Satellite identity legitimacy verification method and device of a giant constellation system

Abstract

The application discloses a satellite identity legitimacy verification method and device of a mega constellation system, and relates to the field of satellite communication. The method comprises the following steps: acquiring a satellite data set; the satellite data comprises one or more of satellite signal time difference of arrival (TDoA), Doppler frequency shift (DoDFS) and signal average received power difference (DoRP), and corresponding identity legitimacy labels; the labels comprise legitimate, illegal and attack; dividing the satellite data set into a training set and a test set; constructing an identity legitimacy verification model based on CNN-LSTM deep learning; training the identity legitimacy verification model based on the training set; verifying and classifying the legitimacy of the satellites in the mega constellation system based on the trained identity legitimacy verification model; evaluating the accuracy of the trained identity legitimacy verification model, and if the evaluation result is lower than a preset threshold, retraining the identity legitimacy verification model until convergence. The application can realize high-precision identification of satellite identity legitimacy.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of satellite communication, and particularly relates to a satellite identity legitimacy verification method and device of a mega-constellation system. BACKGROUND

[0002] As an important component of the 6th Generation Mobile Communication System (6G), the scale of satellite communication network is expanding, and the number of mega-constellation satellite systems deployed worldwide is rapidly increasing. However, the rapid increase in the number of satellites also brings new security challenges. Illegal access and spoofing attacks not only affect the normal communication services of the mega-constellation system, but also may pose a serious threat to ground users and other spacecraft. On the other hand, as an effective extension and important supplement to ground networks, mega-constellation systems provide global users with large-capacity, high-coverage, and low-latency continuous communication services in all space and time. With the rapid evolution of navigation and remote sensing satellite technologies, satellite communication application scenarios continue to expand and gradually occupy a core position in global information infrastructure, becoming an increasingly important force in national strategic security. The United States' "Starlink", Europe's "IRIS", China's "Hongmu" and "Xiaohongmu" have all entered the deployment stage, and the number of satellites in orbit is growing rapidly. The number of satellites in orbit has exceeded 2,000, and the number of mega-constellation satellites has exceeded 1,000. The number of satellites in orbit is expected to reach 10,000 by 2027. The number of satellites in orbit is expected to reach 10,000 by 2027. 2Both the satellite communication system and China's "Thousand Sails" program aim to build a large-scale, independently controllable low-Earth orbit (LEO) constellation system to safeguard the nation's leading position in global communications and information security. However, satellite communications typically employ high-power, large-beam antenna designs to achieve wide-area coverage, but this design results in a signal leakage range far exceeding that of ground base stations, creating a wide-area radio frequency (RF) attack surface. Attackers do not need to approach the satellite itself; they can intercept, interfere with, forge, or tamper with the signal within the coverage area using ground equipment or drones. The downlink of satellite communication systems is particularly vulnerable to eavesdropping, interception, and even replay. Traditional terrestrial communication networks can rely on national borders to deploy firewalls and border gateways for protection. However, due to the orbital dynamics of satellites, physical protection boundaries are completely lost. Furthermore, the coverage area of ​​satellite signals spans multiple national sovereign territories and the high seas, making the communication links inherently uncontrollable. The complexity of control has increased significantly, making it difficult to effectively isolate malicious nodes or limit potential attack traffic. The blurring of network boundaries forces satellite communication systems to confront more complex global dynamic attack surface management issues. With countries deploying numerous low-Earth orbit satellite constellations, orbital resources are becoming increasingly scarce. Data link interference, signal spoofing, and frequency contention can occur between giant satellite systems or even between individual satellites; satellites operating in adjacent or co-orbital orbits may cause collisions, signal interference, and resource contention. This is particularly true for military satellites with strategic functions, significantly increasing the risk of network warfare. Therefore, there is an urgent need to conduct in-depth research on satellite identity authentication methods to build a fundamental security mechanism for satellite communication networks. By implementing a complete identity authentication framework, not only can the verifiable access of satellite nodes be ensured, but also effective defense against spoofing attacks by illegal physical entities can be achieved, providing strong support for ensuring the information security of satellite communication systems.

[0003] Satellite identity authentication is a fundamental security mechanism for building satellite communication networks. By implementing a complete identity authentication framework, not only can verifiable access to satellite nodes be ensured, but also deception attacks by unauthorized physical entities can be effectively defended, providing reliable security for the operation of space information communication systems. Current research on satellite communication system security, particularly on identity authentication methods for the Giants constellation satellites, mainly focuses on physical layer (or radio frequency fingerprint) information authentication, identity registration information authentication (based on blockchain), and orbital prior information authentication schemes.

[0004] In terms of authentication methods based on physical layer (or radio frequency fingerprint) information, current research mainly collects satellite system physical layer or radio frequency parameter information, and then combines prior information for identity verification. In existing research, there is a physical layer-based (Physical-Layer Authentication, PLA) identity authentication scheme that can maintain a high authentication success rate and a low false rejection rate in both fixed satellite service (Fixed Satellite Service, FSS) and mobile satellite service (Mobile Satellite Service, MSS) scenarios. In addition, related research also involves a convolutional neural network structure based on in-phase quadrature (In-phase Quadrature, IQ) signals. At the same time, in order to enhance the security of satellite measurement and control links, a method of ground station identity recognition based on radio frequency fingerprint has been proposed.

[0005] In terms of authentication based on identity registration information, this type of method usually relies on identity identification, key agreement and digital signature mechanism in communication protocols, and realizes the verification of the identity of both parties through challenge / response authentication, symmetric or asymmetric encryption technology. In existing research, there is a design of a certificateless authentication model and an identity authentication scheme based on blockchain and cryptocurrency; in addition, for inter-constellation satellite communication scenarios, related research maintains the communication credentials between satellites in each constellation through a blockchain ledger, and introduces a space digital token as the basis for mutual trust between satellite nodes.

[0006] In terms of authentication methods based on orbital prior information, current research mainly analyzes the orbital motion characteristics of satellites, and uses their position and velocity parameters in space for identity verification. In existing research, a Time-Difference-of-Arrival (TDoA) based orbit authentication mechanism has been proposed; another research combines the orbit dynamics model and the observation model to establish a fixed orbit observation model, and uses the Kalman filter algorithm to estimate the satellite orbit state; in addition, research has also integrated multi-source measurement data of satellites and used a federated filtering algorithm to achieve high-precision determination of satellite attitude and orbit.

[0007] In summary, the existing satellite identity verification scheme researches generally have the problems of large data collection quantity and high computational complexity, which limits the real-time authentication of the system. At the same time, the strict synchronization requirement between multiple ground stations also does not fully exploit the space-time correlation of the data, which restricts the authentication accuracy. Against this background, the present application focuses on the satellite identity authentication problem in the mega-constellation environment and designs a satellite identity authentication scheme based on ephemeris information (TLE data) and statistical prior information of the satellite-ground channel. The scheme integrates a deep learning model into the low-orbit satellite identity legitimacy monitoring and authentication process, adopts a periodic and deep learning model updating strategy based on the identification accuracy performance threshold, reduces the requirement for data collection quantity and computational complexity, and improves the overall security of the mega-constellation satellite system while ensuring high authentication accuracy. SUMMARY

[0008] The purpose of the present application is to provide a satellite identity legitimacy verification method, device, equipment and medium for a mega-constellation system, which can realize high-precision identification of satellite identity.

[0009] To achieve the above purpose, the present application provides the following solutions.

[0010] In a first aspect, the present application provides a satellite identity legitimacy verification method for a mega-constellation system, comprising:

[0011] obtaining a satellite data set; the satellite data includes one or more of satellite signal time difference of arrival (TDoA), Doppler frequency shift (DoDFS) and signal average received power difference (DoRP), and a corresponding identity legitimacy label; the label includes legitimate, illegitimate and attack;

[0012] dividing the satellite data set into a training set and a test set;

[0013] constructing an identity legitimacy verification model based on CNN-LSTM deep learning;

[0014] training the identity legitimacy verification model based on the training set;

[0015] verifying and classifying the legitimacy of the satellites in the mega-constellation system based on the trained identity legitimacy verification model;

[0016] evaluating the accuracy of the trained identity legitimacy verification model, and if the evaluation result is lower than a preset threshold, retraining the identity legitimacy verification model until convergence.

