Non-cooperative star chain signal matching identification method combined with known satellite constellation signal measurement

Abstract

The invention discloses a non-cooperative star chain signal matching identification method combined with known satellite constellation signal measurement, and solves the problem of non-cooperative star chain signal matching identification when the position of a receiver is unknown and the receiver cannot depend on frame synchronization and demodulation information. And the technical problems of fuzzy star chain carrier frequency and difficulty in real-time identification in a dynamic scene are solved. The method comprises the following implementation steps: establishing a measurement data set; constructing a receiver speed resolving equation set; generating a non-cooperative star chain signal matching test equation set; generating a receiver speed candidate set; screening a receiver speed candidate set; determining a probability speed; and obtaining a matching identification result. According to the method, the satellite with the known identity and the intercepted signal of the satellite are used for assisting non-cooperative star chain signal matching recognition, real-time matching recognition of the moving platform under the carrier frequency fuzzy condition of the non-cooperative star chain satellite can be achieved, calculation complexity is linearly controllable, the method does not depend on the real position of a receiver, and reliability is high. The method can be used for receiver navigation and positioning based on non-cooperative star chain signals.

Description

Technical Field

[0001] This invention belongs to the field of satellite navigation and positioning technology, and mainly relates to the identification of non-cooperative satellite signals. Specifically, it is a method for matching and identifying non-cooperative Starlink signals by combining signal measurements from known satellite constellations. This invention utilizes the known identities of satellites and their intercepted signals to assist in matching and identifying non-cooperative Starlink satellite signals. It is used for receiver navigation and positioning based on non-cooperative Starlink signals. Background Technology

[0002] Matching received signals from non-cooperative satellites with their transmitting satellites is fundamental to navigation and positioning using non-cooperative satellites. There are two main methods for satellite signal matching and identification: one is based on spatial location information, and the other is based on the structural characteristics of the satellite signals. These two methods utilize the spatiotemporal information and electromagnetic characteristics of the satellite signals, respectively. However, in non-cooperative navigation and positioning scenarios, due to the lack of accurate receiver position information, it is difficult to extract spatiotemporal information from the signals using only Doppler measurements. Furthermore, it is difficult to collect a large number of non-cooperative satellite signals, severely impacting the extraction and analysis of electromagnetic characteristics. It should also be noted that the difficulty of satellite identification varies across different constellations. For example, if spatiotemporal information is used for identification, Iridium and Orbcomm satellites, due to their small number of satellites, require less precise receiver position accuracy and can be identified by directly comparing the approximate receiver position and received Doppler frequencies with possible satellites. However, for the Starlink constellation, with its large number of satellites and the ambiguity of transmitted carrier frequencies, identification is difficult.

[0003] Existing technical solutions:

[0004] Nanjing University of Aeronautics and Astronautics disclosed a "method for identifying non-cooperative navigation satellites based on Doppler frequency shift characteristics" in its patent application (patent application number: CN201510368579.1, application publication number: CN105044744A). The implementation steps are as follows: single-carrier signals of different frequencies and their Doppler frequency shift measurements are obtained through codeless carrier recovery; prior Doppler frequency shift values ​​are calculated based on the initial orbital parameters of the low-Earth orbit user satellite and the periodically annotated navigation ephemeris; the measured Doppler frequency shift values ​​containing receiver frequency differences and the prior values ​​are arranged into two groups of samples in descending order, and a bisection method is used to efficiently find the frequency difference estimate to achieve a consistent correspondence between the two groups of frequencies; and based on the proportion of each Doppler frequency shift matching residual, a global matching and local elimination method is used to improve the global reliability of satellite identification. This method can accurately identify navigation satellites and accurately estimate receiver frequency difference; however, since it relies on the precise position of the receiver as a prerequisite when calculating the prior Doppler frequency shift, it cannot be applied to non-cooperative satellite navigation and positioning scenarios where the receiver position is unknown.

[0005] Xi'an University of Electronic Science and Technology disclosed a method for identifying non-cooperative low-Earth orbit mega-constellation satellites in its patent application (patent application number: CN202410291259.X, application publication number: CN118191881A). The implementation steps are as follows: Constructing a set of relative position vectors of the nearest and second-nearest satellites, and a set of vector angles between target satellites; filtering relative position vectors in the satellite orbit database whose relative position vector magnitudes between any two satellites are less than a set threshold; constructing a set of vector angles to be matched for each target satellite based on the filtering results, and performing correlation matching with the vector angles of the target satellites; verifying the correlation matching results using a matching check operator based on internal rotation coordinate transformation to determine the satellite number, thus completing the identification of non-cooperative satellites. This method achieves identity matching by comparing the relative position vectors and the angle between the vectors of satellites. The idea is simple and easy to implement. However, it does not consider the carrier frequency offset and cannot cope with the frequency hopping characteristics of Starlink. At the same time, its process based on batch angle screening is difficult to execute in real time on dynamic platforms. In addition, in the dense Starlink layout, the relative vectors and angles of multiple satellites are highly similar, which poses a risk of matching ambiguity.

