CN112910567B - Interception classification monitoring method based on recurrent neural network and related equipment - Google Patents
Interception classification monitoring method based on recurrent neural network and related equipment Download PDFInfo
- Publication number
- CN112910567B CN112910567B CN202110114438.2A CN202110114438A CN112910567B CN 112910567 B CN112910567 B CN 112910567B CN 202110114438 A CN202110114438 A CN 202110114438A CN 112910567 B CN112910567 B CN 112910567B
- Authority
- CN
- China
- Prior art keywords
- eavesdropping
- parameters
- neural network
- analysis result
- optical fiber
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04B—TRANSMISSION
- H04B10/00—Transmission systems employing electromagnetic waves other than radio-waves, e.g. infrared, visible or ultraviolet light, or employing corpuscular radiation, e.g. quantum communication
- H04B10/80—Optical aspects relating to the use of optical transmission for specific applications, not provided for in groups H04B10/03 - H04B10/70, e.g. optical power feeding or optical transmission through water
- H04B10/85—Protection from unauthorised access, e.g. eavesdrop protection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Molecular Biology (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Computer Security & Cryptography (AREA)
- Electromagnetism (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Optical Communication System (AREA)
Abstract
One or more embodiments of the present specification provide a recurrent neural network-based eavesdropping classification monitoring method, including: collecting transmission signals transmitted on an optical fiber channel, and determining parameters of the transmission signals; the parameters comprise optical signal to noise ratio, chromatic dispersion, polarization mode dispersion, wavelength and bit error rate; preprocessing the parameters to obtain input values; analyzing and calculating the input value by using a pre-trained recurrent neural network to obtain an analysis result; according to the analysis result, performing intervention on eavesdropping behavior; and acquiring the real-time stress of the optical fiber, and determining the eavesdropping classification result according to the analysis result and the corresponding real-time stress of the optical fiber. The scheme disclosed by the invention is beneficial to eliminating the error of system change and reducing the occurrence of false detection, and a specific eavesdropping prevention method is formulated aiming at a specific eavesdropping mode, so that the normal transmission of signals can not be influenced while the safety of an optical communication system is improved.
Description
Technical Field
One or more embodiments of the present disclosure relate to the field of eavesdropping monitoring technologies, and in particular, to an eavesdropping classification monitoring method and related device based on a recurrent neural network.
Background
Optical fiber communication is widely used in many real-life scenarios due to the characteristics of large-capacity, high-rate transmission, and thus many lawless persons may try to obtain information transmitted in an optical fiber by various means to gain profits. The eavesdropping method can be divided into two categories, light leakage eavesdropping and coupling eavesdropping. The light leakage eavesdropping is to destroy the optical fiber and restore the original signal by using the leaked optical signal, and the method comprises an optical fiber bending method and a V-shaped groove notching method. The coupling eavesdropping is to couple the transmission fiber and the eavesdropping fiber by physical or chemical means, and split the light to obtain an optical signal, including a fused biconical taper method, a hydrofluoric acid etching method and an edge polishing method.
In the existing monitoring technology, the method comprises the following steps: the optical time domain reflectometer and the optical fiber eavesdropping monitoring system based on the support vector machine have certain limitation on the detection capability because the optical time domain reflectometer has an event blind area. For non-reflective events, i.e. for non-optical fibre pressure tapping, the loss in the channel after signal leakage also changes continuously, so that OTDR cannot detect this type of tapping. When the support vector machine is used for eavesdropping monitoring, because the support vector machine does not support time behavior, when an eavesdropper eavesdrops by using a coupling method, the loss changes slowly along with time, and a monitoring module may think that the eavesdropping is caused by instability of a system, and the situation of misjudgment may occur. Therefore, a more effective and accurate optical fiber eavesdropping monitoring method is urgently needed, and a specific eavesdropping prevention scheme is favorably formulated aiming at a specific eavesdropping mode.
Disclosure of Invention
In view of the above, one or more embodiments of the present disclosure are directed to an eavesdropping classification monitoring method based on a recurrent neural network, so as to solve the problem that in the prior art, an eavesdropping means cannot be identified, and thus a specific eavesdropping prevention method cannot be formulated for a specific eavesdropping manner.
In view of the above, one or more embodiments of the present disclosure provide a method and related device for monitoring eavesdropping classification based on a recurrent neural network, including:
collecting transmission signals transmitted on an optical fiber channel, and determining parameters of the transmission signals; the parameters comprise optical signal to noise ratio, chromatic dispersion, polarization mode chromatic dispersion, wavelength and bit error rate;
preprocessing the parameters to obtain input values;
analyzing and calculating the input value by using a pre-trained recurrent neural network to obtain an analysis result;
and acquiring the bending radius and the real-time stress of the optical fiber, and determining the eavesdropping classification result according to the analysis result and the corresponding bending radius or the real-time stress.
Optionally, the training process of the recurrent neural network includes:
acquiring parameters for training and parameters for testing;
determining an interception mode corresponding to the training parameters according to the training parameters;
constructing a training set according to the parameters for training and the corresponding eavesdropping mode;
according to the training set, carrying out model training by using a gradient descent method; and after the training is finished, performing mode testing by using the testing parameters to obtain the recurrent neural network.
Optionally, the analysis result includes: no wiretap, light leakage wiretap and coupling wiretap; after the analysis result is obtained, the method further comprises the following steps:
and interrupting the current communication when the analysis result is light leakage interception or coupling interception.