[0017] Optionally, the step of obtaining a satellite data set specifically includes the following steps:

[0018] Step 1: import the ephemeris TLE data of the legal satellite constellation, the ephemeris TLE data of the simulated illegal satellite constellation and the unmanned aerial vehicle track information of the simulated spoofing attack;

[0019] Step 2: traverse the ephemeris TLE data of the legal satellite constellation, the ephemeris TLE data of the simulated illegal satellite constellation and the unmanned aerial vehicle track information of the simulated spoofing attack, and perform screening of candidate signal sources:

[0020] For the legal signal source, the simulated illegal signal source and the simulated spoofing unmanned aerial vehicle signal source, the ground station control center GS-CC generates the corresponding ground receiver coordinate list and measurement time list according to the pre-set ground receiver number and measurement times;

[0021] Step 3: based on the ground receiver coordinate list and the measurement time list, the ground station control center calls the coordinate calculation function of the current class belonging to the signal source based on the ephemeris TLE data of the legal satellite constellation, the ephemeris TLE data of the simulated illegal satellite constellation and the unmanned aerial vehicle track information of the simulated spoofing attack, to obtain the coordinate information and index serial number of the current specific signal source;

[0022] Step 4: perform visibility check on the specific signal source, and the specific check logic is: if the specific signal source is visible to any receiver in the ground station receiver coordinate list at the selected measurement time list, the specific signal source is taken as a candidate signal source and stored in the list; otherwise, the specific signal source is re-found until a preset number of candidate signal sources are successfully screened and the process ends;

[0023] For each type of signal source, the corresponding identity legitimacy real label is known before the above-mentioned screening of candidate signal sources is performed, when the data of the corresponding legal satellite constellation is collected, it is marked as a "True" label; when the data of the corresponding illegal satellite constellation is collected, it is marked as a "False" label; when the unmanned aerial vehicle signal source data is collected, it is marked as an "Attacker" label, and all satellites used to collect the data of the legal satellite constellation constitute a satellite white list;

[0024] Step 5: Calculate the root mean square error between the actual TDoA signature matrix of each candidate signal source in the three types of signal sources and the theoretical TDoA signature matrix of all imported satellites, the root mean square error between the actual Doppler frequency offset DoDFS signature matrix and the theoretical Doppler frequency offset signature DoDFS matrix of all imported satellites, and the root mean square error between the actual average received power DoRP signature matrix and the theoretical average received power DoRP signature matrix of all imported satellites, respectively denoted as the first root mean square error, the second root mean square error and the third root mean square error. The calculation of the above three types of root mean square errors utilizes the extension rule of the broadcast mechanism. The obtained root mean square error data of each type is a one-dimensional array, and the length of the array is equal to the number of all imported satellites. Each element in the array corresponds to the root mean square error between a theoretical satellite and an actual candidate satellite. The authentication result for each specific candidate signal source is:

[0025] Step 6: Select the satellite authentication tag with the smallest first root mean square error as the satellite first preliminary authentication result.

[0026] Step 7: Select the satellite authentication tag with the smallest second root mean square error as the satellite second preliminary authentication result.

[0027] Step 8: Select the satellite authentication tag with the smallest third root mean square error as the satellite third preliminary authentication result.

[0028] Step 9: Determine whether the above satellite first, second and third preliminary authentication results are in the satellite whitelist. If the satellite first, second and third preliminary authentication results are determined to exist in the legal satellite whitelist through retrieval, it is determined that the satellite first, second and third preliminary authentication results are legal satellite signal sources. At this time, return the "True" label and coordinate data for the satellite signal source. If not, it is determined that the satellite first, second and third preliminary authentication results are illegal satellite signal sources. At this time, temporarily return the "False" label and coordinate data for the illegal satellite signal source.

[0029] Step 10: The ground station control center further analyzes the legality of the signal source according to the satellite first, second and third preliminary authentication results in the previous step: first, the ground station control center constructs a three-dimensional grid between the geometric center of the ground receiver and the current candidate signal source. The three-dimensional grid is in the form of a cube, and the central axis is connected to the direction vector of the geometric center of the ground receiver and the current candidate signal source. By applying spatial rotation and scale to the local grid, it is accurately aligned in the space along the central axis direction, and covers all grid points to be measured along the line.

[0030] Step 11: Assuming each grid point in the three-dimensional grid is a corresponding stationary signal source, calculate its corresponding theoretical TDoA signature matrix, DoDFS signature matrix and DoRP signature matrix respectively; the ground station control center further calculates the root mean square error (RMSE) between the theoretical TDoA signature matrix, DoDFS signature matrix and DoRP signature matrix and the actual TDoA signature matrix, DoDFS signature matrix and DoRP signature matrix of the candidate signal source;

[0031] Step 12: According to the determination method of the satellite first, second and third preliminary authentication results, the ground station control center finds out the identity tags corresponding to the grid point signal sources with the smallest RMSE as the preliminary authentication results of the candidate signal source, which are called grid point first, second and third preliminary authentication results. This process is essentially to find the best matching grid point signal source in the three-dimensional grid;

[0032] Step 13: The ground station control center further analyzes the grid point first, second and third preliminary authentication results. If the height of the signal source corresponding to the grid point first, second and third preliminary authentication results is less than 100 kilometers, and the residual error improvement ratio is higher than the set value, the candidate signal source is finally determined as a spoofing attack, and the “Attacker” label and the grid coordinates corresponding to the grid point first, second and third preliminary authentication results are returned. Otherwise, it is determined that there is no other unknown signal source more suitable than the satellite first, second and third preliminary authentication results under the current time slot, and the satellite first, second and third preliminary authentication results are returned, i.e. the “True” label, or the “False” label, the satellite serial number index and the corresponding coordinate data.

[0033] At this point, the identity authentication for this specific signal source is completed, and the final authentication results of TDoA, DoDFS and DoRP are obtained. If two or more of the TDoA authentication results, DoDFS authentication results and DoRP authentication results are consistent, it means that the current candidate signal source data is valid, and the corresponding data is stored. Otherwise, it is considered as invalid data, and the data is discarded.

[0034] The above process is repeated until a sufficient number of labeled data is obtained.

[0035] Optionally, the expression of the TDoA signature matrix is as follows:

[0036]

[0037] where M TDoA represents the TDoA signature matrix, represents the TDoA value of the nth ground station relative to the reference station in the mth measurement, represents the dimension of the matrix, which is m rows and n columns;

[0038] The expression of the DoDFS signature matrix is as follows:

[0039]

[0040] Wherein, The difference between the Doppler frequency offset value of the satellite to be measured measured by the nth ground station in the mth measurement and the Doppler frequency offset value of the satellite to be measured measured by the reference ground station.

[0041] The expression of the DoRP signature matrix is as follows:

[0042]

[0043] Wherein, The difference between the signal average received power value of the satellite to be measured received by the nth ground station in the mth measurement and the signal average received power value of the satellite to be measured received by the reference ground station.

[0044] Optionally, the satellite identity legitimacy verification method of the mega constellation system further comprises the following steps after the step of "labeling the mixed signature matrix of the illegal satellite, the mixed signature matrix of the attacking drone and the mixed signature matrix of the legal satellite to obtain a satellite data set":

[0045] The mixed signature matrix of the illegal satellite, the mixed signature matrix of the attacking drone and the mixed signature matrix of the legal satellite are divided into a plurality of sub-matrices with fixed time steps along the time dimension by a sliding window.

[0046] Optionally, the identity legitimacy verification model comprises a CNN part, an LSTM part and a fully connected layer.

[0047] The CNN part, the LSTM part and the fully connected layer are connected in sequence.

[0048] Optionally, the training of the identity legitimacy verification model based on the training set comprises the following steps:

[0049] The cross-entropy loss function is used as the optimization objective, and the Adam optimizer is used for identity legitimacy verification model weight update.