[0006] In summary, existing methods for identifying non-cooperative low-Earth orbit (LEO) satellites mainly fall into two categories: one is satellite signal matching based on spatial location information, which requires knowledge of the receiver's position; the other is signal matching and identification based on the structural characteristics of the satellite signals. The Nanjing University of Aeronautics and Astronautics (NUAA) scheme utilizes uncoded carrier waves to recover Doppler measurements and performs bidirectional sorting and residual elimination with the prior frequency shift of the navigation satellites. While achieving high accuracy, it is heavily dependent on the receiver's position, and the computational load increases dramatically with the constellation size. The Xidian University (XDU) scheme filters candidates using the angle between relative position vectors and verifies them using an internal rotation transformation. It does not require the receiver's position but does not consider the carrier frequency shift of non-cooperative starlinks, which can easily lead to multiple interpretations in dense starlink scenarios. Summary of the Invention

[0007] The purpose of this invention is to address the shortcomings of the prior art by providing a non-cooperative Starlink signal matching and identification method that supports real-time matching and identification in dynamic scenes, can cope with the uncertainty of Starlink satellite carrier frequency, and has high reliability by combining known satellite constellation signal measurements.

[0008] This invention is a method for matching and identifying non-cooperative Starlink signals by combining signal measurements from known satellite constellations. Its key feature is that it utilizes the known identities of satellites and their intercepted signals to assist in matching and identifying non-cooperative Starlink satellite signals, comprising the following steps:

[0009] Step 1: Establish a measurement dataset: The receiver detects and collects signals from known satellite constellations and non-cooperative Starlink satellites in real time, and establishes a measurement dataset for known satellite constellation signals and non-cooperative Starlink satellite signals;

[0010] Step 2: Construct a set of receiver velocity equations: Estimate the approximate location of the receiver, and use the state vectors of the known satellite constellation of M satellites and the frequencies of the known satellite constellation signals received by the receiver to establish a set of receiver velocity equations for auxiliary calculation.

[0011] Step 3: Generate a set of non-cooperative Starlink signal matching test equations: For the n non-cooperative Starlink signals to be matched and the signal frequencies received by their receivers, combined with the approximate location of the receivers, the state vectors of the N non-cooperative Starlink satellites, and the carrier frequency ambiguity assumptions of the non-cooperative Starlink satellites, construct the signal-satellite-carrier frequency combination relationship one by one, construct the non-cooperative Starlink signal matching test equations under each combination, and summarize them to generate a set of non-cooperative Starlink signal matching test equations.

[0012] Step 4: Solve the system of equations simultaneously to generate the receiver velocity candidate set RES: Traverse each non-cooperative Starlink signal matching test equation and solve it simultaneously with the receiver velocity solution equations to obtain a set of receiver velocity solutions to be screened, forming the receiver velocity candidate set RES.

[0013] Step 5: Filter the receiver velocity candidate set RES and generate the effective velocity solution set. Thresholds are selected based on the approximate speed of the receiver and the speed difference. Error judgment is performed on all solutions in the receiver velocity candidate set RES, and solutions whose difference from the approximate velocity is less than a set threshold are retained to form the receiver effective velocity solution set.

[0014] Step 6: Determine the probability velocity V p : Effective velocity set of the receiver Histograms were plotted for the velocity components along the X, Y, and Z axes. The frequency distribution of each axis was statistically analyzed, and the center velocity value corresponding to the mode with the highest frequency was selected. The three center velocity values ​​were combined to form the probability velocity V of the receiver. p ;

[0015] Step 7: Obtain the non-cooperative Starlink signal matching and identification results: For each of the n non-cooperative Starlink signals to be matched, in the receiver's effective velocity solution set, use the Euclidean distance to measure the relationship between each velocity solution and the probability velocity V. p The difference between the two is used to select the velocity solution with the smallest distance. The signal-satellite-frequency combination corresponding to this velocity solution is the matching and identification result of the non-cooperative starlink signal.

[0016] This invention solves both the problem of non-cooperative Starlink signal matching and identification when the receiver location is unknown and frame synchronization and demodulation information cannot be relied upon, and the technical problems of Starlink carrier frequency ambiguity and difficulty in real-time identification in dynamic scenarios.

[0017] Compared with the prior art, the present invention has the following advantages:

[0018] Achieving matching and identification under the condition of carrier frequency ambiguity of non-cooperative Starlink satellites: This invention first enumerates all "signal-satellite-carrier frequency" combination assumptions, constructs and solves the joint equations by least squares for each set of assumptions, and generates a candidate set of receiver velocities; then, it uses the inertial navigation approximate velocity and screening threshold to eliminate pseudo-solutions with excessive deviations, and performs histogram statistics on the remaining solution set to extract the velocity mode and select the combination with the smallest residual, thus achieving high-reliability matching under the condition of carrier frequency uncertainty.