Optionally, no wiretapping, light leakage wiretapping and coupling wiretapping; the eavesdropping classification result is: bending and eavesdropping optical fibers, eavesdropping a V-shaped groove opening, eavesdropping by hydrofluoric acid corrosion and eavesdropping by melting and tapering;
the determining of the eavesdropping classification result according to the analysis result and the corresponding real-time stress of the optical fiber specifically comprises:
when the analysis result is the light leakage eavesdropping, acquiring the bending radius; if the bending radius is smaller than a preset threshold value, the eavesdropping classification result is the optical fiber bending eavesdropping, and if the real-time stress is larger than the preset threshold value, the eavesdropping is performed through a V-shaped notch;
and when the analysis result is the coupling eavesdropping, acquiring the real-time stress, if the real-time stress is 0, judging that the eavesdropping classification result is hydrofluoric acid corrosion eavesdropping, and otherwise, judging that the eavesdropping classification result is fused biconical eavesdropping.
Optionally, the digital signal is optimized to obtain a complete digital signal;
and carrying out symbol judgment and recoding on the complete digital signal to obtain complete information, and realizing the recovery of the intercepted part in the transmission information.
Optionally, the optimizing includes: dispersion compensation, carrier recovery and equalization with constant modulus algorithm
Based on the same inventive concept, one or more embodiments of the present specification further provide an eavesdropping classification monitoring apparatus based on a recurrent neural network, including:
the optical performance monitoring module is configured to collect a transmission signal transmitted on an optical fiber channel and determine a parameter of the transmission signal; the parameters comprise optical signal to noise ratio, chromatic dispersion, polarization mode chromatic dispersion, wavelength and bit error rate;
the preprocessing module is configured to preprocess the parameters to obtain input values;
the eavesdropping monitoring module is configured to analyze and calculate the input value by using a pre-trained recurrent neural network to obtain an analysis result;
and the monitoring result processing module is configured to acquire the bending radius and the real-time stress of the optical fiber and determine the eavesdropping classification result according to the analysis result and the corresponding bending radius or the real-time stress.
Based on the same inventive concept, one or more embodiments of the present specification further provide an electronic device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the recursive neural network-based eavesdropping classification monitoring method as described in any one of the above.
Based on the same inventive concept, one or more embodiments of the present specification further provide a non-transitory computer-readable storage medium storing computer instructions for causing the computer to perform any one of the above-described recursive neural network-based eavesdropping classification monitoring methods.
As can be seen from the above description, one or more embodiments of the present disclosure provide a recursive neural network-based eavesdropping classification monitoring method by collecting a transmission signal transmitted over a fiber channel and determining a parameter of the transmission signal; the parameters comprise optical signal to noise ratio, chromatic dispersion, polarization mode chromatic dispersion, wavelength and bit error rate; preprocessing the parameters to obtain input values; analyzing and calculating an input value by using a pre-trained recurrent neural network to obtain an analysis result; and accurately determining the eavesdropping classification result according to the analysis result and the acquired real-time stress of the optical fiber, and formulating an eavesdropping prevention method according to the classification result. The method can formulate specific anti-eavesdropping methods aiming at different eavesdropping modes so as to improve the safety of the communication system, and the monitoring and classifying method of the method does not influence the normal transmission of signals; the method is favorable for eliminating the error of system change and reducing the occurrence of false detection.
Drawings
In order to more clearly illustrate one or more embodiments or prior art solutions of the present specification, the drawings that are needed in the description of the embodiments or prior art will be briefly described below, and it is obvious that the drawings in the following description are only one or more embodiments of the present specification, and that other drawings may be obtained by those skilled in the art without inventive effort from these drawings.
FIG. 1 is a flow diagram of a recursive neural network based eavesdropping classification monitoring method according to one or more embodiments of the present disclosure;
FIG. 2 is a schematic diagram of a recursive neural network-based eavesdropping classification monitoring method according to one or more embodiments of the present disclosure;
FIG. 3 is a decoding recovery diagram in accordance with one or more embodiments of the present disclosure;
FIG. 4 is a schematic diagram of recurrent neural network training in accordance with one or more embodiments of the present disclosure;
FIG. 5 is a schematic representation of a recurrent neural network in accordance with one or more embodiments of the present disclosure;
FIG. 6 is a schematic diagram of an eavesdropping classification monitoring device based on a recurrent neural network according to one or more embodiments of the present disclosure;
fig. 7 is a schematic structural diagram of an electronic device according to one or more embodiments of the present disclosure.
Detailed Description
For the purpose of promoting a better understanding of the objects, aspects and advantages of the present disclosure, reference is made to the following detailed description taken in conjunction with the accompanying drawings.
It is to be noted that unless otherwise defined, technical or scientific terms used in one or more embodiments of the present specification should have the ordinary meaning as understood by those of ordinary skill in the art to which this disclosure belongs. The use of "first," "second," and similar terms in one or more embodiments of the specification is not intended to indicate any order, quantity, or importance, but rather is used to distinguish one element from another. The word "comprising" or "comprises", and the like, means that the element or item listed before the word covers the element or item listed after the word and its equivalents, but does not exclude other elements or items. The terms "connected" or "coupled" and the like are not restricted to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "upper", "lower", "left", "right", and the like are used merely to indicate relative positional relationships, and when the absolute position of the object being described is changed, the relative positional relationships may also be changed accordingly.
As described in the background section, optical fiber communication is widely used in many real-life scenes due to its transmission characteristics, such as high capacity, high speed, high electromagnetic interference resistance, and high radiation resistance, so that many lawless persons can obtain benefits by acquiring information transmitted in optical fibers by various means. Thus, there is a need for an eavesdropping monitoring method that accurately learns the eavesdropping classification.
However, in the prior art, for a non-reflection event, that is, a mode of eavesdropping on the pressure generated by the optical fiber is not performed, the loss in a channel after the signal leakage is continuously changed, and an optical time domain reflectometer cannot detect the eavesdropping mode of the type; because the support vector machine does not support time behavior, when eavesdropping is carried out by using a coupling method, the change of loss time is slow, and misjudgment may occur.