[0050] In each round of training, samples are loaded from the training set in sequence, and the identity legitimacy verification model parameters are updated through the forward propagation and backward propagation processes of the identity legitimacy verification model.

[0051] Optionally, the accuracy of the trained identity legitimacy verification model is evaluated, and if the evaluation result is lower than a preset threshold, the identity legitimacy verification model is retrained until convergence, which comprises the following steps:

[0052] The identity legitimacy verification model is verified by using the test set, and a loss value and an accuracy of the identity legitimacy verification model on the verification set are calculated; the identity legitimacy verification model is not trained by using a Dropout regularization method during the training process, so as to ensure the integrity of time sequence continuity;

[0053] If the verification accuracy of the current round is better than that of the best model in the past, the current model parameter is saved;

[0054] The loss curve and the accuracy curve of the identity legitimacy verification model on the training set and the test set are drawn, and the authentication accuracy, the false alarm rate and the missed alarm rate of the test set are calculated based on a confusion matrix;

[0055] If the authentication accuracy, the false alarm rate and the missed alarm rate are lower than a preset threshold, the satellite data set is updated, and the identity legitimacy verification model is retrained.

[0056] In a second aspect, the application provides a satellite identity legitimacy verification device of a mega-constellation system, comprising:

[0057] A satellite data set acquisition module is configured to acquire a satellite data set; the satellite data comprises one or more of a satellite signal time difference of arrival (TDoA), a Doppler frequency shift (DoDFS) and a signal average received power difference (DoRP), and a corresponding identity legitimacy label; the label comprises legitimate, illegitimate and attack;

[0058] A data set division module is configured to divide the satellite data set into a training set and a test set;

[0059] An identity legitimacy verification model construction module is configured to construct an identity legitimacy verification model based on a CNN-LSTM deep learning;

[0060] A training module is configured to train the identity legitimacy verification model based on the training set;

[0061] A legitimacy verification and classification module is configured to verify and classify the legitimacy of a satellite in the mega-constellation system based on the trained identity legitimacy verification model;

[0062] A model evaluation module is configured to evaluate the accuracy of the trained identity legitimacy verification model; if the evaluation result is lower than a preset threshold, the identity legitimacy verification model is retrained until convergence.

[0063] According to the specific embodiments provided in the application, the application has the following technical effects:

[0064] The application provides a satellite identity legitimacy verification method and device of a mega constellation system. Local spatial features in a mixing matrix are extracted by using a CNN, and deep temporal and spatial correlations between elements in the mixing matrix are further captured by an LSTM, so that high-precision satellite identity discrimination is realized. Simulation results show that the scheme provided in the application has advantages in accuracy, low complexity and real-time performance compared with a benchmark scheme, and has good generalization ability and anti-interference robustness. Compared with a traditional method of matching judgment by calculating a root mean square error, the method provided in the application has advantages in identity discrimination accuracy, model convergence speed and running time delay. Under the condition of reasonable selection of the number of ground stations and the combination of measurement parameters, the authentication accuracy of the model on the test set reaches 99.51% taking TDoA data as an example. The method is especially suitable for a mega constellation satellite system with high real-time and security requirements. The scheme designed in the application can provide a technical reference for satellite identity authentication mechanisms in a mega constellation system. BRIEF DESCRIPTION OF DRAWINGS

[0065] In order to more clearly illustrate the technical solutions in the embodiments of the application or the prior art, the drawings needed in the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the application, and other drawings can be obtained by those skilled in the art without creative labor on the basis of the following drawings.

[0066] Figure 1 A flowchart of a satellite identity legitimacy verification method of a mega constellation system provided by an embodiment of the application;

[0067] Figure 2 A network model diagram of a mega constellation satellite communication system provided by an embodiment of the application;

[0068] Figure 3 A flowchart of a satellite identity authentication algorithm provided by an embodiment of the application;

[0069] Figure 4 A block diagram of an identity authentication system of a mega constellation satellite communication system provided by an embodiment of the application;

[0070] Figure 5 A data processing and training framework diagram of a satellite identity authentication framework based on a CNN-LSTM provided by an embodiment of the application;

[0071] Figure 6 A statistical performance diagram of a satellite identity legitimacy verification method of a mega constellation system provided by an embodiment of the application;

[0072] Figure 7 A structure diagram of a satellite identity legitimacy verification device of a mega constellation system provided by an embodiment of the application. DETAILED DESCRIPTION

[0073] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of protection of the present application.

[0074] The above purposes, features and advantages of the present application will be more apparent and understandable. The present application will be described in further detail below with reference to the drawings and specific embodiments.

[0075] In order to solve the problems mentioned in the background, while fully improving the performance of the satellite identity legality authentication system, the data required for identity authentication is reduced as much as possible, the real-time and response rate of the satellite identity authentication system are improved, the present application proposes a satellite identity legality authentication method for a mega constellation system based on deep learning. The method mainly includes three parts of data collection, model training and authentication and classification decision. Specifically, first, based on the satellite identity authentication system of the mega constellation system, the satellite signal Time-of-Arrival (ToA) data, Doppler frequency offset and average received power measured by the ground station are collected, and a satellite data set specific to a particular mega constellation is constructed. Then, the deep learning model is trained using the data set. Finally, the trained model is used as input with the mixed matrix of the satellite signal to be measured, and the satellite identity legality authentication and classification decision are performed. Similarly, the model can also learn the abnormal space-time correlation features in the mixed matrix corresponding to the signals sent by illegal satellites, low-altitude unmanned aerial vehicles and other attackers, thereby achieving effective classification and identification of legal satellites, illegal satellites and ground attack sources.

[0076] To achieve the above purpose, the present application proposes a deep learning authentication scheme combining Convolutional Neural Network (CNN) and Long Short-term Memory Networks (LSTM). The TDoA, Doppler frequency offset, and average received power mixed signature matrix data are used to train the deep learning model based on CNN and LSTM to capture the space-time correlation features in the data, so that the model can automatically determine the legality of the satellite according to one or more of the calculated TDoA, Doppler frequency offset, and average received power data measurements.

[0077] Specifically, the satellite identity legality verification method of the mega constellation system in the present application includes a ground station, a satellite to be measured, a low-altitude unmanned aerial vehicle, and a CNN-LSTM based deep learning model.

[0078] The ground station is a station on the ground that receives satellite signals in the system. In actual simulation, the positions of the ground receiving stations are regenerated at the beginning of each experiment. First, a random position is selected from the existing 21 hard-coded positions in the program as the center (reference ground station). After selecting the distribution diameter and the number of ground stations, the remaining n ground receiving stations are generated by copying. The positions of the ground stations are uniformly arranged on a circle with the reference ground station as the center and the selected distribution diameter as the radius. The minimum number of ground stations is 4, and the maximum number is 20. The distribution diameter is an integer between 1 km and 10 km (inclusive).

[0079] The satellites to be tested include legal satellites and illegal satellites of the mega constellation system, which are normally operating on the given orbit and maintaining normal communication status with the ground stations or other communication devices in the system.

[0080] The low-altitude drones are attackers who want to invade the target mega constellation system. They imitate legal satellites in the atmosphere and send fake authentication signals to the ground stations.

[0081] In an exemplary embodiment, as shown in Figure 1 A satellite identity verification method for a mega constellation system is provided, which is executed by a computer device, specifically by a terminal or a server, or by both a terminal and a server. The network model of the satellite communication system of the mega constellation is shown in Figure 2 The satellite network model of the satellite identity verification system mainly consists of three parts: the mega constellation system composed of low-orbit satellites (including legal satellites and illegal satellites of the mega constellation), attack drones in the atmosphere, and a certain number of ground receiving stations (GS).

[0082] In the embodiments of the present application, the following steps 101 to 105 are included. Among them:

[0083] Step 101: Obtain a satellite data set, which includes one or more of satellite signal time difference of arrival (TDoA), Doppler frequency shift (DoDFS), and signal average received power difference (DoRP), and corresponding identity legality labels; the labels include legal, illegal, and attack.