[0019] Real-time matching and identification on dynamic platforms: This invention can reconstruct the receiver velocity calculation and matching verification equations at each observation moment, perform least squares solution in parallel, and then immediately perform threshold screening and histogram mode statistics. Without accumulating multiple frames of data, velocity estimation and signal-satellite matching can be completed within seconds, meeting the real-time identification requirements of high-speed maneuvering platforms for non-cooperative Starlink signals.

[0020] The computational complexity is linearly controllable and easy to parallelize: In this invention, the number of equations only increases linearly with the number of known satellites, the number of signals to be matched, and the number of carrier frequency assumptions, avoiding the quadratic explosion of traditional pairwise matching; the core least squares solution and histogram statistics steps can be decomposed into independent tasks and executed concurrently, which can run efficiently in parallel, ensuring predictable computational overhead and low latency response in a large-scale Starlink environment.

[0021] High reliability, independent of receiver's actual position: When the global navigation satellite system is unavailable, the invention can use an inertial navigation system to calculate the approximate position of the receiver at time t in real time to replace the receiver's actual position, thereby improving reliability and enabling Starlink signal matching and identification when the receiver's state is unknown. Attached Figure Description

[0022] Figure 1 This is a flowchart of the present invention;

[0023] Figure 2 This is a flowchart illustrating the technical solution of the present invention. Detailed Implementation

[0024] To help better understand the present invention, the technical solution of the present invention will be clearly and completely described below with reference to the accompanying drawings.

[0025] With the increasing global demand for communication and navigation, positioning using low-Earth orbit (LEO) satellite signals has become an important supplement. However, the unknown signal formats of non-cooperative constellations such as Starlink, and the unavailability of receiver position and velocity, make it difficult for traditional synchronous demodulation and differential filtering methods to achieve real-time and reliable identity matching. To address this challenge, this invention proposes a non-cooperative Starlink signal matching and identification method that combines measurements of known satellite constellation signals. This method requires no frame synchronization or prior position information and can efficiently and robustly identify non-cooperative Starlink signals in dynamic environments.

[0026] Example 1:

[0027] This invention is a method for matching and identifying non-cooperative Starlink signals by combining signal measurements from known satellite constellations. This invention utilizes the known identities of satellites and their intercepted signals to assist in matching and identifying non-cooperative Starlink satellite signals. See [link to related documentation]. Figure 1 , Figure 1 This is a flowchart of the present invention, which includes the following steps:

[0028] Step 1: Establish a measurement dataset: The receiver detects and collects known satellite constellation signals and non-cooperative Starlink signals in real time, and establishes a measurement dataset of known (such as Iridium, Orbcomm, etc.) satellite constellation signals and non-cooperative Starlink satellite signals. Here, measurement refers to the frequency of the received signal.

[0029] Step 2: Construct the receiver velocity equation set: The approximate position of the receiver is estimated by the inertial navigation system to replace the actual position of the receiver. Using the state vectors of the known satellite constellation of M satellites and the frequency of the known satellite constellation signals received by the receiver, a set of receiver velocity equations is established to assist in the calculation. This set of equations can also be called the constant velocity equations.

[0030] Step 3: Generate a set of non-cooperative Starlink signal matching test equations: For the n non-cooperative Starlink signals to be matched and the signal frequencies received by their receivers, combined with the approximate location of the receivers, the state vectors of the N non-cooperative Starlink satellites, and the carrier frequency ambiguity assumptions of the non-cooperative Starlink satellites, construct signal-satellite-carrier frequency combination relationships one by one, construct non-cooperative Starlink signal matching test equations under each combination, and summarize them to generate a set of non-cooperative Starlink signal matching test equations, which can also be called a test equation set.

[0031] Step 4: Solve the simultaneous equations to generate the receiver velocity candidate set RES: Traverse each non-cooperative starlink signal match test equation and solve the simultaneous equations of each non-cooperative starlink signal match test equation and the receiver velocity solution equations to obtain a set of receiver velocity solutions to be screened, forming the receiver velocity candidate set RES, which can also be called the test result set. Since different combinations of assumptions have varying degrees of error compared to the actual receiver velocity, the velocity distribution range in this candidate set is large, and the accuracy is low. This receiver velocity candidate set RES needs further screening in subsequent steps.

[0032] Step 5: Filter the receiver velocity candidate set RES and generate the effective velocity solution set. Based on the inertial navigation system, the approximate speed of the receiver is estimated, and the speed difference is used to select a threshold. Error judgment is performed on all solutions in the receiver velocity candidate set RES, and solutions whose difference from the approximate velocity is less than a set threshold are retained to form the receiver effective velocity solution set. This solution set contains receiver velocity solutions for matching satellites of the signal to be identified, providing reliable data support for subsequent probabilistic velocity estimation and signal matching. The value is selected based on the actual speed of the platform.

