choose other several groups of data, non-training data is as the test data of network, and network performance is tested; 7) whole air traffic control system is carried out comprehensive assessment, provide assessment result and assessment report; 8) problem and the suggestion of proposition to existing.
3. according to the neural network assessment models of the blank pipe infosystem described in the claim 1, it is characterized in that described BP neural network error back propagation algorithm, wherein the weights adjustment is to combine the use of gradient descent algorithm.
1. network error
When the network input does not wait with desired output, there is output error E, define as follows:
Further being expanded to input layer has:
2. weights adjustment
The network error originated from input is each layer weights ω
Jk, v
IjFunction, therefore adjust weights and can change error E.Obviously, the principle of adjustment weights is that error is constantly diminished, and therefore should make the adjustment amount of weights and the gradient of error be declined to become direct ratio, that is:
Negative sign representes that gradient descends in the formula, and constant η ∈ (0,1) representes scale-up factor, in training, has reflected learning rate.Can find out that the BP algorithm belongs to δ learning law class, is the gradient descent algorithm of error.The weights adjustment formula of the BP learning algorithm of three-layer network is:
4. according to 12 indexs in the information system integrated evaluation index system of the blank pipe described in the claim 1; With its input as network; And screen corresponding blank pipe service data, and these data are carried out normalization handle, the neuron of BP net all adopts the sigmoid transfer function; Can prevent after the conversion to make neuron output saturated, make the weights adjustment get into the flat region of error curved surface then because of the absolute value of clean input is excessive; The output of Sigmoid transfer function 0:1 or-1:1 between; As the output data of teacher signal as not carrying out conversion process; Certainly will make the big output component absolute error of numerical value big, the output component absolute error that numerical value is little is little, only adjusts piece value to the total error of output during network training; It is bigger consequently in total mistake Qiang, to account for the little output component relative error of share, output quantity is carried out change of scale after this problem can be readily solved.In addition, when each component dimension of input or defeated little vector not simultaneously, tackle different components and in its span, carry out conversion respectively, identical and when being same dimension when each component physical significance. should in whole data area, confirm maximal value x
MaxWith minimum value x
MinThe conversion process of uniting.
Inputoutput data is transformed to [0,1] interval value following transforming function transformation function commonly used:
X wherein
iRepresent inputoutput data, x
MinRepresent the minimum value of variation range, x
MaxRepresent the maximal value of variation range.With the real input of the data after handling as network, adopt the method in the claim 3 that network is trained, training stops after error satisfies allowed band, and this moment, network training finished.
5. for the blank pipe infosystem in the claim 1 with according to the blank pipe infosystem neural metwork training network in the claim 4, adopt proper data that network is tested.If test result is satisfied then network performance is good, if test result is unsatisfied with then returns 4 from new training network.Through repeatedly repetition training, find the neural network structure of the most suitable blank pipe infosystem, and set up the neural network assessment models.
6. according to blank pipe infosystem in the claim 1 and the neural network blank pipe assessment models in the claim 5, the air traffic control system of reality is assessed, provided the comprehensive assessment result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201010519836.4A CN102456158B (en) | 2010-10-26 | 2010-10-26 | Based on the air traffic control atm information system security assessment method of ANN BP model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201010519836.4A CN102456158B (en) | 2010-10-26 | 2010-10-26 | Based on the air traffic control atm information system security assessment method of ANN BP model |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102456158A true CN102456158A (en) | 2012-05-16 |