[0084] Step 102: Divide the satellite data set into a training set and a test set.

[0085] The construction of the data set includes data acquisition, labeling, and data preprocessing. Specifically as follows:

[0086] (a) Import other (non-lawful) satellite list and attack unmanned aerial vehicle, collect the signal time of arrival, Doppler frequency offset and signal average received power of illegal satellite or attack unmanned aerial vehicle, and establish the corresponding TDoA, DoDFS and DoRP hybrid signature matrix data;

[0087] (b) Import the list of legal satellites, use the TLE data of the satellite and the channel statistical model to calculate the theoretical signal time of arrival, Doppler frequency offset data and the theoretical value of the signal average received power, and establish the corresponding TDoA, DoDFS and DoRP hybrid matrix;

[0088] (c) According to the known satellite identity verification classification result, add a label to each of the above hybrid matrix: legal True / illegal False / attack Attack.

[0089] Among them, the satellite identity legality verification classification result is as follows:

[0090] The orbit parameters (TLE data) of the legal satellite are known, so the theoretical position and velocity information at a specific time can be obtained by orbit calculation method, and the corresponding hybrid matrix can be calculated combined with the measurement data of the ground station received signal. Since the above hybrid matrix comes from the identity verified legal satellite, its corresponding label is marked as "legal". The above hybrid matrix and its label are stored in the form of key-value pair dictionary (such as can be saved in.npy format data file), for subsequent model training. The elements in each hybrid matrix are collected in time sequence, forming time series data with fixed time interval, so that all elements in the matrix share consistent space-time correlation characteristics. In addition to legal satellites, the hybrid matrix constructed by satellites from other satellite constellation systems is processed in the same way and is uniformly labeled as "illegal" label; the hybrid matrix obtained by measuring the signal sent by the simulated attack device (such as low-altitude unmanned aerial vehicle) through the ground station is uniformly labeled as "attack" label.

[0091] According to the known satellite identity legality classification result, adding a label to each of the above hybrid matrix specifically includes:

[0092] Taking TDoA measurement as an example, after obtaining the TDoA signature matrix of the satellite under test, the system first loads a list containing TLE data for all satellites. Then, based on the reference time t0 and the time interval Δt, the specific execution period of this measurement is obtained. Afterward, the entire TLE data list of the satellite is traversed, and the specific positional changes of all satellites in the list are calculated within the measurement period to determine whether the satellites are visible during the measurement period, that is, whether the ground station can receive the signals transmitted by the satellites within the measurement period. In actual simulation, the elevation angle for measuring satellite visibility can be manually set, and then the maximum penalty is applied to invisible satellites. Then, using the TLE data of all satellites in the list, the theoretical TDoA signature matrix of the satellite at the measurement time is calculated. The TDoA of each satellite is subtracted from the TDoA of the satellite under test to calculate the root mean square error (RMSE) of each satellite. Then, the entire candidate satellite list is traversed to find the satellite with the smallest RMSE and its index. It is then determined whether the satellite belongs to the target constellation and a label of whether it is valid (True or False) is given. After that, by judging whether the altitude of the invalid satellite and the residual between the theoretical value and the calculated value are abnormal, it is evaluated whether the TDoA signature matrix is ​​calculated by a fake authentication signal emitted by the UAV and a label of whether it is an attack signal (False or Attack) is given.

[0093] See Figure 3 The complete authentication algorithm process is as follows:

[0094] Step 1: Import ephemeris TLE data of legitimate satellite constellations, ephemeris TLE data of simulating illegitimate satellite constellations, and drone flight path information simulating spoofing attacks;

[0095] Step 2: Traverse the ephemeris TLE data of the legitimate satellite constellation, the ephemeris TLE data of the simulated illegitimate satellite constellation, and the flight path information of the drone simulating a spoofing attack to filter candidate signal sources:

[0096] For legitimate signal sources, simulated illegal signal sources, and simulated deception drone signal sources, the ground station control center GS-CC generates a corresponding list of ground receiver coordinates and a list of measurement times based on the pre-set number of ground receivers and the number of measurements.

[0097] Step 3: Based on the ground receiver coordinate list and measurement time list, the ground station control center, based on the ephemeris TLE data of legitimate satellite constellations, the ephemeris TLE data of simulated illegal satellite constellations, and the UAV trajectory information of simulated deception attacks, calls the coordinate calculation function of the signal source to which the current class belongs, to obtain the coordinate information and index number of the current specific signal source;

[0098] Step 4: Perform a visibility check on the specific signal source, and the specific check logic is: if the specific signal source is visible to any receiver in the selected list of measurement time instances and the list of ground station receiver coordinates, then the specific signal source is taken as a candidate signal source and stored in the list; otherwise, the specific signal source is re-searched until a preset number of candidate signal sources are successfully screened.

[0099] For each type of signal source, the corresponding identity legitimacy and authenticity label is known before the above-mentioned screening of candidate signal sources. When data corresponding to a legal constellation satellite is collected, it is marked as a "True" label; when data corresponding to an illegal constellation satellite is collected, it is marked as a "False" label; and when unmanned aerial vehicle signal source data is collected, it is marked as an "Attacker" label. All satellites used to collect legal constellation satellite data constitute a satellite whitelist.

[0100] Step 5: Calculate the root mean square error between the actual TDoA signature matrix of each candidate signal source in the three types of signal sources and the theoretical TDoA signature matrix of all imported satellites, the root mean square error between the actual Doppler frequency shift DoDFS signature matrix and the theoretical Doppler frequency shift signature DoDFS matrix of all imported satellites, and the root mean square error between the actual average received power DoRP signature matrix and the theoretical average received power DoRP signature matrix of all imported satellites, respectively denoted as the first root mean square error, the second root mean square error, and the third root mean square error. The calculation of the above three types of root mean square errors utilizes the expansion rule of the broadcast mechanism, and the obtained root mean square error data of each type is a one-dimensional array, the length of the array is equal to the number of all imported satellites, and each element in the array corresponds to the root mean square error between a theoretical satellite and an actual candidate satellite. The authentication result for each specific candidate signal source is:

[0101] Step 6: Select the satellite authentication label with the smallest first root mean square error as the first preliminary satellite authentication result.

[0102] Step 7: Select the satellite authentication label with the smallest second root mean square error as the second preliminary satellite authentication result.

[0103] Step 8: Select the satellite authentication label with the smallest third root mean square error as the third preliminary satellite authentication result.

[0104] Step 9: Determine whether the first, second, and third preliminary authentication results of the satellite are in the satellite whitelist. If the first, second, and third preliminary authentication results of the satellite are determined to exist in the legal satellite whitelist through searching, it is determined that the first, second, and third preliminary authentication results of the satellite are legal satellite signal sources, and at this time, the "True" label and coordinate data are returned for the satellite signal source; if not, it is determined that the first, second, and third preliminary authentication results of the satellite are illegal satellite signal sources, and at this time, the "False" label and coordinate data are temporarily returned for the illegal satellite signal source;

[0105] Step 10: The ground station control center further analyzes the legality of the signal source according to the first, second, and third preliminary authentication results of the satellite in the previous step: first, the ground station control center constructs a three-dimensional grid between the geometric center of the ground receiver and the current candidate signal source, and the three-dimensional grid is in a cubic structure, and the direction vector connecting the geometric center of the ground receiver and the current candidate signal source is taken as the central axis, and the local grid is subjected to spatial rotation and scale zoom to accurately align the central axis direction in space and cover all grid points to be measured along the line;

[0106] Step 11: Assuming that each grid point in the three-dimensional grid is a corresponding stationary signal source, the corresponding theoretical TDoA signature matrix, DoDFS signature matrix, and DoRP signature matrix are calculated respectively; the ground station control center calculates the root mean square error RMSE between the theoretical TDoA signature matrix, DoDFS signature matrix, and DoRP signature matrix and the actual TDoA signature matrix, DoDFS signature matrix, and DoRP signature matrix of the candidate signal source;

[0107] Step 12: According to the determination method of the first, second, and third preliminary authentication results of the satellite, the ground station control center finds out the identity labels corresponding to the grid points with the smallest RMSE in the three categories as the preliminary authentication results of the candidate signal source, which are called grid point first, second, and third preliminary authentication results. This process is essentially to find the best matching grid point signal source in the above three-dimensional grid points;

[0108] Step 13: The ground station control center further analyzes the first, second, and third preliminary authentication results of the grid point. If the signal source height corresponding to the first, second, and third preliminary authentication results of the grid point is less than 100 kilometers, and the residual improvement ratio is higher than the set value, the final determination is that the candidate signal source is a spoofing attack, and the "Attacker" label and the grid coordinates corresponding to the first, second, and third preliminary authentication results of the grid point are returned. Otherwise, it is determined that there is no other unknown signal source more suitable than the first, second, and third preliminary authentication results of the satellite in the current time slot, and the first, second, and third preliminary authentication results of the satellite are returned, i.e., the "True" label, or the "False" label, the satellite serial number index, and the corresponding coordinate data.