[0033] Step 6: Determine the probability velocity V p : Effective velocity set of the receiver Histograms were plotted for the velocity components along the X, Y, and Z axes. The frequency distribution of each axis was statistically analyzed, and the center velocity value corresponding to the mode with the highest frequency was selected. The three center velocity values ​​were combined to form the probability velocity V of the receiver. p .

[0034] Step 7: Obtain the non-cooperative Starlink signal matching and identification results: For each of the n non-cooperative Starlink signals to be matched, in the receiver's effective velocity solution set, use the Euclidean distance to measure the relationship between each velocity solution and the probability velocity V. p The difference between the two is used to select the velocity solution with the smallest distance. The signal-satellite-frequency combination corresponding to this velocity solution is the matching and identification result of the non-cooperative starlink signal.

[0035] This invention presents a comprehensive technical solution. It constructs a receiver velocity calculation equation set based on measurements of a known satellite constellation, generates a signal-satellite-carrier frequency matching verification equation set by combining carrier frequency offset assumptions, and solves these equation sets jointly using the least squares method to generate a velocity candidate set. Through multi-level threshold filtering and histogram mode extraction, the probabilistic velocity of the receiver is quickly obtained. Using this probabilistic velocity as a benchmark, the velocity solution closest to each signal to be matched is identified, thus completing the identification of non-cooperative Starlink signals. This invention solves the technical problem of achieving non-cooperative satellite signal matching under conditions where the receiver's prior position is unknown and frame synchronization and demodulation information cannot be relied upon. It also addresses the technical problems of existing methods' inability to handle inaccurate matching caused by Starlink signal carrier frequency ambiguity and their inability to complete real-time matching identification in dynamic scenarios.

[0036] Example 2:

[0037] The non-cooperative Starlink signal matching and identification method based on known satellite constellation signal measurements is the same as in Example 1. The construction of the receiver velocity solution equation set in step (2) of this invention is described in [reference needed]. Figure 2 , Figure 2 This is a flowchart illustrating the technical solution of the present invention, which includes the following steps:

[0038] 2a) Approximate location of the receiver r r Estimate: Using a geocentric coordinate system, when global satellite navigation systems are unavailable, the approximate position of the receiver at time t is calculated in real time through an inertial navigation system, which will have a certain error compared to the actual position.

[0039] 2b) Known satellite signal reception: At the same time t, the receiver receives signals from M known satellites, numbered q = 1, 2, ..., M. Since there are 3 unknowns in the solution of the receiver speed, at least 3 equations are required to solve it. Therefore, M is greater than or equal to 2.

[0040] 2c) Obtaining the state vectors of known satellites: In the geocentric coordinate system, import the two-line element TLE file of the M known satellites into the simplified conventional perturbation model SGP4 to obtain the state vectors of the M known satellites at time t, including the position vector and the velocity vector.

[0041] 2d) Constructing the velocity solution equation system: Using the state vectors of the known constellation of M satellites and the frequency of the known satellite constellation signals received by the receiver, for the q-th known satellite, f r,q f is the frequency at which the receiver receives the q-th known satellite. 0,q It is the carrier frequency of the q-th known satellite, which can be found in the published satellite documents, based on the approximate location r of the receiver. r The state vector of the q-th known satellite Construct the receiver speed calculation equation:

[0042]

[0043] Where c is the speed of light constant. The receiver speed is the value to be solved. The above M equations are unified into a set of receiver speed solution equations to assist in solving for the receiver speed.

[0044] This invention constructs a set of receiver velocity equations by introducing the state vectors of M known satellites and their received signal frequencies, combined with the approximate location of the receiver. Compared to traditional methods that rely on external ephemeris or precise location, this invention can quickly generate a set of velocity equations using only readily available TLE data and local frequency measurements. Subsequently, these equations are solved in parallel with the matching verification equations for non-cooperative Starlink signals, effectively providing a stable and accurate velocity reference for subsequent signal-satellite matching.

[0045] Example 3:

[0046] The non-cooperative Starlink signal matching identification method based on known satellite constellation signal measurements is the same as in Examples 1-2. Step (3) of generating the non-cooperative Starlink signal matching verification equation set includes the following steps:

[0047] 3a) Receiving signals from non-cooperative Starlink satellites: At time t, the receiver simultaneously detects signals emitted by N non-cooperative Starlink satellites. The observed non-cooperative Starlink satellites are numbered j = 1, 2, ..., N.

[0048] 3b) Selecting satellite signals to be matched: Select n satellite signals to be matched from the above N non-cooperative Starlink satellite signals, and denot them as i = 1, 2, ..., n, f r,i It is the receiving frequency of the i-th satellite signal to be matched.

[0049] 3c) Obtain the state vectors of non-cooperative Starlink satellites: In the geocentric-ground-fixed coordinate system, import the TLE files of the N non-cooperative Starlink satellites into the SGP4 model to obtain the position vectors and velocity vectors of each non-cooperative Starlink satellite at time t.