In view of the above-mentioned problems in the prior art and based on the findings of the applicant, one or more embodiments in this specification provide an eavesdropping classification monitoring method based on a recurrent neural network, which analyzes parameters extracted by a receiving end in combination with a stress sensor according to the characteristics of each eavesdropping method and the characteristic that the recurrent neural network has memorability, determines whether eavesdropping occurs or not by time-varying transmission signal parameters, and realizes eavesdropping monitoring and eavesdropping method determination in combination with feedback of a distributed optical fiber stress sensor received at this time, thereby facilitating an eavesdropping prevention method designed for different eavesdropping modes to accurately solve the eavesdropping problem, and thus improving the security of an optical communication system.
The technical solutions of one or more embodiments of the present specification are further described in detail below with reference to specific examples.
As shown in fig. 1, one or more embodiments in this specification provide an eavesdropping classification monitoring method based on a recurrent neural network, which specifically includes the following steps:
s101, collecting a transmission signal transmitted on an optical fiber channel, and determining a parameter of the transmission signal; the parameters include optical signal to noise ratio, dispersion, polarization mode dispersion, wavelength, and bit error rate.
In this embodiment, first, a receiving end receives a transmission signal amplified by an amplifier on an optical fiber channel of an optical fiber, where the optical fiber may be a communication optical fiber or a sensing optical fiber; the transmission signals are sent to an eavesdropping monitoring module, and the eavesdropping monitoring module sequentially carries out photoelectric detection on the transmission signals to obtain electric signals; then, carrying out analog-to-digital conversion on the obtained electric signal to obtain a digital signal; and finally, the optical performance monitoring module extracts parameters of the digital signal to obtain the parameters, wherein the parameters comprise an optical signal to noise ratio, chromatic dispersion, polarization mode chromatic dispersion, wavelength and a bit error rate. When the dispersion is serious, the front and the back of the optical pulse are mutually overlapped, so that intersymbol interference is caused, and the error rate is increased; the optical signal to noise ratio directly reflects the interference degree of the signal in the transmission process, and the higher the optical signal to noise ratio is, the lower the interference degree of the signal by noise is, and the fewer error codes are; polarization film dispersion can cause transmission delays in communication systems; wavelength is an important characteristic index of a wave, and is a measure of the properties of the wave, and when eavesdropping occurs, the wavelength changes rapidly in a short time, which affects the transmission of information. Therefore, osnr, dispersion, polarization mode dispersion, wavelength and bit error rate are important criteria for measuring the communication quality of the fiber optic communication system.
As shown in fig. 2, the transmission signal is transmitted by an optical transmitter, amplified by an optical fiber amplifier, and then reaches a receiving end, and then the optical signal as the transmission signal is subjected to photoelectric detection by a photoelectric detector and converted into an electrical signal, and then the electrical signal is subjected to analog-to-digital conversion to obtain a digital signal. The analog-to-digital conversion corresponds to digital-to-analog conversion, which is the inverse process of the analog-to-digital conversion, and the digital-to-analog conversion is used for converting a digital signal into an analog signal.
As shown in fig. 2, the transmission signal is transmitted by an optical transmitter, amplified by an optical fiber amplifier, and then reaches a receiving end, and then the optical signal as the transmission signal is subjected to photoelectric detection by a photoelectric detector, and is converted into an electrical signal, and then the electrical signal is subjected to analog-to-digital conversion, so as to obtain a digital signal. Extracting parameters according to the digital signals, transmitting the extracted parameters to a trained recurrent neural network, and judging whether eavesdropping occurs or not by the recurrent neural network according to the parameters; when the eavesdropping is found, the transmission is interrupted immediately, and corresponding parameters are further obtained to determine eavesdropping classification results. The analog-to-digital conversion corresponds to digital-to-analog conversion, which is the inverse process of the analog-to-digital conversion, and the digital-to-analog conversion is used for converting a digital signal into an analog signal.
As an alternative embodiment, as shown in fig. 3, the present application can accurately implement classification of eavesdropping modes. The system is also provided with a decoding recovery function, information of the digital signal is obtained before parameter extraction is carried out on the digital signal, another branch of digital signal is sequentially subjected to dispersion compensation, carrier recovery and constant modulus algorithm equalization to compensate information loss caused by eavesdropping in the transmission process, information recovery is realized, a complete digital signal is obtained, then the complete digital signal is decoded through symbol judgment and recoding to obtain complete information, and follow-up research is carried out according to the complete information. The eavesdropping is accurately classified, and meanwhile, the information transmission of the optical fiber communication system is guaranteed.
And S102, preprocessing the parameters to obtain input values.
In this embodiment, the snr, the dispersion, the polarization mode dispersion, the wavelength, and the bit error rate of the parameters are preprocessed to obtain an input value X, where X is [ X ═ X1，…，xt-1，xt，xt+1，…，xn]Wherein x is1，…，xt-1，xt，xt+1，…，xnIs 1 to n input vectors, and any one input vector contains five numerical values, including: the osnr, the dispersion, the polarization mode dispersion, the wavelength and the bit error rate are important criteria for measuring the optical fiber communication.
And S103, analyzing and calculating the input value by using a pre-trained recurrent neural network to obtain an analysis result.
In this embodiment, the recurrent neural network is generated by training an initial recurrent neural network model in advance.
Specifically, referring to fig. 4, the initial training of the recurrent neural network model specifically includes:
s401, acquiring parameters for training and parameters for testing;
s402, determining an eavesdropping mode corresponding to the training parameters according to the training parameters;
s403, constructing a training set according to the parameters for training and the corresponding eavesdropping mode;
s404, performing model training by using a gradient descent method according to the training set; and after the training is finished, performing mode testing by using the testing parameters to obtain the recurrent neural network.