[0109] At this point, the identity authentication for this specific signal source is completed, and the final authentication results of TDoA, DoDFS, and DoRP are obtained. If two or more of the TDoA authentication result, DoDFS authentication result, and DoRP authentication result are consistent, it means that the current candidate signal source data is valid, and the corresponding data is stored. Otherwise, it is considered as invalid data and discarded.

[0110] The above process is repeated until a sufficient number of labeled data is obtained.

[0111] The above process details the complete method of satellite identification, legitimacy judgment, and attack signal detection based on TDoA signature matrix. It is emphasized that for Doppler frequency offset and signal average received power measurement data, the system will use the same logical framework for processing: first load TLE data, determine the measurement period, traverse the satellite list for visibility determination and punishment, calculate the theoretical signature matrix, compare with the measured signature matrix to obtain RMSE, select the minimum RMSE candidate star, perform constellation legitimacy judgment, and finally evaluate whether it is an attack signal. Next, specific steps will be introduced for Doppler frequency offset measurement data and signal average received power measurement data.

[0112] Next, TDoA signature matrix, DoDFS signature matrix, and DoRP signature matrix will be introduced in detail:

[0113] (1) TDoA signature matrix:

[0114] First, the distance d g and the speed of light c, the time of the signal from the satellite to the ground station, i.e., the Time-of-Flight (ToF), can be calculated:

[0115]

[0116] Secondly, if the satellite under test sends an authentication signal to ground station A at time t0, the time of arrival of the signal at ground station A can be calculated according to ToF, i.e. ToA:

[0117]

[0118] Similarly, when other ground stations in the system receive the authentication signal sent by the satellite under test, the corresponding ToA can be obtained. The difference between the ToA obtained by ground station A and the ToA obtained by ground station B can be obtained as the time difference of arrival TDoA between ground station A and ground station B gg′ :

[0119]

[0120] Then, the ToA of all the remaining ground stations is subtracted from the ToA of the reference ground station, so as to obtain the TDoA of the reference ground station and all the ground stations. After that, the corresponding TDoA values are arranged in a row according to the order in which the ground stations receive the satellite signals, so as to obtain a TDoA row vector about the satellite under test at the measurement time t0. Assuming that the time interval (time step) between each measurement is Δt, a similar TDoA row vector can be obtained every Δt time.

[0121] The expression of the TDoA row vector is:

[0122]

[0123] Thus, after the number of measurements is set, a TDoA signature matrix about the satellite under test can be obtained. The elements in a TDoA signature matrix are m·n measurement data generated by n+1 ground receiving stations relative to the orbital positions of m transmitting satellites, wherein m is the number of rows of the TDoA matrix, indicating the number of measurements of the satellite signals by the ground stations, and n+1 is the number of columns of the TDoA matrix, indicating the number of ground stations.

[0124] The expression of the TDoA signature matrix is:

[0125]

[0126] wherein, represents the TDoA value of the nth ground station relative to the reference station in the mth measurement.

[0127] It is particularly pointed out that after the ToA of each ground station is obtained, the data format needs to be converted from Universal Time Coordinated (UTC) to Julian day (JD) format, which is a time format adopted by astronomers. The conversion converts the format of the time into a floating-point number, which is convenient for subsequent calculation.

[0128] (2) DoDFS signature matrix:

[0129] The Doppler shift calculation in the satellite signal authentication system is based on the classical Doppler effect principle:

[0130] The system first calculates the geometric vector pointing from each ground station to the satellite and normalizes it to a unit vector to represent the signal propagation direction, where represents the position vector pointing from the Earth center to the satellite in the True Equator Mean Equinox (TEME) coordinates; represents the position vector pointing from the Earth center to the ground station in the TEME coordinates.

[0131] Then the relative velocity vector of the satellite with respect to the ground station is calculated where the ground station velocity has considered the Earth rotation effect.

[0132] Finally, the radial velocity component is extracted through vector dot product: which is the only velocity component affecting the Doppler shift. The final Doppler shift calculation formula is:

[0133]

[0134] where f0 is the carrier frequency (1575.42 MHz), c is the speed of light, and the negative sign is used to ensure a negative frequency shift when the satellite is moving away from the receiver.

[0135] The system uses vector calculation to handle multiple receivers, multiple measurement times, and multiple candidate satellites simultaneously, forming a complete Doppler "fingerprint" matrix. By comparing the measured frequency shift with the theoretical frequency shift of all satellites in the database, the system can identify the most matched satellite and detect potential spoofing attacks.

[0136] The DoDFS signature matrix expression is:

[0137]

[0138] where represents the difference between the Doppler frequency shift of the satellite being measured measured by the nth ground station in the mth measurement and the Doppler frequency shift of the satellite being measured measured by the reference ground station.

[0139] (3) DoRP signature matrix:

[0140] Distance calculation: The system first calculates the Euclidean distance between the receiver and the satellite based on their positions. Given the receiver position rec pos(x, y, z) and satellite position sat pos (x, y, z) and satellite position sat

[0141]

[0142] Free space path loss calculation: The system uses the standard free space path loss (FSPL) model to calculate the power attenuation of the signal propagating in the vacuum. The formula is:

[0143] FSPL (dB) = 20 x log 10 (d meters ) + 20 x log 10 (f Hz ) - 147.55 (9)

[0144] where d is the propagation distance (m), f Hz is the signal frequency (the system uses a frequency of 1.57542 GHz), and the constant term 147.55 comes from the calculation result of .

[0145] Received power calculation: According to the link budget principle, the received power is equal to the transmitted power minus the path loss: P received (dBW) = EIRP (dBW) - FSPL (dB), where EIRP (Equivalent Isotropically Radiated Power) is set to the typical value of the satellite, which is 14.3 dBW.

[0146] Import error model: In order to simulate the measurement uncertainty in the real environment, the system superimposes an error model on the calculated theoretical power value, usually using a Gaussian distribution or a lognormal distribution of power error: P measured = P theoretical -error (dB).

[0147] The expression of the DoRP signature matrix is:

[0148]

[0149] where represents the difference between the average received power value of the signal of the satellite to be measured received by the nth ground station in the mth measurement and the average received power value of the signal of the satellite to be measured received by the reference ground station.

[0150] The above calculation method is based on the ideal free space propagation assumption, although it ignores complex factors such as atmospheric attenuation and multipath effect, but it is sufficient and effective for the relative power feature comparison in satellite signal authentication, and can provide a reliable theoretical basis for subsequent signal authenticity discrimination.

[0151] Finally, in actual measurement, considering that the system is in the same condition, based on TDoA, DoDFS and DoRP signature matrix, there may be inconsistencies in generating satellite legitimacy and attack signal label. The final judgment adopts the fusion strategy: in the case of obtaining three types of labels at the same time, adopt two or more consistent labels; if the three types of labels are different from each other, give up this group of data and regard it as invalid. Similarly, when only two types of labels can be obtained, only when they are completely consistent can they be adopted, otherwise they are also abandoned and marked as invalid data. This rule verifies the three independent physical quantities of TDoA, DoDFS and DoRP through cross-validation, and uses the majority consensus principle to improve the robustness and credibility of the final judgment in complex environment and confrontation scene.