[0050] 3d) Constructing the signal-satellite-carrier frequency hypothetical combination: Since it is unknown which satellite and which carrier frequency the signal to be matched was transmitted from, the combination needs to be traversed; the signal to be matched i = 1, 2, ..., n, the non-cooperative Starlink satellite j = 1, 2, ..., N, and the hypothetical carrier frequency k = 1, 2, ..., K are used to construct the signal-satellite-carrier frequency hypothetical combination.

[0051] 3e) Construct a set of equations for matching non-cooperative Starlink signals: For each signal i to be matched, each candidate satellite j, and each carrier frequency offset, assuming k = 1, 2, ..., K, based on the approximate receiver position r r The state vector of the j-th non-cooperative Starlink satellite The receiver measures the frequency f of the satellite to be matched. r,i and the transmission frequency (f0+Δf) under the kth carrier frequency offset assumption k Construct a matching test equation;

[0052]

[0053] Where c is the speed of light constant. The receiver velocity to be solved is f0, which is the nominal carrier frequency of the non-cooperative Starlink satellite, and Δf is... k For the k-th carrier frequency offset assumption, f0 and Δf k It can be estimated from publicly available satellite documents. The equations under the combination of all satellite signals i = 1, 2, ..., n to be matched, non-cooperative Starlink satellites j = 1, 2, ..., N, and assumed carrier frequencies k = 1, 2, ..., K are summarized to form a set of non-cooperative Starlink signal matching verification equations for joint solution.

[0054] To achieve matching and identification of non-cooperative Starlink signals, this invention traverses all signal-satellite-carrier frequency ternary hypothesis combinations and constructs a complete set of matching verification equations based on the approximate receiver location, the TLE / SGP4 state vector of each Starlink satellite, and the received frequency and assumed transmit frequency. This process is entirely based on existing navigation and orbit estimation standards, ensuring that full-coverage hypothesis testing is completed with controllable computational overhead in real-world environments with carrier frequency ambiguity. This provides a comprehensive verification foundation compatible with all possibilities for subsequent joint solutions to receiver speed and signal identification.

[0055] Example 4:

[0056] The non-cooperative Starlink signal matching and identification method based on known satellite constellation signal measurements is the same as in Examples 1-3. Step (4) involves solving the simultaneous equations to generate the receiver velocity candidate set RES, which includes the following steps:

[0057] 4a) Constructing a joint equation set: For each signal i–satellite j–carrier frequency k assumed combination, the receiver velocity solution equation set and the matching test equation corresponding to the combination are merged to form a system of equations about the receiver velocity vector. A system of linear equations.

[0058] 4b) Least Squares Solution: Apply the least squares method to the linear equations for each of the above hypothesis combinations to obtain the candidate solutions for the receiver velocity under the corresponding hypothesis combination (i,j,k).

[0059] 4c) Candidate set storage: Save all obtained candidate sets The receiver velocity candidate set is stored sequentially according to the assumed order for subsequent screening.

[0060] This invention addresses each signal-satellite-carrier frequency hypothesis combination by merging the receiver velocity calculation equations and matching verification equations into a linear system. Least squares are then used to solve this system, generating and storing a candidate set of receiver velocities under all hypotheses. This process eliminates the need for batch or offline computation and can be solved in parallel on multi-core or hardware-accelerated platforms, providing a comprehensive and high-performance candidate velocity set for subsequent threshold filtering and signal matching.

[0061] Example 5:

[0062] The non-cooperative Starlink signal matching and identification method based on known satellite constellation signal measurements is the same as in Examples 1-4. Step (5) involves filtering the receiver velocity candidate set RES to generate an effective velocity solution set. The steps include the following:

[0063] 5a) Setting a filtering threshold: Assuming the satellite signal i to be matched is transmitted by Starlink satellite j, the receiver velocities calculated under various assumed carrier frequencies are close and have little error compared to the actual receiver velocities. Conversely, if the satellite signal i to be matched is not transmitted by Starlink satellite j, the calculated receiver velocities will have a very large error compared to the actual receiver velocities, making them more likely to be filtered out. The approximate receiver velocity v at time t is obtained in real-time through the inertial navigation system. rec Based on the permissible speed error level in practical applications, a speed difference screening threshold is set.

[0064] 5b) Filtering the receiver speed candidate set RES: Filtering thresholds based on the speed differences in 5a). exist Select the receiver velocity candidate set RES under the given conditions.

[0065] 5c) Obtain the effective velocity solution set The selected combination of hypotheses that satisfies the screening criteria is retained in the receiver velocity candidate set RES, and this is denoted as the effective velocity solution set.