In step S401, a large number of values of the snr, the chromatic dispersion, the polarization mode dispersion, the wavelength, and the bit error rate of the transmission signal are collected in advance as parameters under various conditions, including the case where the eavesdropping occurs and the case where the eavesdropping does not occur; and defining numerical similarities and differences of optical signal to noise ratio, chromatic dispersion, polarization mode chromatic dispersion, wavelength and error rate under the condition of occurrence of eavesdropping and no eavesdropping and different eavesdropping modes, preprocessing the parameters, converting the parameters into an encoding form which can be identified by a recurrent neural network model, dividing the obtained parameters into parameters for training and parameters for testing, taking 80% of the parameters as parameters for training, and taking 20% of the parameters as parameters for training. Wherein, the optical signal-to-noise ratio is the ratio of the optical signal power to the noise power within the optical effective bandwidth of 0.1 nm; dispersion: the phenomenon that the complex color light is decomposed into monochromatic light to form a spectrum is called dispersion, and the dispersion in the optical fiber is composed of material dispersion, waveguide dispersion, refractive index distribution dispersion and the like, and can cause distortion of a transmission signal; polarization mode dispersion: when two light pulses at the input end are split into two vertical polarized outgoing pulses and are transmitted at the same transmission speed, the refractive indexes in the X-axis direction and the Y-axis direction are different, so that polarization mode dispersion is caused; wavelength: the distance of the wave propagating in a vibration period is referred, and the waves with the same frequency propagate in different media at different speeds, so the wavelengths are different; the error rate is the ratio of the errors in the transmission to the total number of codes transmitted. The bit error rate in this application is the ratio of the bit errors in the transmitted signal to the total transmitted signal multiplied by 100%.
In step S404, training the training parameters based on time back propagation, performing a round of iteration by a gradient descent method, determining the parameters of the model when the loss function is small enough and the number of iteration rounds reaches a preset value, completing the model training, performing a mode test by using the test parameters, and finally obtaining the recurrent neural network.
In the embodiment, according to the characteristics of each interception method, whether interception occurs is analyzed by the change of the transmission signal along with time by utilizing the characteristic that the recurrent neural network has memorability, so that corresponding processing measures can be taken according to the analysis result; when light leakage eavesdropping or coupling eavesdropping is monitored, the current communication is interrupted to prevent illegal personnel from continuing illegal eavesdropping, and loss caused by eavesdropping is timely reduced.
As an alternative embodiment, the gradient descent method is also called a steepest descent method, and is an iterative method, and the main operation of each step is to solve a gradient vector of an objective function, and take the negative gradient of the current position as a search direction, and the closer to the target value, the smaller the step size is, the slower the descent speed is. Among them, the gradient descent method is divided into batch gradient descent and random gradient descent. The gradient descent method is commonly used for solving a minimum value in the algorithm, and is not the minimum value, and the more appropriate the preset threshold value is, the more accurate the final result value is; and when the iteration times reach a preset threshold value, outputting a final result.
In the present embodiment, the recurrent neural network is divided into an input layer, a hidden layer and an output layer, x, as shown in fig. 5tIs an n-dimensional vector at time t, and is also an input value, where X is [ X ] in the figure1，…，xt-1，xt，xt+1，…，xn]Wherein x is1，…，xt-1，xt，xt+1，…，xnThe input vectors are from 1 to n corresponding time, and each input vector comprises an optical signal to noise ratio, chromatic dispersion, polarization mode chromatic dispersion, wavelength and an error rate; h istRepresents the hidden state at the time t, which is the hidden state at the time immediately before the time t, ht+1A hidden state at a time next to the time t; y istOutput representing time t, yt-1Is the previous time of time t; the weight U is the weight from an input layer to a hidden layer, the original input of the user is abstracted to be used as the input of the hidden layer, the weight V is the weight from the hidden layer to the hidden layer, the memory and the scheduling memory of a network are controlled by the weight V, the weight W is the weight from the hidden layer to an output layer, the weight W is normalized through the data obtained by the hidden layer, and the result Y is finally output, wherein the value of Y is Y1、y2、y3Composition, y, when no eavesdropping action takes place1、y2、y3Corresponding to 1, 0 and 0 respectively, the output result Y has a value of 100, and Y is detected when light leakage eavesdropping occurs1、y2、y3Corresponding to 0, 1 and 0 respectively, the value of output result Y is 010, when coupling eavesdropping occurs, Y1、y2、y3Corresponding to 0, 1, respectively, the value of the output result Y is 001 at this time.
And S104, acquiring the bending radius and the real-time stress of the optical fiber, and determining the eavesdropping classification result according to the analysis result and the corresponding bending radius or the real-time stress.
In this embodiment, the analysis result includes: no wiretap, light leakage wiretap and coupling wiretap; eavesdropping classification results: bending and eavesdropping optical fibers, eavesdropping a V-shaped groove opening, eavesdropping by hydrofluoric acid corrosion and eavesdropping by melting and tapering; the step of obtaining the corresponding real-time stress or bending radius according to the analysis result and determining the eavesdropping classification result comprises the following steps:
the determining of the eavesdropping classification result according to the analysis result and the corresponding real-time stress of the optical fiber specifically comprises:
when the analysis result is the light leakage eavesdropping, acquiring the bending radius; if the bending radius is smaller than a preset threshold value, the eavesdropping classification result is the optical fiber bending eavesdropping, and if the real-time stress is larger than the preset threshold value, the eavesdropping is performed through a V-shaped notch;
and when the analysis result is the coupling eavesdropping, acquiring the real-time stress, if the real-time stress is 0, judging that the eavesdropping classification result is hydrofluoric acid corrosion eavesdropping, and otherwise, judging that the eavesdropping classification result is fused biconical eavesdropping.