[0152] Step 103: Construct an identity legitimacy verification model based on CNN-LSTM deep learning.

[0153] The identity legitimacy verification model consists of two parts. The first part is CNN, which is used to extract local spatial features in the sliding window sub-matrix; the second part is LSTM, which is used to model the time series output by CNN to capture the feature dependence relationship across time steps.

[0154] The classification of the input sub-matrix is completed through the full connection layer, and the legitimacy label is output.

[0155] Step 104: Train the identity legitimacy verification model based on the training set.

[0156] Referring to Figure 4 and Figure 5 , the model training process is as follows:

[0157] During model training, the label adopts integer encoding method: the attacking unmanned aerial vehicle is 0, the legitimate satellite is 1, and the illegal satellite is 2; the mixed matrix is divided into multiple sub-matrices with fixed time steps through sliding window along the time dimension, so as to enhance the learning ability of the model to local time sequence change;

[0158] The cross-entropy loss function is used as the optimization objective, and the Adam optimizer is used for model weight update;

[0159] In each round of training, samples are loaded from the training set in turn, and network parameters are updated through the forward propagation and backward propagation process of CNN-LSTM network;

[0160] After each round of training, the test set is used to verify the model, and the loss value and accuracy of the model on the validation set are calculated; no regularization means such as Dropout is used in the model training process to ensure the integrity of time sequence continuity;

[0161] If the current round of verification accuracy is better than the best model ever, save the current model parameters;

[0162] Step 105: Based on the trained identity legitimacy verification model, the legitimacy of the satellite in the giant constellation system is verified and classified.

[0163] Collect the ToA, Doppler frequency offset and received power data of the satellite signal to be tested, construct a hybrid matrix corresponding to the data, use the model trained in step 104 to directly determine the legitimacy of the satellite identity to be tested, and finally take corresponding response actions according to the classification results.

[0164] The expression of the hybrid matrix is:

[0165]

[0166] In order to make the local time correlation feature of the hybrid signature matrix more obvious, after obtaining the hybrid signature matrix data of the satellite to be tested, the sliding window is used along the time sequence to divide the matrix again. The time step between each adjacent window is one, that is, a sub-matrix is collected every other row with a sliding window. In this process, the space-time correlation in a hybrid signature matrix is amplified, effectively improving the training effect of the model. The sub-matrix after division still adds the label of the original matrix and is saved in the.npy data file in the form of a dictionary.

[0167] The present application builds an identity authentication system for a giant constellation satellite communication system. The input of the system is the signal arrival time, Doppler frequency offset, and average signal received power measured by multiple ground stations, and the output is a three-class label after system authentication. After the ground station obtains the signal arrival time, Doppler frequency offset, and average signal received power, it subtracts the signal arrival time, Doppler frequency offset, and average signal received power of the reference ground station to obtain hybrid signature matrix data. Then, the data is input into a deep learning model, which captures the space-time correlation in the data. Finally, the model outputs the satellite legitimacy label after authentication.

[0168] Step 106: Evaluate the accuracy of the trained identity legitimacy verification model. If the evaluation result is lower than the preset threshold, retrain the identity legitimacy verification model until convergence.

[0169] After training, draw the loss curve and accuracy curve of the model on the training set and test set, and calculate the authentication accuracy, false alarm rate and missed alarm rate of the test set based on the confusion matrix, which are used to monitor the accuracy and timeliness of the trained model. If the performance indicators are lower than the performance threshold, the original training data set needs to be updated (especially as the space environment changes, the space-time correlation features contained in the received power data may change), and then the identity legitimacy verification model is retrained.

[0170] The application extracts local spatial features in the mixing matrix using a CNN, and further captures deep temporal and spatial correlations between elements in the mixing matrix through an LSTM, to achieve high-precision identification of satellite identity. Simulation results show that the proposed scheme has advantages in accuracy, low complexity and real-time performance compared to the baseline scheme, and shows good generalization ability and anti-interference robustness. Compared with the traditional method of relying on root mean square error for matching judgment, the method of the application shows advantages in identity identification accuracy, model convergence speed and running delay. Under the condition of reasonable selection of the number of ground stations and the combination of measurement parameters, taking TDoA data as an example, the authentication accuracy of the model on the test set reaches 99.51%, and it is especially suitable for giant constellation satellite systems with high real-time and security requirements. The scheme designed in the application can provide technical reference for satellite identity authentication mechanism in giant constellation system.

[0171] Figure 6 The statistical performance diagram of the satellite identity legality verification based on the TDoA data and the CNN-LSTM deep learning-based identity legality verification model shows the changes of the test set authentication accuracy, false alarm rate and missed alarm rate with the TDoA window size. The accuracy shows a trend of first rising and then falling with the window size: when the window size gradually increases from 1 to 5, the accuracy steadily increases and reaches the highest value of 99.51% when the window size is 5, indicating that the model recognition effect of the sample is best at this time. Subsequently, as the window size continues to increase, the accuracy begins to decrease, indicating that too large window size may have a negative impact on the model performance, and the reason may be that the increase of the window size increases the number of time slots in the TDoA matrix, which in turn causes the model to not be able to capture the internal relationship of the data well. When the window size is 1, the false alarm rate is as high as 4.0%, and the missed alarm rate is 1.47%. As the window size increases, the false alarm rate generally shows a downward trend and reaches the lowest value of about 0.4% when the window size is 5; at the same time, the value of the missed alarm rate also decreases synchronously and reaches the lowest value of 0.18% when the window size is 5. However, when the window size continues to increase beyond 5, both the false alarm rate and the missed alarm rate begin to rise, and when the window size is 8, the false alarm rate rises to about 2.71% and the missed alarm rate rises to 1.61%. From the three curves, it can be seen that when the window size is 5, the model reaches a good balance in accuracy, false alarm rate and missed alarm rate, which is a relatively optimal parameter selection.

[0172] Based on the same inventive concept, the embodiments of the present application also provide a satellite identity legitimacy verification device of a mega-constellation system for implementing the satellite identity legitimacy verification method of the mega-constellation system. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme described in the above method, so the specific limitations in one or more satellite identity legitimacy verification device embodiments of the mega-constellation system provided below can refer to the limitations of the satellite identity legitimacy verification method of the mega-constellation system described above, and will not be repeated here.

[0173] In one exemplary embodiment, as shown in Figure 7 a satellite identity legitimacy verification device of a mega-constellation system is provided, comprising:

[0174] a satellite data set acquisition module 201 for acquiring a satellite data set; the satellite data includes one or more of satellite signal time difference of arrival TDoA, Doppler frequency shift DoDFS and signal average received power difference DoRP, and a corresponding identity legitimacy label; the label includes legitimate, illegal and attack;

[0175] a data set division module 202 for dividing the satellite data set into a training set and a test set;

[0176] an identity legitimacy verification model construction module 203 for constructing an identity legitimacy verification model based on CNN-LSTM deep learning;

[0177] a training module 204 for training the identity legitimacy verification model based on the training set;

[0178] a legitimacy verification and classification module 205 for verifying and classifying the legitimacy of the satellites in the mega-constellation system based on the trained identity legitimacy verification model;

[0179] a model evaluation module 206 for evaluating the accuracy of the trained identity legitimacy verification model, and if the evaluation result is lower than a preset threshold, retraining the identity legitimacy verification model until convergence.

[0180] The principles and implementation modes of the present application are described by applying specific examples in this paper, and the above example description is only for helping to understand the method and its core idea of the present application; at the same time, for those skilled in the art, according to the idea of the present application, there are changes in specific implementation modes and application scope. In conclusion, the content of the present description should not be understood as a limitation of the present application.

Claims

1. A method for verifying the legitimacy of satellite identities in the Giants constellation system, characterized in that, The satellite identity verification method of the Giants constellation system includes: Obtain satellite datasets; the satellite data includes one or more of the following: time difference of arrival (TDoA), Doppler frequency offset (DoDFS), and average received power difference (DoRP), as well as corresponding identity legitimacy labels; the labels include legitimate, illegitimate, and attack. The satellite dataset is divided into a training set and a test set; Construct an identity verification model based on CNN-LSTM deep learning; The identity verification model is trained based on the training set; The legitimacy of satellites in the Giants constellation system is verified and classified based on a trained identity legitimacy verification model. The accuracy of the trained identity verification model is evaluated. If the evaluation result is lower than a preset threshold, the identity verification model is retrained until convergence. The acquisition of the satellite dataset specifically includes the following steps: Step 1: Import ephemeris TLE data of legitimate satellite constellations, ephemeris TLE data of simulating illegitimate satellite constellations, and drone flight path information simulating spoofing attacks; Step 2: Traverse the ephemeris TLE data of the legitimate satellite constellation, the ephemeris TLE data of the simulated illegitimate satellite constellation, and the flight path information of the drone simulating a spoofing attack to filter candidate signal sources: For legitimate signal sources, simulated illegal signal sources, and simulated deception drone signal sources, the ground station control center GS-CC generates a corresponding list of ground receiver coordinates and a list of measurement times based on the pre-set number of ground receivers and the number of measurements. Step 3: Based on the ground receiver coordinate list and measurement time list, the ground station control center, based on the ephemeris TLE data of legitimate satellite constellations, the ephemeris TLE data of simulated illegal satellite constellations, and the UAV trajectory information of simulated deception attacks, calls the coordinate calculation function of the signal source to which the current class belongs, to obtain the coordinate information and index number of the current specific signal source; Step 4: Perform a visibility check on the specific signal source. The specific check logic is as follows: if the specific signal source is visible to any receiver in the ground station receiver coordinate list under the selected measurement time list, then the specific signal source is selected as a candidate signal source and stored in the list; otherwise, the specific signal source is searched again until a preset number of candidate signal sources are successfully filtered and the process ends. For each type of signal source, before screening the candidate signal sources, their corresponding identity legitimacy labels are known. When data from a legitimate constellation satellite is collected, it is marked as "True"; when data from an illegitimate constellation satellite is collected, it is marked as "False"; and when data from a drone signal source is collected, it is marked as "Attacker". All satellites used to collect data from legitimate constellation satellites constitute a legitimate satellite whitelist. Step 5: Calculate the root mean square error (RMSE) between the actual TDoA signature matrix and the theoretical TDoA signature matrix of all imported satellites for each candidate signal source in the three signal source categories; the RMSE between the actual Doppler frequency offset (DoDFS) signature matrix and the theoretical Doppler frequency offset (DoDFS) signature matrix of all imported satellites; and the RMSE between the actual average received power (DoRP) signature matrix and the theoretical average received power (DoRP) signature matrix of all imported satellites. These are denoted as the first RMSE, second RMSE, and third RMSE, respectively. Each type of RMSE data is a one-dimensional array, with the array length equal to the number of imported satellites. Each element in the array corresponds to the RMSE between a theoretical satellite and the actual candidate satellite. The authentication results for each specific candidate signal source are as follows: Step 6: Select the satellite certification label with the smallest root mean square error as the first preliminary certification result for the satellite; Step 7: Select the satellite certification label with the smallest second root mean square error as the second preliminary certification result for the satellite; Step 8: Select the satellite certification label with the smallest third root mean square error as the third preliminary certification result of the satellite; Step 9: Determine whether the first, second, and third preliminary authentication results of the satellite are in the legal satellite whitelist. If the first, second, and third preliminary authentication results of the satellite are found to be in the legal satellite whitelist after retrieval, then the first, second, and third preliminary authentication results of the satellite are determined to be legal satellite signal sources. At this time, a "True" label and coordinate data are returned for the satellite signal source. If not, then the first, second, and third preliminary authentication results of the satellite are determined to be illegal satellite signal sources. At this time, a "False" label and coordinate data are temporarily returned for the illegal satellite signal source. Step 10: Based on the preliminary authentication results of the first, second, and third satellites in the previous step, the ground station control center further analyzes the legitimacy of the signal source: First, the ground station control center constructs a three-dimensional grid between the geometric center of the ground receiver and the current candidate signal source. The three-dimensional grid has a cubic structure, with the direction vector connecting the geometric center of the ground receiver and the current candidate signal source as the central axis. By applying spatial rotation and scaling to the local grid, it is made to accurately align with the central axis direction in space and cover all grid points to be tested along the line. Step 11: Assuming that each grid point in the three-dimensional grid is a corresponding stationary signal source, calculate its corresponding theoretical TDoA signature matrix, DoDFS signature matrix, and DoRP signature matrix; the ground station control center then calculates the root mean square error (RMSE) between its theoretical TDoA signature matrix, DoDFS signature matrix, and DoRP signature matrix and the actual TDoA signature matrix, DoDFS signature matrix, and DoRP signature matrix of the candidate signal source. Step 12: According to the judgment method of the first, second and third preliminary authentication results of the satellite mentioned above, the ground station control center finds the identity label corresponding to the grid point signal source with the minimum RMSE in the three categories as the preliminary authentication result of the candidate signal source, which is called the first, second and third preliminary authentication results of the grid point. This process is essentially to find the best matching grid point signal source in the above three-dimensional grid points. Step 13: The ground station control center further analyzes the first, second, and third preliminary authentication results of the grid points. If the signal source altitude corresponding to the first, second, and third preliminary authentication results of the grid points is less than 100 kilometers and the residual improvement ratio is higher than the set value, then the candidate signal source is finally determined to be a spoofing attack, and the "Attacker" label and the grid coordinates of the corresponding first, second, and third preliminary authentication results of the grid points are returned. Otherwise, it is determined that there is no other unknown signal source more suitable than the first, second, and third preliminary authentication results of the satellite in the current time slot, and the first, second, and third preliminary authentication results of the satellite are returned, i.e., the "True" label or the "False" label, the satellite serial number index, and the corresponding coordinate data. At this point, the authentication of the specific signal source is complete, and the final authentication results of TDoA, DoDFS, and DoRP are obtained. If two or more of the authentication results of TDoA, DoDFS, and DoRP are consistent, it means that the current candidate signal source data is valid and the corresponding data is stored; otherwise, the data is considered invalid and is discarded. Repeat the above process until a sufficient amount of labeled data is obtained.

2. The satellite identity verification method for the Giants constellation system according to claim 1, characterized in that, The expression for the TDoA signature matrix is ​​as follows: ； in, Represents the TDoA signature matrix. Indicates the first In the second measurement TDoA value of each ground station relative to the base station This indicates that the matrix has dimensions of m rows and n columns; The expression for the DoDFS signature matrix is ​​as follows: ； in, Indicates the first In the second measurement The difference between the Doppler frequency offset value of the satellite under test measured by each ground station and the Doppler frequency offset value of the satellite signal under test measured by the reference ground station; The expression for the DoRP signature matrix is ​​as follows: ； in, Indicates the first In the second measurement The difference between the average received power of the satellite signal received by the ground station and the average received power of the satellite signal received by the reference ground station.

3. The satellite identity verification method for the Giants constellation system according to claim 1, characterized in that, The satellite identity legitimacy verification method for the Giant Star system further includes the following steps after "adding labels to the hybrid signature matrix of the illegal satellite, the hybrid signature matrix of the attacking drone, and the hybrid signature matrix of the legitimate satellite to obtain the satellite dataset": The hybrid signature matrix of the illegal satellite, the hybrid signature matrix of the attacking drone, and the hybrid signature matrix of the legitimate satellite are divided into multiple sub-matrices with fixed time steps along the time dimension using a sliding window.

4. The satellite identity verification method for the Giant Pisces system according to claim 3, characterized in that, The identity verification model includes: a CNN part, an LSTM part, and a fully connected layer; The CNN part, LSTM part, and fully connected layer are connected sequentially.