[0066] This invention utilizes the approximate velocity calculated by inertial navigation and a preset error threshold to filter the receiver velocity candidate set generated in step 4, retaining the hypothetical combinations with errors within the threshold, forming a highly reliable effective velocity solution set. This significantly reduces the burden of subsequent histogram statistics and matching calculations, providing reliable data support for subsequent probabilistic velocity calculations and signal matching identification.

[0067] Example 6:

[0068] The non-cooperative Starlink signal matching and identification method based on known satellite constellation signal measurements is the same as in Examples 1-5, except that step (6) involves determining the probability velocity V. p It includes the following steps:

[0069] 6a) Frequency statistics of velocity components: for 5c) effective velocity solution set Histograms were used to count the frequency of each velocity component along the X, Y, and Z axes.

[0070] 6b) Determine the probability velocity V p Take the center velocity value corresponding to the highest frequency mode from each of the three histograms, and combine these three modes to obtain the receiver's probabilistic velocity V. p .

[0071] This invention creates histograms for the X, Y, and Z components of the effective velocity solution set, counts their frequency of occurrence, and extracts the mode (center value of the highest frequency box) of each component. These three components are then combined to form the receiver's "probabilistic velocity," achieving an innovative screening strategy that automatically locks the most reliable velocity solution in a "frequency-first" manner.

[0072] Example 7:

[0073] The method for identifying non-cooperative Starlink signals by combining known satellite constellation signal measurements is the same as in Example 1, step 7) of obtaining the non-cooperative Starlink signal matching identification results, and its implementation is as follows:

[0074] 7a) Error Calculation: For each satellite signal i to be identified, calculate... The magnitude of the error between probability and velocity.

[0075] 7b) Obtain matching and identification results: Traverse the satellite number j corresponding to the minimum velocity modulus and save the matching and identification results.

[0076] This invention uses probabilistic velocity to sort the Euclidean distances of each candidate solution and selects the satellite number with the smallest distance as the final matching result for the signal, thus ensuring the accuracy of non-cooperative Starlink signal identification.

[0077] This invention is a non-cooperative Starlink signal matching and identification method that combines known satellite constellation signal measurements. It is used to solve problems in low-Earth orbit satellite navigation and positioning scenarios, such as the failure of pseudorange / carrier phase matching due to unknown receiver position and velocity, the difficulty of synchronous demodulation and decoding due to unknown non-cooperative signal format, the poor real-time performance and large computational load of filtering algorithms in high dynamic environments, and the surge in computational load caused by the explosive increase in the assumed combination of visible Starlink constellation satellite carrier frequencies. This invention detects and collects various low-Earth orbit satellite signals in real time, establishing a signal measurement dataset. It constructs a velocity-determining equation using M known constellations (e.g., Iridium, Orbcomm constellations) and their corresponding signal measurements. It then iterates through Starlink satellite signal measurements and the satellites to be matched, constructing a test equation set. Each test equation and the velocity-determining equation in the test equation set are solved simultaneously to construct a test result set. The test result set is filtered based on the receiver's approximate velocity range, retaining data with approximate velocities less than the filtering threshold, resulting in a result dataset. The midpoint of the densest region in the filtered result dataset is calculated as the probabilistic velocity. Finally, the test result set is searched for the satellite signal measurement closest to the probabilistic velocity, achieving non-cooperative Starlink signal matching and identification. This invention offers advantages such as high flexibility, good real-time performance, wide applicability, and strong parallel processing capabilities.

[0078] Example 8:

[0079] The non-cooperative Starlink signal matching and identification method, which combines known satellite constellation signal measurements, is the same as in Examples 1-7. The technical effects of this invention are verified through simulation below. In the MATLAB simulation experiment, the approximate position of the receiver at time t is set to [-1856080, 5152630, 3345530], and the approximate velocity is set to [10, 10, 10]. The field of view includes M = 2 known reference satellites and N = 146 non-cooperative Starlink satellites, of which 10 Starlink signals are randomly selected as targets to be identified. The carrier centers of the known satellites are f... 0,1 =1616MHz, f 0,2 =150MHz Starlink carrier center f0=12.95GHz, carrier frequency offset Δf=43945.5Hz, a total of K=20 assumptions are enumerated; speed screening threshold. =50m / s. Experimental results show that 98.6% of the 10×146×20 hypotheses were eliminated by the screening threshold, leaving approximately 400 groups, or about 20 satellites, awaiting further, more precise matching. After histogram mode extraction, the average deviation between the probability velocity and the actual velocity is approximately 2m / s, and all 10 test starlink signals can be correctly matched, as shown in Table 1. The MATLAB serial execution time is approximately 5 seconds, but the solution tasks are independent of each other. If deployed in parallel on hardware platforms such as GPUs or FPGAs, the computational latency can be reduced to the level of hundreds of milliseconds or even tens of milliseconds, fully meeting the real-time recognition requirements of dynamic platforms.