As an alternative embodiment, when no eavesdropping behavior occurs, the recurrent neural network may output an instruction "100" to the eavesdropping monitoring module to indicate that no eavesdropping occurs, and at this time, the monitoring result processing module does not respond, and the communication system operates normally; when light leakage eavesdropping occurs, the recurrent neural network outputs an instruction '010' to the eavesdropping monitoring module, the monitoring result processing module starts an alarm in time and interrupts communication, meanwhile, the bending radius of the optical fiber is called, if the bending radius is smaller than a preset threshold value, the optical fiber is subjected to bending eavesdropping, and if the real-time stress is larger than the preset threshold value, the optical fiber is subjected to eavesdropping in a V-shaped notch; wherein the preset threshold is set to 5 mm.
As an alternative embodiment, the preset threshold of the bending radius may be adjustable, specifically determined according to the thickness of the optical fiber and the thickness of the outer cladding, under the condition that the magnitude of the acting force is not changed, the magnitude of the radius of the optical fiber and the thickness of the outer cladding of the optical fiber are different, resulting in different bending radii, when the radius of the optical fiber is larger or the outer cladding of the optical fiber is thicker, the formed bending radius may become smaller by using the same force, and at this time, the preset threshold may be greater than 5mm, and may be set to be 7mm, 8mm, or 9 mm; the bend radius may be formed to be larger when the radius of the optical fiber is smaller or the outer cladding of the optical fiber is thinner, and the predetermined threshold may be less than 5mm, and may be set to be 4mm, 3mm, or 2 mm.
As an alternative embodiment, the optical fiber bending method is to remove the coating layer and the cladding of the optical fiber to leave only the bare fiber, then bend the bare fiber properly, leak out the light in the fiber core without satisfying the total reflection condition, and recover the original signal by 100% as long as the leaked light power reaches 1% of the light power of the original signal. The V-shaped groove cutting method is to perform V-shaped cutting on the cladding to the fiber core, so that the refractive index around the fiber core is changed, and signal leakage occurs at the V-shaped groove. FBT (fused biconical taper) means that two (or more) optical fibers from which coating layers are removed are brought together in a certain method, melted under high-temperature heating and simultaneously drawn to both sides, and finally a special waveguide structure in the form of a biconical body is formed in a heating zone, and different splitting ratios can be obtained by controlling the twisting angle and the drawing length of the optical fibers. The hydrofluoric acid etching method uses hydrofluoric acid as an etchant to dissolve a coating layer and a cladding of the optical fiber for coupling.
As an alternative example, prior art optical fibers have been subjected to a different range of forces when using fiber bending and V-notch eavesdropping. The fusion tapering method and the chemical erosion method are different in that the evanescent field refractive index profiles of the two leakage optical fibers are different, the influence of the cladding on the loss of the optical power is different, and the optical power loss is larger when the cladding is thinner. The influence of leaky and coupled eavesdropping on the transmitted optical signal is also different, the former causing abrupt changes in the power of the optical signal and the latter affecting the optical signal relatively slowly. Therefore, if the eavesdropping is identified by an instrument or an artificial intelligence algorithm without time characteristics, the eavesdropping behavior may not be monitored because the coupled eavesdropping method has a slow influence on the signal, thereby generating a misjudgment. Meanwhile, even if the existence of the eavesdropping is known, the eavesdropping can continue to occur because accurate protection cannot be made.
In the application, the characteristic of each eavesdropping method is utilized, the characteristic that the recurrent neural network has memorability is utilized, whether eavesdropping occurs or not is analyzed through the change of transmission signals along with time, parameters extracted by a receiving end are analyzed in combination with a stress sensor, and the eavesdropping monitoring and the eavesdropping method judgment are realized, so that a detection result processing module can take corresponding processing measures according to an analysis result, the eavesdropping prevention method designed aiming at different eavesdropping modes is favorable for accurately solving the eavesdropping problem, and the safety of an optical communication system is improved.
It should be noted that the method of one or more embodiments of the present disclosure may be performed by a single device, such as a computer or server. The method of the embodiment can also be applied to a distributed scene and completed by the mutual cooperation of a plurality of devices. In such a distributed scenario, one of the devices may perform only one or more steps of the method of one or more embodiments of the present disclosure, and the devices may interact with each other to complete the method.
It should be noted that the above description describes certain embodiments of this disclosure. Other embodiments are within the scope of the following claims. In some cases, the actions or steps recited in the claims may be performed in a different order than in the embodiments and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing may also be possible or may be advantageous.
Based on the same inventive concept, corresponding to any embodiment method, one or more embodiments of the present specification further provide an eavesdropping classification monitoring device based on a recurrent neural network.
Referring to fig. 6, the apparatus for monitoring eavesdropping classification based on recurrent neural network includes:
the optical performance monitoring module is configured to collect a transmission signal transmitted on an optical fiber channel and determine a parameter of the transmission signal; the parameters comprise optical signal to noise ratio, chromatic dispersion, polarization mode chromatic dispersion, wavelength and bit error rate;
the preprocessing module is configured to preprocess the parameters to obtain input values;
the eavesdropping monitoring module is configured to analyze and calculate the input value by using a pre-trained recurrent neural network to obtain an analysis result;
and the monitoring result processing module is configured to acquire the bending radius and the real-time stress of the optical fiber and determine the eavesdropping classification result according to the analysis result and the corresponding bending radius or the real-time stress.
For convenience of description, the above devices are described as being divided into various modules by functions, and are described separately. Of course, the functionality of the modules may be implemented in the same one or more software and/or hardware implementations in implementing one or more embodiments of the present description.