5. The satellite identity verification method for the Giants constellation system according to claim 1, characterized in that, Training the identity legitimacy verification model based on the training set specifically includes the following steps: The cross-entropy loss function is used as the optimization objective, and the Adam optimizer is used to update the weights of the identity verification model. In each round of training, samples are loaded sequentially from the training set, and the identity verification model parameters are updated through the forward and backward propagation processes of the identity verification model.

6. The satellite identity verification method for the Giant Pisces system according to claim 1, characterized in that, The accuracy of the trained identity verification model is evaluated. If the evaluation result is lower than a preset threshold, the identity verification model is retrained until convergence. This specifically includes the following steps: The identity legitimacy verification model is validated using a test set, and the loss and accuracy of the identity legitimacy verification model on the validation set are calculated. Dropout regularization was not used during the training of the identity legitimacy verification model to ensure the integrity of temporal continuity. If the current validation accuracy is better than the previous best model, then save the current model parameters; Plot the loss curve and accuracy curve of the identity legitimacy verification model on the training set and test set, and calculate the authentication accuracy, false alarm rate and false alarm rate of the test set based on the confusion matrix; If the authentication accuracy, false alarm rate, and missed alarm rate are lower than preset thresholds, the satellite dataset is updated, and the identity legitimacy verification model is retrained.

7. A satellite identity verification device for the Giants constellation system, characterized in that, The satellite identity verification device of the Giant Star system includes: A satellite dataset acquisition module is used to acquire satellite datasets; the satellite data includes one or more of the following: time difference of arrival (TDoA), Doppler frequency offset (DoDFS), and average received power difference (DoRP), as well as corresponding identity legitimacy labels; the labels include legitimate, illegitimate, and attack. The dataset partitioning module is used to divide the satellite dataset into a training set and a test set; The identity legitimacy verification model building module is used to build an identity legitimacy verification model based on CNN-LSTM deep learning. The training module is used to train the identity legitimacy verification model based on the training set; The legitimacy verification and classification module is used to verify and classify the legitimacy of satellites in the Giants constellation system based on a trained identity legitimacy verification model. The model evaluation module is used to evaluate the accuracy of the trained identity verification model. If the evaluation result is lower than a preset threshold, the identity verification model is retrained until convergence. The acquisition of the satellite dataset specifically includes the following steps: Step 1: Import ephemeris TLE data of legitimate satellite constellations, ephemeris TLE data of simulating illegitimate satellite constellations, and drone flight path information simulating spoofing attacks; Step 2: Traverse the ephemeris TLE data of the legitimate satellite constellation, the ephemeris TLE data of the simulated illegitimate satellite constellation, and the flight path information of the drone simulating a spoofing attack to filter candidate signal sources: For legitimate signal sources, simulated illegal signal sources, and simulated deception drone signal sources, the ground station control center GS-CC generates a corresponding list of ground receiver coordinates and a list of measurement times based on the pre-set number of ground receivers and the number of measurements. Step 3: Based on the ground receiver coordinate list and measurement time list, the ground station control center, based on the ephemeris TLE data of legitimate satellite constellations, the ephemeris TLE data of simulated illegal satellite constellations, and the UAV trajectory information of simulated deception attacks, calls the coordinate calculation function of the signal source to which the current class belongs, to obtain the coordinate information and index number of the current specific signal source; Step 4: Perform a visibility check on the specific signal source. The specific check logic is as follows: if the specific signal source is visible to any receiver in the ground station receiver coordinate list under the selected measurement time list, then the specific signal source is selected as a candidate signal source and stored in the list; otherwise, the specific signal source is searched again until a preset number of candidate signal sources are successfully filtered and the process ends. For each type of signal source, before screening the candidate signal sources, their corresponding identity legitimacy labels are known. When data from a legitimate constellation satellite is collected, it is marked as "True"; when data from an illegitimate constellation satellite is collected, it is marked as "False"; and when data from a drone signal source is collected, it is marked as "Attacker". All satellites used to collect data from legitimate constellation satellites constitute a legitimate satellite whitelist. Step 5: Calculate the root mean square error (RMSE) between the actual TDoA signature matrix and the theoretical TDoA signature matrix of all imported satellites for each candidate signal source in the three signal source categories; the RMSE between the actual Doppler frequency offset (DoDFS) signature matrix and the theoretical Doppler frequency offset (DoDFS) signature matrix of all imported satellites; and the RMSE between the actual average received power (DoRP) signature matrix and the theoretical average received power (DoRP) signature matrix of all imported satellites. These are denoted as the first RMSE, second RMSE, and third RMSE, respectively. Each type of RMSE data is a one-dimensional array, with the array length equal to the number of imported satellites. Each element in the array corresponds to the RMSE between a theoretical satellite and the actual candidate satellite. The authentication results for each specific candidate signal source are as follows: Step 6: Select the satellite certification label with the smallest root mean square error as the first preliminary certification result for the satellite; Step 7: Select the satellite certification label with the smallest second root mean square error as the second preliminary certification result for the satellite; Step 8: Select the satellite certification label with the smallest third root mean square error as the third preliminary certification result of the satellite; Step 9: Determine whether the first, second, and third preliminary authentication results of the satellite are in the legal satellite whitelist. If the first, second, and third preliminary authentication results of the satellite are found to be in the legal satellite whitelist after retrieval, then the first, second, and third preliminary authentication results of the satellite are determined to be legal satellite signal sources. At this time, a "True" label and coordinate data are returned for the satellite signal source. If not, then the first, second, and third preliminary authentication results of the satellite are determined to be illegal satellite signal sources. At this time, a "False" label and coordinate data are temporarily returned for the illegal satellite signal source. Step 10: Based on the preliminary authentication results of the first, second, and third satellites in the previous step, the ground station control center further analyzes the legitimacy of the signal source: First, the ground station control center constructs a three-dimensional grid between the geometric center of the ground receiver and the current candidate signal source. The three-dimensional grid has a cubic structure, with the direction vector connecting the geometric center of the ground receiver and the current candidate signal source as the central axis. By applying spatial rotation and scaling to the local grid, it is made to accurately align with the central axis direction in space and cover all grid points to be tested along the line. Step 11: Assuming that each grid point in the three-dimensional grid is a corresponding stationary signal source, calculate its corresponding theoretical TDoA signature matrix, DoDFS signature matrix, and DoRP signature matrix; the ground station control center then calculates the root mean square error (RMSE) between its theoretical TDoA signature matrix, DoDFS signature matrix, and DoRP signature matrix and the actual TDoA signature matrix, DoDFS signature matrix, and DoRP signature matrix of the candidate signal source. Step 12: According to the judgment method of the first, second and third preliminary authentication results of the satellite mentioned above, the ground station control center finds the identity label corresponding to the grid point signal source with the minimum RMSE in the three categories as the preliminary authentication result of the candidate signal source, which is called the first, second and third preliminary authentication results of the grid point. This process is essentially to find the best matching grid point signal source in the above three-dimensional grid points. Step 13: The ground station control center further analyzes the first, second, and third preliminary authentication results of the grid points. If the signal source altitude corresponding to the first, second, and third preliminary authentication results of the grid points is less than 100 kilometers and the residual improvement ratio is higher than the set value, then the candidate signal source is finally determined to be a spoofing attack, and the "Attacker" label and the grid coordinates of the corresponding first, second, and third preliminary authentication results of the grid points are returned. Otherwise, it is determined that there is no other unknown signal source more suitable than the first, second, and third preliminary authentication results of the satellite in the current time slot, and the first, second, and third preliminary authentication results of the satellite are returned, i.e., the "True" label or the "False" label, the satellite serial number index, and the corresponding coordinate data. At this point, the authentication of the specific signal source is complete, and the final authentication results of TDoA, DoDFS, and DoRP are obtained. If two or more of the authentication results of TDoA, DoDFS, and DoRP are consistent, it means that the current candidate signal source data is valid and the corresponding data is stored; otherwise, the data is considered invalid and is discarded. Repeat the above process until a sufficient amount of labeled data is obtained.

Images