[0080] Table 1: Randomly Selected Signals to be Identified and Identification Results

[0081] Identification signal 110 130 105 58 133 36 75 132 37 35 Recognition results 110 130 105 58 133 36 75 132 37 35

[0082] Based on the above analysis and simulation verification, this invention proposes a non-cooperative Starlink signal matching and identification method that combines known satellite constellation signal measurements. This method can achieve highly reliable signal matching and identification even when the Starlink satellite carrier frequency is unclear and the actual location of the receiver is unknown. It also maintains linearly controllable computational load and has the ability to identify signals in real time at the second level.

[0083] In summary, this invention is a non-cooperative Starlink signal matching and identification method that combines known satellite constellation signal measurements. This invention solves both the problem of non-cooperative Starlink signal matching and identification when the receiver position is unknown and frame synchronization and demodulation information cannot be relied upon, and the technical problems of Starlink carrier frequency ambiguity and difficulty in real-time identification in dynamic scenarios. The implementation steps include: establishing a measurement dataset; constructing a receiver velocity calculation equation set; generating a non-cooperative Starlink signal matching verification equation set; solving the equations simultaneously to generate a receiver velocity candidate set RES; filtering the receiver velocity candidate set RES; and determining the probabilistic velocity V. p This invention obtains matching and identification results for non-cooperative Starlink signals. It utilizes known-identity satellites and their intercepted signals to assist in the matching and identification of non-cooperative Starlink signals. This enables real-time matching and identification of moving platforms under conditions of carrier frequency ambiguity for non-cooperative Starlink satellites, with linearly controllable computational complexity, independent of the receiver's actual position, and high reliability. This invention can be used for receiver navigation and positioning based on non-cooperative Starlink signals.

Claims

1. A method for matching and identifying non-cooperative Starlink signals by combining known satellite constellation signal measurements, characterized in that: The matching and identification of non-cooperative Starlink satellite signals using known satellites and their intercepted signals includes the following steps: Step 1: Establish a measurement dataset: The receiver detects and collects signals from known satellite constellations and non-cooperative Starlink satellites in real time, and establishes a measurement dataset for known satellite constellation signals and non-cooperative Starlink satellite signals; Step 2: Construct a set of receiver velocity equations: Estimate the approximate location of the receiver, and use the state vectors of the known satellite constellation of M satellites and the frequencies of the known satellite constellation signals received by the receiver to establish a set of receiver velocity equations for auxiliary calculation. Step 3: Generate a set of non-cooperative Starlink signal matching test equations: For the n non-cooperative Starlink signals to be matched and the signal frequencies received by their receivers, combine the approximate location of the receiver, the state vectors of the N non-cooperative Starlink satellites, and the carrier frequency ambiguity assumptions of the non-cooperative Starlink satellites, construct the signal-satellite-carrier frequency assumption combination relationship one by one, construct the non-cooperative Starlink signal matching test equations under each combination, and summarize them to generate a set of non-cooperative Starlink signal matching test equations; Step 4: Solve the system of equations simultaneously to generate the receiver velocity candidate set RES: Traverse each non-cooperative Starlink signal matching test equation and solve it simultaneously with the receiver velocity solution equations to obtain a set of receiver velocity solutions to be screened, forming the receiver velocity candidate set RES. Step 5: Filter the receiver velocity candidate set RES and generate the effective velocity solution set. Thresholds are selected based on the approximate speed of the receiver and the speed difference. Error judgment is performed on all solutions in the receiver velocity candidate set RES, and solutions whose difference from the approximate velocity is less than a set threshold are retained to form the receiver effective velocity solution set. Step 6: Determine the probability velocity V p : Effective velocity set of the receiver Histograms were plotted for the velocity components along the X, Y, and Z axes. The frequency distribution of each axis was statistically analyzed, and the center velocity value corresponding to the mode with the highest frequency was selected. The three center velocity values ​​were combined to form the probability velocity V of the receiver. p ; Step 7: Obtain the non-cooperative Starlink signal matching and identification results: For each of the n non-cooperative Starlink signals to be matched, in the receiver's effective velocity solution set, use the Euclidean distance to measure the relationship between each velocity solution and the probability velocity V. p The difference between the two is used to select the velocity solution with the smallest distance. The signal-satellite-frequency combination corresponding to this velocity solution is the matching and identification result of the non-cooperative starlink signal.

2. The non-cooperative Starlink signal matching and identification method based on known satellite constellation signal measurements according to claim 1, characterized in that, Step (2) involves constructing the receiver velocity solution equation set, which includes the following steps: 2a) Approximate location of the receiver r r Estimation: Using a geocentric coordinate system, the approximate position of the receiver at time t is calculated in real time through an inertial navigation system, which has a certain error compared with the actual position; 2b) Known satellite signal reception: At the same time t, the receiver receives signals from M known satellites, with satellite numbers q = 1, 2, ..., M, where M is greater than or equal to 2; 2c) Obtaining the known satellite state vector: Import the two-line element TLE file of the M known satellites into the simplified conventional perturbation model SGP4 to obtain the state vector of the known satellites at time t, including the position vector and velocity vector; 2d) Constructing the velocity solution equation system: Using the state vectors of the known constellation of M satellites and the frequency of the known satellite constellation signals received by the receiver, for the q-th known satellite, f r,q f is the frequency at which the receiver receives the q-th known satellite. 0,q It is the carrier frequency of the q-th known satellite, based on the approximate position r of the receiver. r The state vector of the q-th known satellite Construct the receiver speed calculation equation: Where c is the speed of light constant. The receiver speed is the value to be solved. The above M equations are unified into a set of receiver speed solution equations to assist in solving for the receiver speed.