The apparatus of the foregoing embodiment is used to implement the corresponding method for monitoring eavesdropping classification based on a recurrent neural network in any of the foregoing embodiments, and has the beneficial effects of the corresponding method embodiment, which are not described herein again.
Based on the same inventive concept, corresponding to the method of any embodiment, one or more embodiments of the present specification further provide an electronic device, which includes a memory, a processor, and a computer program stored in the memory and running on the processor, and when the processor executes the computer program, the method for monitoring eavesdropping classification based on a recurrent neural network according to any embodiment is implemented.
Fig. 7 is a schematic diagram illustrating a more specific hardware structure of an electronic device according to this embodiment, where the device may include: a processor 1010, a memory 1020, an input/output interface 1030, a communication interface 1040, and a bus 1050. Wherein the processor 1010, memory 1020, input/output interface 1030, and communication interface 1040 are communicatively coupled to each other within the device via bus 1050.
The processor 1010 may be implemented by a general-purpose CPU (Central Processing Unit), a microprocessor, an Application Specific Integrated Circuit (ASIC), or one or more Integrated circuits, and is configured to execute related programs to implement the technical solutions provided in the embodiments of the present disclosure.
The Memory 1020 may be implemented in the form of a ROM (Read Only Memory), a RAM (Random Access Memory), a static Memory device, a dynamic Memory device, or the like. The memory 1020 may store an operating system and other application programs, and when the technical solution provided by the embodiments of the present specification is implemented by software or firmware, the relevant program codes are stored in the memory 1020 and called to be executed by the processor 1010.
The input/output interface 1030 is used for connecting an input/output module to input and output information. The input/output module may be configured as a component in a device (not shown) or may be external to the device to provide a corresponding function. The input devices may include a keyboard, a mouse, a touch screen, a microphone, various sensors, etc., and the output devices may include a display, a speaker, a vibrator, an indicator light, etc.
The communication interface 1040 is used for connecting a communication module (not shown in the drawings) to implement communication interaction between the present device and other devices. The communication module can realize communication in a wired mode (for example, USB, network cable, etc.), and can also realize communication in a wireless mode (for example, mobile network, WIFI, bluetooth, etc.).
It should be noted that although the above-mentioned device only shows the processor 1010, the memory 1020, the input/output interface 1030, the communication interface 1040 and the bus 1050, in a specific implementation, the device may also include other components necessary for normal operation. In addition, those skilled in the art will appreciate that the above-described apparatus may also include only those components necessary to implement the embodiments of the present description, and not necessarily all of the components shown in the figures.
The electronic device of the foregoing embodiment is used to implement the corresponding method for monitoring eavesdropping classification based on a recurrent neural network in any of the foregoing embodiments, and has the beneficial effects of the corresponding method embodiment, which are not described herein again.
Based on the same inventive concept, corresponding to any of the above-described embodiment methods, one or more embodiments of the present specification further provide a non-transitory computer-readable storage medium storing computer instructions for causing the computer to perform the recursive neural network-based eavesdropping classification monitoring method according to any of the above-described embodiments.
Computer-readable media of the present embodiments, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), Read Only Memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), Digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device.
The storage medium of the above embodiment stores computer instructions for causing the computer to execute the interception classification monitoring method based on the recurrent neural network according to any of the above embodiments, and has the beneficial effects of the corresponding method embodiments, which are not described herein again.
Those of ordinary skill in the art will understand that: the discussion of any embodiment above is meant to be exemplary only, and is not intended to intimate that the scope of the disclosure, including the claims, is limited to these examples; within the spirit of the present disclosure, features from the above embodiments or from different embodiments may also be combined, steps may be implemented in any order, and there are many other variations of different aspects of one or more embodiments of the present description as described above, which are not provided in detail for the sake of brevity.
In addition, well-known power/ground connections to Integrated Circuit (IC) chips and other components may or may not be shown in the provided figures, for simplicity of illustration and discussion, and so as not to obscure one or more embodiments of the disclosure. Furthermore, devices may be shown in block diagram form in order to avoid obscuring the understanding of one or more embodiments of the present description, and this also takes into account the fact that specifics with respect to implementation of such block diagram devices are highly dependent upon the platform within which the one or more embodiments of the present description are to be implemented (i.e., specifics should be well within purview of one skilled in the art). Where specific details (e.g., circuits) are set forth in order to describe example embodiments of the disclosure, it should be apparent to one skilled in the art that one or more embodiments of the disclosure can be practiced without, or with variation of, these specific details. Accordingly, the description is to be regarded as illustrative instead of restrictive.
While the present disclosure has been described in conjunction with specific embodiments thereof, many alternatives, modifications, and variations of these embodiments will be apparent to those of ordinary skill in the art in light of the foregoing description. For example, other memory architectures (e.g., dynamic ram (dram)) may use the discussed embodiments.
It is intended that the one or more embodiments of the present specification embrace all such alternatives, modifications and variations as fall within the broad scope of the appended claims. Therefore, any omissions, modifications, substitutions, improvements, and the like that may be made without departing from the spirit and principles of one or more embodiments of the present disclosure are intended to be included within the scope of the present disclosure.
Claims (10)
1. An eavesdropping classification monitoring method based on a recurrent neural network is characterized by comprising the following steps:
collecting transmission signals transmitted on an optical fiber channel, and determining parameters of the transmission signals; the parameters comprise optical signal to noise ratio, chromatic dispersion, polarization mode chromatic dispersion, wavelength and bit error rate;
preprocessing the parameters to obtain input values;
analyzing and calculating the input value by using a pre-trained recurrent neural network to obtain an analysis result; wherein the analysis result comprises: no wiretap, light leakage wiretap and coupling wiretap;
and acquiring the bending radius and the real-time stress of the optical fiber, acquiring the corresponding bending radius or the real-time stress according to the analysis result, and determining the eavesdropping classification result.