| CN102456158B CN102456158B (en) | 2015-11-11 |
Family
ID=46039329
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201010519836.4A Active CN102456158B (en) | 2010-10-26 | 2010-10-26 | Based on the air traffic control atm information system security assessment method of ANN BP model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102456158B (en) |
Cited By (24)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103226900A (en) * | 2013-03-21 | 2013-07-31 | 北京工业大学 | Space domain sector partitioning method based on weighted graph model |
| CN105095408A (en) * | 2015-07-09 | 2015-11-25 | 百度在线网络技术（北京）有限公司 | Method and apparatus for judging reliability of network expert |
| CN105118333A (en) * | 2015-09-30 | 2015-12-02 | 中国民用航空总局第二研究所 | Air traffic control analog simulation abnormality detection method and device based on multiple-regression model |
| CN105261241A (en) * | 2015-09-30 | 2016-01-20 | 中国民用航空总局第二研究所 | Analog simulation anomaly detection method and analog simulation anomaly detection device for air traffic control based on Hopfield neural network |
| CN105575186A (en) * | 2015-12-08 | 2016-05-11 | 成都民航空管科技发展有限公司 | Air traffic control automatic system secondary surveillance radar code directional allocation method |
| CN106027550A (en) * | 2016-06-29 | 2016-10-12 | 北京邮电大学 | Defense strategy system analysis method and device |
| CN106897474A (en) * | 2015-12-21 | 2017-06-27 | 中国航空工业集团公司西安飞机设计研究所 | A kind of aircraft cabin pressure control system control performance appraisal procedure |
| CN107430710A (en) * | 2014-11-10 | 2017-12-01 | 联邦快递公司 | Risk assessment framework |
| CN107729949A (en) * | 2017-11-03 | 2018-02-23 | 中国民航大学 | A kind of cluster and analysis method of control operation sub-health state |
| CN108133623A (en) * | 2018-01-31 | 2018-06-08 | 中国民航大学 | A kind of overhead crossing point graded index method for building up |
| CN108459582A (en) * | 2018-03-01 | 2018-08-28 | 中国航空无线电电子研究所 | Comprehensive health assessment method towards IMA systems |
| CN109409587A (en) * | 2018-10-09 | 2019-03-01 | 南京航空航天大学 | A kind of airport excavated based on weather data is into traffic flow forecasting method of leaving the theatre |
| CN109495296A (en) * | 2018-11-02 | 2019-03-19 | 国网四川省电力公司电力科学研究院 | Intelligent substation communication network state evaluation method based on clustering and neural network |
| CN109544999A (en) * | 2019-01-14 | 2019-03-29 | 中国民航大学 | A kind of air traffic networks method for evaluating reliability based on cloud model |
| CN110689032A (en) * | 2018-07-04 | 2020-01-14 | 北京京东尚科信息技术有限公司 | Data processing method and system, computer system and computer readable storage medium |
| CN110928752A (en) * | 2019-11-14 | 2020-03-27 | 青岛民航空管实业发展有限公司 | Method, device and equipment for evaluating health degree of air traffic control station |
| CN111091266A (en) * | 2019-11-12 | 2020-05-01 | 青岛民航空管实业发展有限公司 | Grade determination method and device of empty management station and storage medium |
| CN111610517A (en) * | 2020-06-09 | 2020-09-01 | 电子科技大学 | Secondary radar signal processing method based on deep four-channel network |
| CN111610518A (en) * | 2020-06-09 | 2020-09-01 | 电子科技大学 | Secondary radar signal denoising method based on depth residual separation convolutional network |
| CN111785095A (en) * | 2020-07-31 | 2020-10-16 | 北京航空航天大学 | Monitoring performance index evaluation method |
| CN112069740A (en) * | 2020-09-21 | 2020-12-11 | 吴学松 | Rock burst danger assessment method and platform |
| CN113256143A (en) * | 2021-06-07 | 2021-08-13 | 中国传媒大学 | Assessment method for media-melting organization, electronic equipment and readable storage medium |
| CN113850350A (en) * | 2021-11-30 | 2021-12-28 | 中哲国际工程设计有限公司 | Urban building land intelligent planning system and method |
| CN114818396A (en) * | 2022-06-29 | 2022-07-29 | 湖南大佳数据科技有限公司 | Network security shooting range system and drilling method for satellite navigation system |
-
2010
- 2010-10-26 CN CN201010519836.4A patent/CN102456158B/en active Active
Non-Patent Citations (5)
| Title |
|---|
| I. S. SAEH, A. KHAIRUDDIN: "《Power and Energy Conference, 2008, PECon 2008. IEEE 2nd International》", 3 December 2008 * |
| 中国民用航空总局: "《民用航空空中交通通信导航监视设备使用》", 1 August 2002 * |
| 王浩锋: "基于BP神经网络的民用航空航段安全风险评估", 《信息与电子工程》 * |
| 王起全: "航空企业基于SHEL模型的神经网络安全评价研究", 《中国安全科学学报》 * |