3. The non-cooperative Starlink signal matching and identification method based on known satellite constellation signal measurements according to claim 1, characterized in that, Step (3) of generating the non-cooperative Starlink signal matching test equation set includes the following steps: 3a) Receiving signals from non-cooperative Starlink satellites: At time t, the receiver simultaneously detects signals emitted by N non-cooperative Starlink satellites. The observed non-cooperative Starlink satellites are numbered j = 1, 2, ..., N. 3b) Selecting satellite signals to be matched: Select n signals to be matched from the above N non-cooperative Starlink satellite signals, and number them as i = 1, 2, ..., n, f r,i It is the receiving frequency of the i-th satellite signal to be matched; 3c) Obtain the state vectors of non-cooperative Starlink satellites: Import the TLE files of the N non-cooperative Starlink satellites into the SGP4 model to obtain the position and velocity vectors of each non-cooperative Starlink satellite at time t; 3d) Constructing the signal-satellite-carrier frequency hypothesis combination: The signal to be matched i = 1, 2, ..., n, the non-cooperative Starlink satellites j = 1, 2, ..., N, and the hypothetical carrier frequency k = 1, 2, ..., K are used to construct the signal-satellite-carrier frequency hypothesis combination. 3e) Construct a set of equations for matching non-cooperative Starlink signals: For each signal i to be matched, each candidate satellite j, and each carrier frequency offset, assuming k = 1, 2, ..., K, based on the approximate receiver position r r The state vector of the j-th non-cooperative Starlink satellite The receiver measures the frequency f of the satellite to be matched. r,i and the transmission frequency (f0+Δf) under the kth carrier frequency offset assumption k Construct a matching test equation; Where c is the speed of light constant. The receiver velocity to be solved is f0, which is the nominal carrier frequency of the non-cooperative Starlink satellite, and Δf is... k For the k-th carrier frequency offset hypothesis, the equations under all combinations of signals to be matched i = 1, 2, ..., n, non-cooperative Starlink satellites j = 1, 2, ..., N, and assumed carrier frequencies k = 1, 2, ..., K are summarized to form a set of non-cooperative Starlink signal matching verification equations for joint solution.

4. The non-cooperative Starlink signal matching and identification method based on known satellite constellation signal measurements according to claim 1, characterized in that, Step (4) involves solving the simultaneous equations to generate the receiver velocity candidate set RES, which includes the following steps: 4a) Constructing a joint equation set: For each signal i–satellite j–carrier frequency k assumed combination, the receiver velocity solution equation set and the matching test equation corresponding to the combination are merged to form a system of equations about the receiver velocity vector. A system of linear equations; 4b) Least Squares Solution: Apply the least squares method to the linear equations for each of the above hypothesis combinations to obtain the candidate solutions for the receiver velocity under the corresponding hypothesis combination (i,j,k). 4c) Candidate set storage: Save all obtained candidate sets The receiver velocity candidate set is stored sequentially according to the assumed order for subsequent screening.

5. The non-cooperative Starlink signal matching and identification method based on known satellite constellation signal measurements according to claim 1, characterized in that, Step (5) involves filtering the receiver velocity candidate set RES to generate an effective velocity solution set. The steps include the following: 5a) Setting the filtering threshold: The approximate velocity v of the receiver at time t is obtained in real time through the inertial navigation system. rec Based on the permissible speed error level in practical applications, a speed difference screening threshold is set. 5b) Filtering the receiver speed candidate set RES: Filtering thresholds based on the speed differences in 5a). exist Selecting the receiver velocity candidate set RES under certain conditions; 5c) Obtain the effective velocity solution set The selected combination of hypotheses that satisfies the screening criteria is retained in the receiver velocity candidate set RES, and this is denoted as the effective velocity solution set.

6. The non-cooperative Starlink signal matching and identification method based on known satellite constellation signal measurements according to claim 1, characterized in that, The determination of probability velocity V in step (6) p It includes the following steps: 6a) Frequency statistics of velocity components: for 5b) effective velocity solution set Histograms were used to count the frequency of each velocity component along the X, Y, and Z axes respectively. 6b) Determine the probability velocity V p Take the center velocity value corresponding to the highest frequency mode from each of the three histograms, and combine these three modes to obtain the receiver's probabilistic velocity V. p .

Images