2. The method of claim 1, wherein the training process of the recurrent neural network comprises:
acquiring parameters for training and parameters for testing;
determining an eavesdropping mode corresponding to the training parameters according to the training parameters;
constructing a training set according to the parameters for training and the corresponding eavesdropping mode;
according to the training set, carrying out model training by using a gradient descent method; and after the training is finished, performing mode testing by using the testing parameters to obtain the recurrent neural network.
3. The method according to claim 1, wherein the collecting the transmission signal of the optical fiber channel and determining the parameter of the transmission signal specifically comprises:
carrying out photoelectric detection on the transmission signal to obtain an electric signal;
performing analog-to-digital conversion on the electric signal to obtain a digital signal;
and extracting parameters of the digital signals to obtain the parameters.
4. The method of claim 1, wherein the analysis results comprise: no wiretap, light leakage wiretap and coupling wiretap; after the analysis result is obtained, the method further comprises the following steps:
and interrupting the current communication when the analysis result is light leakage wiretapping or coupling wiretapping.
5. The method according to claim 1, wherein said eavesdropping classification result comprises: bending and eavesdropping optical fibers, eavesdropping a V-shaped groove opening, eavesdropping by hydrofluoric acid corrosion and eavesdropping by melting and tapering;
the obtaining the corresponding bending radius or the real-time stress according to the analysis result and determining the eavesdropping classification result specifically comprise:
when the analysis result is the light leakage eavesdropping, acquiring the bending radius; if the bending radius is smaller than a preset threshold value, the eavesdropping classification result is the optical fiber bending eavesdropping, and if the real-time stress is larger than the preset threshold value, the eavesdropping is performed through a V-shaped notch;
and when the analysis result is the coupling eavesdropping, acquiring the real-time stress, if the real-time stress is 0, judging that the eavesdropping classification result is hydrofluoric acid corrosion eavesdropping, and otherwise, judging that the eavesdropping classification result is fused biconical eavesdropping.
6. The method of claim 3, further comprising:
optimizing the digital signal to obtain a complete digital signal;
and carrying out symbol judgment and recoding on the complete digital signal to obtain complete information.
7. The method of claim 6, wherein the optimizing comprises: dispersion compensation, carrier recovery and constant modulus algorithm equalization.
8. An eavesdropping classification monitoring device based on a recurrent neural network, comprising:
the optical performance monitoring module is configured to collect a transmission signal transmitted on an optical fiber channel and determine a parameter of the transmission signal; the parameters comprise optical signal to noise ratio, chromatic dispersion, polarization mode chromatic dispersion, wavelength and bit error rate;
the preprocessing module is configured to preprocess the parameters to obtain input values;
the eavesdropping monitoring module is configured to analyze and calculate the input value by using a pre-trained recurrent neural network to obtain an analysis result; wherein the analysis result comprises: no wiretap, light leakage wiretap and coupling wiretap;
and the monitoring result processing module is configured to acquire the bending radius and the real-time stress of the optical fiber and determine the eavesdropping classification result according to the analysis result and the corresponding bending radius or the real-time stress.
9. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the method according to any of claims 1 to 7 when executing the program.
10. A non-transitory computer readable storage medium storing computer instructions for causing a computer to perform the method of any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110114438.2A CN112910567B (en) | 2021-01-25 | 2021-01-25 | Interception classification monitoring method based on recurrent neural network and related equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110114438.2A CN112910567B (en) | 2021-01-25 | 2021-01-25 | Interception classification monitoring method based on recurrent neural network and related equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112910567A CN112910567A (en) | 2021-06-04 |
| CN112910567B true CN112910567B (en) | 2022-07-01 |
Family
ID=76120687
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110114438.2A Active CN112910567B (en) | 2021-01-25 | 2021-01-25 | Interception classification monitoring method based on recurrent neural network and related equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112910567B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113992277B (en) * | 2021-10-22 | 2023-09-22 | 安天科技集团股份有限公司 | Method, system, equipment and medium for detecting data transmission in optical signal |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107132478A (en) * | 2017-04-13 | 2017-09-05 | 南京师范大学 | A kind of wind-driven generator method for diagnosing faults measured based on polarization maintaining optical fibre with neural network classification |
| CN108089962A (en) * | 2017-11-13 | 2018-05-29 | 北京奇艺世纪科技有限公司 | A kind of method for detecting abnormality, device and electronic equipment |
| CN108596229A (en) * | 2018-04-13 | 2018-09-28 | 北京华电智慧科技产业有限公司 | Online abnormal monitoring, diagnosing method and system |
| WO2019164906A1 (en) * | 2018-02-22 | 2019-08-29 | Subcom, Llc | Fault detection and reporting in line monitoring systems |
| CN110518973A (en) * | 2019-07-26 | 2019-11-29 | 北京光锁科技有限公司 | A kind of optical fiber eavesdropping monitoring method and relevant device based on machine learning |
| CN111008648A (en) * | 2019-11-12 | 2020-04-14 | 国网湖南省电力有限公司 | Fault identification method, system and medium for pure optical fiber electronic current transformer |