| 黄丽: "BP神经网络算法改进及应用研究", 《中国优秀硕士学位论文全文数据库信息科技辑》 * |
Cited By (39)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103226900B (en) * | 2013-03-21 | 2015-10-28 | 北京工业大学 | A kind of space domain sector division methods based on weighted graph model |
| CN103226900A (en) * | 2013-03-21 | 2013-07-31 | 北京工业大学 | Space domain sector partitioning method based on weighted graph model |
| US11897625B2 (en) | 2014-11-10 | 2024-02-13 | Federal Express Corporation | Risk assessment framework |
| US10647443B2 (en) | 2014-11-10 | 2020-05-12 | Federal Express Corporation | Risk assessment framework |
| CN107430710B (en) * | 2014-11-10 | 2021-02-05 | 联邦快递公司 | Risk assessment architecture |
| CN107430710A (en) * | 2014-11-10 | 2017-12-01 | 联邦快递公司 | Risk assessment framework |
| US11299290B2 (en) | 2014-11-10 | 2022-04-12 | Federal Express Corporation | Risk assessment framework |
| US11338934B2 (en) | 2014-11-10 | 2022-05-24 | Federal Express Corporation | Risk assessment framework |
| CN105095408A (en) * | 2015-07-09 | 2015-11-25 | 百度在线网络技术（北京）有限公司 | Method and apparatus for judging reliability of network expert |
| CN105118333A (en) * | 2015-09-30 | 2015-12-02 | 中国民用航空总局第二研究所 | Air traffic control analog simulation abnormality detection method and device based on multiple-regression model |
| CN105261241A (en) * | 2015-09-30 | 2016-01-20 | 中国民用航空总局第二研究所 | Analog simulation anomaly detection method and analog simulation anomaly detection device for air traffic control based on Hopfield neural network |
| CN105261241B (en) * | 2015-09-30 | 2017-11-07 | 中国民用航空总局第二研究所 | Air traffic control analog simulation method for detecting abnormality and device based on Hopfield neural network |
| CN105575186A (en) * | 2015-12-08 | 2016-05-11 | 成都民航空管科技发展有限公司 | Air traffic control automatic system secondary surveillance radar code directional allocation method |
| CN105575186B (en) * | 2015-12-08 | 2018-10-19 | 成都民航空管科技发展有限公司 | Air traffic control automation system secondary radar answering machine encodes fixed direction allocation method |
| CN106897474A (en) * | 2015-12-21 | 2017-06-27 | 中国航空工业集团公司西安飞机设计研究所 | A kind of aircraft cabin pressure control system control performance appraisal procedure |
| CN106027550B (en) * | 2016-06-29 | 2019-04-12 | 北京邮电大学 | A kind of defence policies systematic analytic method and device |
| CN106027550A (en) * | 2016-06-29 | 2016-10-12 | 北京邮电大学 | Defense strategy system analysis method and device |
| CN107729949A (en) * | 2017-11-03 | 2018-02-23 | 中国民航大学 | A kind of cluster and analysis method of control operation sub-health state |
| CN108133623A (en) * | 2018-01-31 | 2018-06-08 | 中国民航大学 | A kind of overhead crossing point graded index method for building up |
| CN108459582A (en) * | 2018-03-01 | 2018-08-28 | 中国航空无线电电子研究所 | Comprehensive health assessment method towards IMA systems |
| CN110689032A (en) * | 2018-07-04 | 2020-01-14 | 北京京东尚科信息技术有限公司 | Data processing method and system, computer system and computer readable storage medium |
| CN109409587A (en) * | 2018-10-09 | 2019-03-01 | 南京航空航天大学 | A kind of airport excavated based on weather data is into traffic flow forecasting method of leaving the theatre |
| CN109495296B (en) * | 2018-11-02 | 2022-05-13 | 国网四川省电力公司电力科学研究院 | Intelligent substation communication network state evaluation method based on clustering and neural network |
| CN109495296A (en) * | 2018-11-02 | 2019-03-19 | 国网四川省电力公司电力科学研究院 | Intelligent substation communication network state evaluation method based on clustering and neural network |
| CN109544999A (en) * | 2019-01-14 | 2019-03-29 | 中国民航大学 | A kind of air traffic networks method for evaluating reliability based on cloud model |
| CN111091266A (en) * | 2019-11-12 | 2020-05-01 | 青岛民航空管实业发展有限公司 | Grade determination method and device of empty management station and storage medium |
| CN110928752A (en) * | 2019-11-14 | 2020-03-27 | 青岛民航空管实业发展有限公司 | Method, device and equipment for evaluating health degree of air traffic control station |
| CN111610517B (en) * | 2020-06-09 | 2022-06-07 | 电子科技大学 | Secondary radar signal processing method based on deep four-channel network |
| CN111610517A (en) * | 2020-06-09 | 2020-09-01 | 电子科技大学 | Secondary radar signal processing method based on deep four-channel network |
| CN111610518A (en) * | 2020-06-09 | 2020-09-01 | 电子科技大学 | Secondary radar signal denoising method based on depth residual separation convolutional network |
| CN111610518B (en) * | 2020-06-09 | 2022-08-05 | 电子科技大学 | Secondary radar signal denoising method based on depth residual separation convolutional network |
| CN111785095A (en) * | 2020-07-31 | 2020-10-16 | 北京航空航天大学 | Monitoring performance index evaluation method |
| CN111785095B (en) * | 2020-07-31 | 2021-06-01 | 北京航空航天大学 | Method for forming monitoring performance evaluation index |
| CN112069740A (en) * | 2020-09-21 | 2020-12-11 | 吴学松 | Rock burst danger assessment method and platform |
| CN112069740B (en) * | 2020-09-21 | 2023-11-24 | 吴学松 | Rock burst risk assessment method and platform |
| CN113256143A (en) * | 2021-06-07 | 2021-08-13 | 中国传媒大学 | Assessment method for media-melting organization, electronic equipment and readable storage medium |
| CN113850350A (en) * | 2021-11-30 | 2021-12-28 | 中哲国际工程设计有限公司 | Urban building land intelligent planning system and method |
| CN114818396A (en) * | 2022-06-29 | 2022-07-29 | 湖南大佳数据科技有限公司 | Network security shooting range system and drilling method for satellite navigation system |
| CN114818396B (en) * | 2022-06-29 | 2022-09-20 | 湖南大佳数据科技有限公司 | Network security shooting range system and drilling method for satellite navigation system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102456158B (en) | 2015-11-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102456158B (en) | Based on the air traffic control atm information system security assessment method of ANN BP model | |
| Turskis et al. | A hybrid model based on fuzzy AHP and fuzzy WASPAS for construction site selection | |
| Chang et al. | A multi criteria group decision-making model for teacher evaluation in higher education based on cloud model and decision tree | |
| CN106411854A (en) | Network security risk assessment method based on fuzzy Bayes | |
| Xu et al. | A computerized risk evaluation model for public-private partnership (PPP) projects and its application | |
| CN107491694A (en) | Method for quantitative evaluation SCADA system information security fragility | |
| Liu et al. | A risk assessment method based on RBF artificial neural network-cloud model for urban water hazard | |
| Li et al. | An assessment method of operator’s situation awareness reliability based on fuzzy logic-AHP | |
| CN107239905A (en) | Onboard networks safety risk estimating method based on advanced AHP GCM | |
| Samani et al. | A fuzzy systematic approach to construction risk analysis | |
| Pessoa et al. | Cost forecasting of public construction projects using multilayer perceptron artificial neural networks: a case study | |
| Nosov et al. | DEVISING AN APPROACH TO SAFETY MANAGEMENT OF VESSEL CONTROL THROUGH THE IDENTIFICATION OF NAVIGATOR’S STATE. | |
| Zhou et al. | Safety assessment of high-risk operations in hydroelectric-project based on accidents analysis, SEM, and ANP | |
| Goodheart | Identification of causal paths and prediction of runway incursion risk by means of Bayesian belief networks | |
| Liu et al. | Systems Evaluation: Methods, Models, and Applications | |
| Gao et al. | Risk assessment of urban rail transit project using interpretative structural modelling: Evidence from China | |
| Shmelova et al. | Artificial intelligence methods and applications in aviation | |
| Ma et al. | Highway construction investment risk evaluation using BP neural network model | |
| Oztekin et al. | Hazard, safety risk, and uncertainty modeling of the integration of unmanned aircraft systems into the national airspace | |
| Shmelova et al. | Artificial neural network for pre-simulation training of air traffic controller | |
| Hrinchenko et al. | Qualimetric approaches to assessing sustainable development indicators | |
| Adnan et al. | Improvement of the method of estimation and forecasting of the state of the monitoring object in intelligent decision support systems | |
| Liao et al. | Probabilistic prediction of unsafe event in air traffic control Department based on the Improved backpropagation neural network | |
| Sohag et al. | Demand forecasting for a domestic airport-A case study | |
| Chen | Environmental applications of granular computing and intelligent systems |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20230505 Address after: Unit 516, Zone A, Comprehensive Building, No. 3 Huatian Road, Huayuan Industrial Zone, Tianjin Binhai High tech Zone, 300300 (settled in Innovation Service Company) Patentee after: TIANJIN YUN'AN TECHNOLOGY DEVELOPMENT Co.,Ltd. Address before: 300300 Tianjin city Dongli District North Road No. 2898 Patentee before: CIVIL AVIATION University OF CHINA |
|
| TR01 | Transfer of patent right |