| CN111178456A (en) * | 2020-01-15 | 2020-05-19 | 腾讯科技（深圳）有限公司 | Abnormal index detection method and device, computer equipment and storage medium |
Family Cites Families (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8494313B2 (en) * | 2009-04-06 | 2013-07-23 | Rockstar Consortium Us Lp | Monitoring eDC polarization inverse filter coefficients to identify real-time physical intrusion into a core or metro optical network |
| RU2428798C2 (en) * | 2009-09-14 | 2011-09-10 | Владимир Васильевич Гришачев | Fibre optic detector of voice information leakage threats via fibre optic communications |
| EP3094038B1 (en) * | 2015-05-12 | 2019-08-07 | ID Quantique S.A. | Apparatus and method for providing eavesdropping detection of an optical fiber communication |
| CN110325929B (en) * | 2016-12-07 | 2021-05-25 | 阿瑞路资讯安全科技股份有限公司 | System and method for signal waveform analysis for detecting changes in wired networks |
| US10187171B2 (en) * | 2017-03-07 | 2019-01-22 | The United States Of America, As Represented By The Secretary Of The Navy | Method for free space optical communication utilizing patterned light and convolutional neural networks |
| GB201720063D0 (en) * | 2017-12-01 | 2018-01-17 | Cambridge Entpr Ltd | Quantum security systems |
| CN110492926B (en) * | 2019-07-26 | 2021-01-05 | 北京光锁科技有限公司 | Optical fiber eavesdropping positioning method and system based on machine learning and electronic equipment |
| CN111786935B (en) * | 2019-11-08 | 2022-03-01 | 国网辽宁省电力有限公司电力科学研究院 | Service flow abnormity detection method for optical cable fiber core remote intelligent scheduling exchange |
| CN111262803B (en) * | 2020-03-04 | 2022-08-26 | 广州番禺职业技术学院 | Physical layer secure communication method, device and system based on deep learning |
-
2021
- 2021-01-25 CN CN202110114438.2A patent/CN112910567B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107132478A (en) * | 2017-04-13 | 2017-09-05 | 南京师范大学 | A kind of wind-driven generator method for diagnosing faults measured based on polarization maintaining optical fibre with neural network classification |
| CN108089962A (en) * | 2017-11-13 | 2018-05-29 | 北京奇艺世纪科技有限公司 | A kind of method for detecting abnormality, device and electronic equipment |
| WO2019164906A1 (en) * | 2018-02-22 | 2019-08-29 | Subcom, Llc | Fault detection and reporting in line monitoring systems |
| CN108596229A (en) * | 2018-04-13 | 2018-09-28 | 北京华电智慧科技产业有限公司 | Online abnormal monitoring, diagnosing method and system |
| CN110518973A (en) * | 2019-07-26 | 2019-11-29 | 北京光锁科技有限公司 | A kind of optical fiber eavesdropping monitoring method and relevant device based on machine learning |
| CN111008648A (en) * | 2019-11-12 | 2020-04-14 | 国网湖南省电力有限公司 | Fault identification method, system and medium for pure optical fiber electronic current transformer |
| CN111178456A (en) * | 2020-01-15 | 2020-05-19 | 腾讯科技（深圳）有限公司 | Abnormal index detection method and device, computer equipment and storage medium |
Non-Patent Citations (2)
| Title |
|---|
| Experimental Demonstration of Optical Fiber Eavesdropping Detection Based on Deep Learning;Mingzhe Liu等;《2019 Asia Communications and Photonics Conference (ACP)》;20200210;全文 * |
| 基于时/频域综合特征提取的分布式光纤入侵监测系统事件识别方法;彭宽等;《光学学报》;20190610;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112910567A (en) | 2021-06-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110492926B (en) | Optical fiber eavesdropping positioning method and system based on machine learning and electronic equipment | |
| CN111442827B (en) | Optical fiber passive online monitoring system for transformer winding vibration | |
| CN106225907B (en) | It is a kind of based on Φ-OTDR technique fiber-optic vibration identifying system and method | |
| CN102879081B (en) | A kind of data processing method in distributed optical fiber vibration system | |
| CN110995339A (en) | Method for extracting and identifying time-space information of distributed optical fiber sensing signal | |
| CN111649817A (en) | Distributed optical fiber vibration sensor system and mode identification method thereof | |
| CN112910567B (en) | Interception classification monitoring method based on recurrent neural network and related equipment | |
| CN104217513A (en) | Method for improving intrusion event identification accuracy based on phase-sensitive optical time domain reflectrometer (phi-OTDR) | |
| EP4135218A1 (en) | Long-distance optical fiber measurement method, apparatus, device and system, and storage medium | |
| Wang et al. | Rapid response DAS denoising method based on deep learning | |
| Lyu et al. | Identification of intrusion events based on distributed optical fiber sensing in complex environment | |
| CN110987038B (en) | Distributed optical fiber disturbance detection device | |
| CN113092879A (en) | Transmission line lightning stroke monitoring method, device, equipment and storage medium | |
| CN110518973B (en) | Optical fiber eavesdropping monitoring method based on machine learning and related equipment | |
| CN110595599B (en) | Method for reducing polarization fading of optical fiber vibration system and detection system applying same | |
| Wu et al. | Multi-source separation under two “blind” conditions for fiber-optic distributed acoustic sensor | |
| Fang et al. | Distributed optical fiber vibration sensing implemented with delayed feedback reservoir computing | |
| Todorović et al. | FSO system performance analysis based on novel Gamma–Chi-square irradiance PDF model | |
| Šlapák et al. | Monitoring of fibre optic links with a machine learning-assisted low-cost polarimeter | |
| Tangudu et al. | Dynamic range enhancement of OTDR using lifting wavelet transform‐modified particle swarm optimisation scheme | |
| CN113124931B (en) | Method and device for improving power optical fiber state monitoring precision and storage medium | |
| CN105551165B (en) | Optical fiber perimeter safety-protection system based on dynamic threshold detection disturbs determination methods | |
| Chen et al. | Combined PMD-PDL effects on BERs in simplified optical systems: an analytical approach | |
| CN205451354U (en) | Optic fibre intelligence anti -theft device | |
| Yuan et al. | Polarization mode dispersion measurement using fixed analyzer method with deep neural network |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |