Low earth orbit satellite gnss anti-jam adaptive ra im method and system

Abstract

This invention provides a method and system for adaptive RAIM (Rapid Interference Detection and Interference Assist) for low-Earth orbit (LEO) GNSS satellites. It requires no hardware modification and acquires data synchronously via a primary and backup dual-machine setup. It employs incremental statistical algorithms to achieve multi-feature dynamic threshold early warning; performs fine-grained detection of cycle slips, multi-frequency variations, and Doppler consistency in the observation domain; isolates faulty satellites using sliding window adaptive RAIM; introduces rigid constraints on orbital dynamics to ensure solution integrity; performs tiered handling based on comprehensive interference confidence; and autonomously switches to orbit extrapolation mode for seamless overflight in the event of severe interference, while also possessing global baseline adaptive calibration capabilities. The advantages of this invention include low resource consumption, rapid response, long-term stable operation, and onboard autonomous closed-loop handling capabilities, effectively ensuring the navigation safety of LEO satellites in complex electromagnetic environments.

Description

Technical Field

[0001] This invention belongs to the field of satellite navigation and anti-interference technology, and in particular relates to a full-link anti-interference and adaptive RAIM (Receiver Autonomous Integrity Monitoring) method and system for low-Earth orbit satellite GNSS (Global Navigation Satellite System) receivers. Background Technology

[0002] Low Earth Orbit (LEO) satellites, with their advantages of low orbital altitude, high speed, and wide coverage, are increasingly widely used in communications, remote sensing, and navigation enhancement. As the core navigation sensor for LEO satellites, the positioning accuracy and anti-jamming capability of GNSS receivers directly affect the safety and reliability of the satellite's on-orbit operation. However, the electromagnetic environment in space where LEO satellites operate is complex, facing two main types of core interference threats: first, suppression interference, which uses high-power radio frequency signals to overwhelm useful signals, causing a sharp drop in the receiver's carrier-to-noise ratio and loss of satellite lock; second, deceptive interference, which uses forged signals to induce the receiver to output incorrect position and time information, possessing extremely high stealth capabilities.

[0003] Currently, anti-jamming technologies for low-Earth orbit (LEO) GNSS receivers suffer from the following problems: First, fixed detection thresholds and poor dynamic adaptability make them unable to adapt to the time-varying characteristics of highly dynamic and electromagnetic environments, resulting in high false alarm and false detection rates. Second, high hardware dependence, with mainstream solutions focusing on RF / baseband domain modifications, leading to high costs and the inability to upgrade the software of already launched satellites. Third, a lack of autonomous closed-loop processing capabilities, relying solely on detection and alarm functions, excessively depending on ground intervention, posing security risks during satellite-to-ground communication outages. Fourth, insufficient long-term stability, lack of baseline adaptive calibration, and performance degradation due to hardware aging. Finally, weak anti-spoofing capabilities, lacking specific detection of features such as clock bias forgery and orbital dynamic constraints, making it difficult to effectively cope with complex interference scenarios. Therefore, an anti-jamming solution with autonomous closed-loop processing capabilities and long-term stability is urgently needed. Summary of the Invention

[0004] In view of this, the present invention aims to propose a low-orbit satellite GNSS anti-interference adaptive RAIM method and system to solve at least one of the problems existing in the above-mentioned prior art.

[0005] To achieve the above objectives, the technical solution of the present invention is implemented as follows: In a first aspect, the present invention provides a low-Earth orbit satellite GNSS anti-interference adaptive RAIM method. The method is executed by a software program running on an onboard embedded processor, without requiring modification of the receiver baseband or RF hardware. The method includes the following steps: Step S1, data acquisition and dual redundancy synchronization: acquiring raw observation data from the primary GNSS receiver and the backup GNSS receiver. The raw observation data includes at least pseudorange, carrier phase, Doppler frequency shift, carrier-to-noise ratio, and automatic gain control (AGC) gain value; achieving synchronization of primary and backup data through time alignment; and employing a joint decision mechanism between the primary and backup receivers, confirming an interference event only when both the primary and backup receivers detect an anomaly simultaneously. Step S2, Multi-feature dynamic threshold fast early warning: Extract multiple core features including AGC gain value, use Welford online incremental algorithm to calculate the mean and standard deviation of each feature in real time, calculate the coefficient of variation based on the mean and standard deviation, dynamically calculate the adaptive threshold based on the coefficient of variation and continuous anomaly count, and combine hard rules and soft rules to make early warning level judgment. Step S3, Fine-grained anti-interference detection of the observation domain: Perform multi-dimensional quality detection on the original observation values, including... Combined cycle slip detection, multi-frequency consistency verification, Doppler-pseudorange rate consistency verification, clock error jump detection, and Doppler residual systematic bias detection are performed to remove observations marked as invalid and output a clean subset of observations. Step S4, Sliding window adaptive RAIM single satellite fault isolation: Calculate pseudorange residuals based on the clean observation subset, maintain the mean and standard deviation of the residuals calculated recursively by the sliding window, and dynamically adjust the RAIM detection threshold according to the carrier cycle skip rate, AGC gain value and anomaly level to identify and remove persistent faulty satellites; Step S5, Extended Epoch Combined with RAIM and Orbital Dynamics Constraints: Cache valid observation data from multiple consecutive epochs, use recursive least squares method to perform receiver state PVT solution, and introduce clock error rate of change constraint, innovation sequence statistical verification and low-Earth orbit dynamics hard constraint to ensure solution integrity; Step S6, System Decision and Closed-Loop Handling: Calculate the comprehensive interference confidence level by integrating multi-source anomaly coefficients, classify the interference level according to the comprehensive interference confidence level and execute the graded handling strategy. When it is determined to be severe interference, the system automatically stops using the raw GNSS observation data and enters the orbit determination extrapolation mode. It autonomously flies over the interference area by relying on the high-precision orbit dynamics model. After the interference disappears, it automatically restores the GNSS positioning mode. Step S7, Global Baseline Adaptation and Telemetry: Monitor the statistical baseline of each feature quantity in real time. When the preset drift condition is met, trigger the baseline adaptation reset process and package the full-link operation status, interference level, confidence level and extrapolation flag into a telemetry frame and transmit it to the ground.

[0006] Furthermore, in step S2, the dynamic adaptive threshold... The expression is as follows: ; Among them, dynamic coefficient The expression is as follows: ; In the formula, Features standard deviation The coefficient of variation is 1. Based on the coefficient, For feature weights, For continuous anomaly counting, and These are the lower and upper limits of the coefficient, respectively. This is the amplitude limiting function.

[0007] Furthermore, the hard rule in step S2 includes any of the following conditions, and anomaly level 2 is directly output when any of the conditions are met: the highest carrier-to-noise ratio of the GPS or BDS system is 0, and its historical baseline mean is ≥42 dB-Hz. The number of tracking satellites in the GPS or BDS system is 0, and its historical baseline average is ≥5. Position accuracy factor (PDOP) > 5.0; AGC gain value < 80.

[0008] Furthermore, in step S3, the Combined cycle slip detection includes: calculation The combined observations are expressed as follows: ; In the formula, , For the frequencies of two points, , These are carrier phase observations at two frequency points. , These are pseudorange observations at two frequency points; if the difference between epochs... If a cycle slip occurs, the satellite is determined to have experienced a cycle slip and all its frequency observations are marked as invalid. The Doppler-pseudorange rate consistency check includes: Calculate the velocity inverted from the Doppler frequency shift velocity calculated by pseudorange difference ,like If so, then mark the observation value at that frequency point as invalid; The clock jump detection includes: If the difference in receiver clock bias estimates between adjacent epochs If so, the marking is suspected of being deceptive or misleading.

[0009] Furthermore, in step S4, the expression for dynamically adjusting the RAIM detection threshold is as follows: ; ; In the formula, As a factor related to cycle slip rate, This is the AGC gain value linkage factor. Anomaly level linkage factor; if the absolute value of the residual mean of a certain satellite is greater than or the standard deviation of the residuals is greater than If the satellite is found to be in a persistent fault, all its frequency observations will be removed.

[0010] Furthermore, in step S5, the hard constraints of low-Earth orbit dynamics include: the orbital height h must satisfy: 300km≤h≤1200km; The velocity amplitude V must satisfy: 7000m / s≤V≤8000m / s; The position difference of the three axes ΔP between the two seconds before and after is ≤15km, and the velocity difference of the three axes ΔV between the two seconds before and after is ≤40m; If either the calculated position or velocity exceeds the above constraints, the calculation result for the current epoch is determined to be invalid.

[0011] Furthermore, in step S6, the overall interference confidence level is... The expression is as follows: ; In the formula, , , , , , , , These are the anomaly coefficients for carrier-to-noise ratio, effective number of satellites, Doppler error rate, residual, AGC gain, cycle slip rate, clock error jump, and Doppler residual, respectively, and each coefficient is normalized to the interval [0, 100].

[0012] Furthermore, in step S6, the system automatically classifies interference levels based on the J value in real time and executes corresponding strategies. The graded handling strategies include: Level 0: If the error rate is less than 10%, the system will output real-time positioning and orbit determination results normally. Level 1: 10% ≤ If the detection threshold is less than 30%, the system will automatically remove the identified faulty satellites and tighten all detection thresholds. Level 2: 30% ≤ If the result is less than 70%, the weight of the orbital dynamics constraint is increased. If the final PVT solution is still invalid, the orbital extrapolation mode is entered. Level 3: ≥70% or both primary and backup machines are in an abnormal state for more than 3 seconds simultaneously or clock difference jump variable >1×10 -4 If the mean value of the s or Doppler residuals is >30 Hz or the orbital constraints are exceeded, the system will be forced to enter the orbital extrapolation mode. The orbit determination extrapolation mode performs continuous extrapolation for 6000 seconds. When 10 consecutive epochs are detected... When the percentage of satellites is less than 30% and the number of valid satellites is no less than 4, the system will automatically revert to GNSS positioning mode.

[0013] Furthermore, in step S7, the conditions for triggering autonomous baseline drift reset are simultaneously met: continuous anomaly count ≥ 1200 seconds; the highest carrier-to-noise ratio of the current GPS or BDS system ≥ 42 dB-Hz; and the coefficient of variation of all features. The relative drift of any feature quantity is >0.3; The reset process includes: determining whether a stable state has been reached based on the fluctuation range of the measured values ​​within the current 30-second window and the overall coefficient of variation of the feature mean; if so, recording the feature mean within the window as a candidate baseline. Then perform a baseline reset operation: the new mean after reset. The new mean after resetting The number of samples in the Welford algorithm is reset to 10, and the second-order central moments are cleared to zero.

[0014] Secondly, based on the same concept, the present invention also provides a low-orbit satellite GNSS anti-interference adaptive RAIM system, which is implemented with a three-layer architecture, including a hardware adaptation layer, a data interaction layer and an anti-interference core algorithm layer. The hardware adaptation layer includes a hardware interface abstraction layer, compilation environment configuration, and computing power scheduling functions, which are used to decouple the core algorithm from different spaceborne processors and GNSS receiver models.

[0015] The data interaction layer includes data synchronization, telemetry management, and parameter management functions, and is responsible for the input and output of internal and external data, as well as the configuration and management of on-orbit parameters.

[0016] The anti-interference core algorithm layer includes the following functional modules: a primary / backup dual receiver redundancy management module, used to perform data synchronization and joint decision-making in step S1; a dynamic threshold interference rapid early warning module, used to perform feature extraction, online statistics, and threshold early warning in step S2; an observation domain fine anti-interference module, used to perform multi-dimensional quality detection and invalid value removal in step S3; a sliding window adaptive RAIM single-satellite fault isolation module, used to perform residual calculation, linkage threshold adjustment, and faulty satellite removal in step S4; an extended epoch joint RAIM and orbit constraint module, used to perform PVT solution and dynamic hard constraint verification in step S5; a system decision and orbit extrapolation module, used to perform confidence calculation, hierarchical handling, and extrapolation mode triggering in step S6; and a global baseline adaptation and telemetry module, used to perform baseline drift monitoring, reset, and full-link status packet downlink in step S7.

[0017] Compared with existing technologies, the low-orbit satellite GNSS anti-interference adaptive RAIM method and system described in this invention have the following advantages: (1) This invention implements anti-interference and RAIM functions entirely based on software architecture, without requiring any physical modification to the RF front-end or baseband hardware of existing GNSS receivers. It can be directly embedded into the existing satellite receiver software framework, or upgraded by software uploading to on-orbit satellites, effectively solving the problems of long modification cycles and high costs of traditional hardware anti-interference solutions, and highly meeting the needs of commercial aerospace for low cost and rapid iteration.

[0018] (2) The traditional fixed threshold detection mechanism is abandoned, and the Welford online incremental statistical algorithm is used in combination with the coefficient of variation to dynamically calculate the adaptive threshold. This mechanism can perceive the time-varying characteristics of the high dynamic flight and electromagnetic environment of low-orbit satellites in real time and automatically adjust the detection sensitivity, fundamentally solving the problem of significantly increased false alarm rate and missed detection rate when orbital maneuvering or geometric configuration changes.

[0019] (3) The AGC gain value is introduced as a fast response feature, and a collaborative decision-making strategy with hard rules as the priority and soft rules as the refinement is constructed. When strong interference occurs suddenly, the system can bypass complex calculations and directly trigger a severe warning. The overall interference identification response delay does not exceed 1 second, effectively ensuring the safety margin of low-orbit satellites under transient strong interference.

[0020] (4) By combining algorithms and software, static memory allocation, circular buffer optimization, and O(1) complexity recursive calculation are adopted to control the static memory usage at runtime to within 6.5KB. No operating system support is required, avoiding complex matrix inversion and iterative operations, and fully adapting to the stringent resource constraints of low computing power and small memory of onboard processors.

[0021] (5) To address the statistical baseline drift caused by the aging of aerospace devices and the gradual changes in the space environment, a global baseline adaptive calibration mechanism was designed. When a continuous drift of the characteristic quantity is detected and the environment is stable, the system can automatically reset the Welford statistical baseline, maintain detection sensitivity without ground intervention, and support stable on-orbit operation for more than five years, thus solving the industry problem of long-term performance degradation of pure software solutions.

[0022] (6) An on-board autonomous closed-loop logic was constructed, which includes interference detection, fault isolation, extrapolation triggering, and state recovery. When severe interference or satellite-to-ground communication interruption is confirmed, the system can automatically cut off abnormal GNSS data, seamlessly switch to orbit determination extrapolation mode, autonomously fly over the interference zone based on a high-precision orbit dynamics model, and smoothly restore positioning after the interference disappears. The entire process is seamless and completely eliminates the excessive dependence on ground telemetry and control.

[0023] (7) Integrating Doppler-pseudorange rate consistency verification, clock error jump detection and Doppler residual systematic deviation detection, the deception features are identified from the physical level of the signal; at the same time, hard constraints of low-orbit orbit dynamics (altitude, velocity amplitude and rate of change limits) are introduced to ensure that the solution results strictly conform to the motion law of low-orbit satellites, effectively eliminating the major risk of incorrect positioning results output by the receiver due to generative / transmitter deception interference.

[0024] (8) A hot standby and joint decision mechanism with dual receivers is adopted, and a single receiver failure will not trigger a false alarm. The master-slave switching delay is ≤1 second. Key on-orbit parameters are read using triple-modular redundancy storage and a two-out-of-three voting method. The command interaction is accompanied by a 32-bit CRC check. It follows the safety design principle of refusing to output invalid solutions rather than outputting erroneous results, and fully meets the high reliability and high security design specifications of aerospace software.

[0025] (9) All threshold parameters and weighting coefficients can be securely injected and modified via ground remote control commands without recompiling and loading the software. At the same time, based on the weighted fusion comprehensive interference confidence model, the system can automatically quantify and distinguish between suppression, deception and hybrid interference, and standardize and package the full-link status parameters for transmission, providing accurate and traceable data support for ground operation and maintenance decisions and on-orbit parameter optimization. Attached Figure Description

[0026] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an undue limitation of the invention. In the drawings: Figure 1 This is a schematic diagram of the overall process of the method described in an embodiment of the present invention; Figure 2 This is a schematic diagram of the functional module architecture of the anti-interference core algorithm layer according to an embodiment of the present invention; Figure 3 This is a schematic diagram of the dynamic threshold rapid early warning process described in an embodiment of the present invention; Figure 4 This is a schematic diagram of the multi-frequency cross-validation process described in an embodiment of the present invention; Figure 5 This is a schematic diagram of the sliding window adaptive RAIM single-star fault isolation process according to an embodiment of the present invention; Figure 6 This is a schematic diagram of the extended epoch joint RAIM and orbit constraint process described in an embodiment of the present invention; Figure 7 This is a schematic diagram of the system-level decision-making and extrapolation triggering process described in an embodiment of the present invention; Figure 8 This is a schematic diagram of the global baseline adaptation and telemetry process described in an embodiment of the present invention; Figure 9 This is a schematic diagram of the overall software architecture described in an embodiment of the present invention; Figure 10 This is a schematic diagram of the core data structure relationship of the software as described in an embodiment of the present invention; Figure 11 This is a schematic diagram illustrating the timing of the detection and autonomous handling of a typical interference event during on-orbit verification as described in an embodiment of the present invention. Figure 12 This is a schematic diagram of the changes in the three-axis position error curves during the orbit extrapolation of a typical disturbance event in the on-orbit verification according to an embodiment of the present invention. Figure 13 This is a schematic diagram of the changes in the three-axis position error curves during the 6000-second orbit determination extrapolation period for on-orbit verification, as described in an embodiment of the present invention. Detailed Implementation

[0027] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other.

[0028] In the description of this invention, it should be understood that the terms "center," "longitudinal," "lateral," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicating orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. Furthermore, the terms "first," "second," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined with "first," "second," etc., may explicitly or implicitly include one or more of that feature. In the description of this invention, unless otherwise stated, "a plurality of" means two or more.

[0029] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art will understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0030] The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0031] like Figures 1 to 13 As shown, in response to the suppressive and deceptive interference faced by low-Earth orbit satellites in complex electromagnetic environments, this invention provides a method and system for adaptive RAIM (Rapid Interference Imaging) for low-Earth orbit satellite GNSS. Its core lies in achieving full-link automation from interference detection, fault isolation, orbit constraint to autonomous closed-loop processing through a multi-layered progressive software processing architecture, which can significantly improve the safety and reliability of satellite navigation systems without modifying hardware.

[0032] like Figure 1 As shown, the low-Earth orbit satellite GNSS anti-jamming adaptive RAIM method includes the following steps: Step S1: Data Acquisition and Dual Redundancy Synchronization At each epoch, raw observation data is collected from both the primary and backup GNSS receivers. This data must include at least pseudorange, carrier phase, Doppler shift, carrier-to-noise ratio, and automatic gain control (AGC) gain values. Synchronization of the observation data between the primary and backup receivers is achieved through time alignment. The primary and backup receivers independently run all subsequent algorithm processes and automatically switch over in case of failure based on mutual inspection results. The real-time joint decision-making rule for the primary and backup receivers is as follows: when only one receiver detects a warning-level anomaly while the other is functioning normally, the system does not trigger an anomaly alarm but only reports the suspicious event via telemetry; when both receivers detect anomalies simultaneously, the system automatically identifies it as a human interference event, thereby effectively improving the accuracy of interference identification and reducing the false alarm rate.

[0033] Step S2: As Figure 3 As shown, rapid early warning using multi-feature dynamic thresholds. This step employs a collaborative decision-making strategy that prioritizes hard rules and refines soft rules: Hard rules for rapid response: If any of the following conditions are met, output an anomaly level of 2, a severe warning: The highest carrier-to-noise ratio of the GPS or BDS system is 0, and its historical baseline mean is ≥42 dB-Hz; The number of tracking satellites in the GPS or BDS system is 0, and its historical baseline average is ≥5. PDOP > 5.0; AGC gain value < 80.

[0034] Soft rules for refined decision-making: Establish a feature set with six core features, including the AGC gain value, specifically: the highest carrier-to-noise ratio of the GPS system, the highest carrier-to-noise ratio of the BDS system, the number of satellites tracked by the GPS system, the number of satellites tracked by the BDS system, the position accuracy factor (PDOP), and the AGC gain value.

[0035] The Welford online incremental algorithm is used to calculate the number of real-time calculations. mean of characteristic and standard deviation Based on the coefficient of variation and continuous anomaly count Dynamically calculate the adaptive threshold for each feature. The dynamic coefficient The expression is as follows: ; In the formula, Features standard deviation The coefficient of variation is 1. Based on the coefficient, For feature weights, For continuous anomaly counting, and These are the lower and upper limits of the coefficient, respectively. This is the amplitude limiting function.

[0036] Calculate the dynamic real-time values ​​of the six core features mentioned above, and score each feature: If the measured value exceeds the dynamic adaptive threshold... An abnormal score is awarded for every 1.5 times or more of the abnormal score, and a warning score is awarded for every 1.5 times or more of the abnormal score. The total abnormal score is calculated by summing these scores. and total early warning points The judgment is as follows: Total abnormal score As the highest priority, the total warning score is... It is the second priority, if This is level 2, and in this case, there is no need to determine the total warning score. ,if only Directly classify it as level 2; if or and The one that is level 1 is considered level 1, and the others are level 0.

[0037] Step S3: As Figure 4 As shown, fine-grained anti-interference detection in the observation domain Multi-dimensional quality checks are performed on the raw observations of each satellite at each frequency point, specifically including: MW combination cycle slip detection: Calculate MW combination observations In the formula, , For the frequencies of two points, , These are carrier phase observations at two frequency points. , Given the pseudorange observations at two frequency points, if the difference between epochs... If a cycle slip occurs, the satellite is determined to have experienced a cycle slip, and all observations at its frequencies are marked as invalid. The cycle slip rate is also calculated. .in, This represents the total number of cycle slips. The total number of satellites included in the statistics. The time interval for the statistics.

[0038] Multi-frequency consistency verification: For the same satellite, if the difference between pseudorange observations at different frequency points... If so, it is determined to be an observational anomaly.

[0039] Doppler-pseudorange rate consistency check: Calculate the velocity inverted from the Doppler frequency shift And the velocity calculated from pseudorange difference In the formula, For wavelength, For Doppler frequency shift, This is the pseudo-distance of the current epoch. If If so, the observation value at that frequency point is marked as invalid.

[0040] Clock bias jump detection: If the difference in receiver clock bias estimates between adjacent epochs... If the marker is suspected of being deceptive or interfering, then the marker is invalid.

[0041] Doppler residual systematic bias detection: Calculate the mean of the residuals between the theoretical Doppler value and the actual measured Doppler value. .like If a systematic deception bias is found, the satellite is marked as invalid. After the above checks, all invalid observations are removed, and a clean subset of observations and corresponding interference characteristic statistics are output.

[0042] Step S4: As Figure 5 As shown, sliding window adaptive RAIM single-star fault isolation Based on the clean observation subset output from step S3, calculate the pseudorange residual of the original observations for each satellite at each frequency. In the formula, For pseudo-range, Geometric distance This represents the receiver clock bias. Simultaneously, a sliding window of length 5 epochs is maintained to recursively calculate the mean of the residuals. and standard deviation Based on the current cycle jump rate The AGC gain value and anomaly level are used to dynamically adjust the RAIM detection threshold. The threshold parameters mainly include... and The two core parameters are the mean threshold and the standard deviation threshold. The formulas for calculating these two core parameters are as follows: , The rules for determining the values ​​of each linkage factor are as follows: Week slip rate factor :when Use 0.8 if the time is times / second / star, otherwise use 1.0; AGC factor : When AGC < 80, take 1.1; otherwise, take 1.0. Abnormality level factor : 0.8 is used when the system is in an abnormal state, otherwise 1.0 is used. If a certain satellite's... or If the satellite is found to be in a persistent fault, all observations at its frequency points will be discarded.

[0043] Step S5: As Figure 6 As shown, the extended epoch joint RAIM and orbital dynamics constraints The system continuously buffers valid observation data for 8 epochs and uses the least squares (RLS) method to perform PVT calculations of the receiver state (position, velocity, clock error, etc.). To ensure the accuracy of the PVT calculation results, the following hard constraints are introduced: Clock bias rate of change constraint: the rate of change of receiver clock bias If the value exceeds the limit, the solution is considered invalid.

[0044] Statistical validation of the innovation sequence: If the standard deviation of the innovation sequence... Greater than its nominal value 3 times, that is If the error occurs, it is considered a system-level anomaly, and the solution is invalid.

[0045] Orbital dynamic constraints: The satellite's orbital data is constrained in the WGS84 coordinate system, and the satellite's position coordinates in this coordinate system are... The velocity coordinates are The coordinates of the Earth's center of mass in this coordinate system are: Earth's radius is .

[0046] orbital height Must meet: ,in The geometric distance from the satellite's position coordinates to the Earth's center of mass coordinates, minus the Earth's radius, is calculated using the following formula: .

[0047] velocity amplitude Must meet: ,in The resultant velocity of the satellite along its three axes in the WGS84 system is calculated using the following formula: If either the calculated position or velocity exceeds the above constraints, the solution result for the current epoch is deemed invalid.

[0048] The difference in position between the three axes two seconds before and after the change, ΔP = P1 - P2, must satisfy: ΔP 15km.

[0049] The three-axis velocity difference ΔV = V1 - V2 within two seconds before and after must satisfy: ΔV 40m. If either the calculated position or velocity exceeds the above constraint range, the calculation result for the current epoch is directly determined to be invalid.

[0050] Step S6: As Figure 7 As shown, system decision-making and closed-loop processing The end-to-end statistics output from the above steps are merged, and eight specific anomaly coefficients are calculated. Each coefficient is normalized to... Interval: ; ; ; ; ; ; ; ; In the formula, , , , , , , , These are the carrier-to-noise ratio, effective number of satellites, Doppler error rate, residual, AGC gain, cycle slip rate, clock error jump, and anomaly coefficient of Doppler residual, respectively. The current observed average carrier-to-noise ratio, The number of satellites currently being tracked. The number of frequency points or satellites where Doppler-pseudorange rate consistency verification failed. is the root mean square of the pseudorange residual, and AGC is the receiver's automatic gain control value. For cycle slip rate, This represents the absolute value of the difference between receiver clock bias estimates between adjacent epochs. It represents the absolute value of the mean of the residuals between the theoretical Doppler value and the actual measured Doppler value.

[0051] Calculate the overall interference confidence level Its weighting coefficients can be configured in orbit: .

[0052] The system automatically in real time according to The system classifies interference levels and executes corresponding strategies, with its core being the autonomous handling logic after interference is confirmed, as follows: Level 0 (i.e., in an interference-free state): The system outputs real-time positioning and orbit determination results normally, and allows automatic updates to the statistical baseline of the feature vectors.

[0053] Level 1 (i.e., under weak interference): The system automatically removes identified faulty satellites and tightens all detection thresholds by 10%.

[0054] It should be noted that if a single satellite malfunctions, no further judgment is needed. The value is directly determined to be at level 1 interference level.

[0055] Level 2 (i.e., under strong interference): The system increases the weight of orbital dynamic constraints. If the final PVT solution is still invalid, the system automatically enters the orbit determination extrapolation mode.

[0056] It should be noted that if the number of valid solution satellites is less than 4, no further judgment is needed. The value is directly determined to be at level 2 interference level.

[0057] Level 3 (i.e., under severe interference): Or, both primary and backup machines simultaneously enter an abnormal state for more than 3 seconds or the clock difference jump variable > If the mean Doppler residual exceeds 30 Hz or the orbital constraints are exceeded, the system automatically stops using raw GNSS observation data and automatically enters orbit extrapolation mode. Orbit extrapolation mode can perform continuous extrapolation for 6000 seconds, ensuring that the satellite can autonomously and safely fly over the entire interference area without relying on GNSS signals, using a high-precision orbital dynamics model, and continue to perform its intended mission. After flying over the interference area, detection is performed for 10 consecutive epochs. When the number of valid satellites is no less than 4, the system will automatically restore GNSS positioning mode.

[0058] At the same time, the system identifies the type of interference, as follows: Suppressing interference: , , , The abnormality values ​​of all four indicators are ≥70% (this threshold supports on-orbit remote injection configuration).

[0059] Deception and interference: , Both indicators have anomaly values ​​of ≥70%, and both trigger orbital constraint out-of-bounds errors.

[0060] Hybrid interference: simultaneously satisfying the characteristics of both of the above and .

[0061] Step S7: As Figure 8 As shown, global baseline adaptation and telemetry The system monitors the statistics of the six core features mentioned in step S2 in real time (GPS system maximum carrier-to-noise ratio, BDS system maximum carrier-to-noise ratio, number of GPS system tracked satellites, number of BDS system tracked satellites, position accuracy factor PDOP, and AGC gain value). The baseline drift autonomous reset process is triggered when all of the following conditions are met simultaneously: Continuous anomaly count ≥ 1200 seconds; The highest carrier-to-noise ratio of current GPS or BDS systems is ≥42 dB-Hz; Coefficient of variation of all features ; The relative drift of any characteristic quantity ,in, For the first The current measured values ​​of each feature quantity The mean of the i-th feature is used to determine the i-th feature. Does the current value of a feature deviate more than 30% from its mean?

[0062] Once all the aforementioned baseline drift triggering conditions are met, the system enters the reset process, which is as follows: First, the six core features mentioned in step S2 are monitored respectively, requiring that the fluctuation range of the measured value of each feature within the current 30-second window does not exceed its historical standard deviation. ±0.5 times, and the overall coefficient of variation of the mean of each feature within the window. This confirms that the system has entered a stable state, and the average value of each feature within the 30-second window is recorded as the candidate baseline. Then, perform a baseline reset operation: the new mean after the reset. The new standard deviation after reset and the number of samples in the Welford algorithm Reset to 10, second-order central moment Reset. After resetting, the continuous anomaly count is reset to zero, and the baseline reset event is reported via telemetry. Finally, key data such as the end-link operating status, interference level, confidence level, extrapolation flag, and interference type are packaged into telemetry frames and transmitted to the ground, allowing the ground station to directly and clearly interpret the satellite anti-interference system status and handling process.

[0063] like Figure 9 As shown, based on the aforementioned low-Earth orbit (LEO) satellite GNSS anti-jamming adaptive RAIM method, this invention also provides a LEO satellite GNSS anti-jamming adaptive RAIM system, which adopts a three-layer modular architecture: (1) Hardware adaptation layer: including hardware interface abstraction layer, compilation environment configuration and computing power scheduling, etc., to decouple the core algorithm from different spaceborne processors and GNSS receiver models.

[0064] (2) Data interaction layer: including functions such as data synchronization, telemetry management and parameter management, responsible for the input and output of internal and external data as well as the configuration and management of on-orbit parameters.

[0065] (3) Anti-interference core algorithm layer: including dynamic threshold warning, multi-frequency verification, adaptive RAIM, baseline adaptation and system decision-making functions, implemented based on standard C99 language, without relying on any operating system, its static memory usage is about 6.5 KB, and the single epoch processing time is about 750 μs, which is fully adapted to low computing power processors in spaceborne scenarios. Figure 2 As shown, the core anti-interference algorithm layer includes the following functional modules: The primary and backup dual receiver redundancy management module is used to perform data synchronization and joint decision-making in step S1. Specifically, it is responsible for data synchronization of the primary and backup receivers, mutual inspection of operating status, and automatic switching in case of failure.

[0066] Dynamic threshold interference rapid early warning module: used to perform feature extraction, online statistics and threshold early warning in step S2. Specifically, based on the Welford incremental statistical algorithm and AGC features, it realizes rapid interference early warning at the second level.

[0067] The observation domain fine anti-interference module is used to perform multi-dimensional quality detection and invalid value removal in step S3. Specifically, it performs tasks such as MW combined cycle slip detection, multi-frequency consistency verification, Doppler-pseudorange rate consistency verification, clock error jump detection, and Doppler residual mean detection.

[0068] Sliding window adaptive RAIM single-satellite fault isolation module: used to perform residual calculation, linkage threshold adjustment and faulty satellite removal in step S4. Specifically, based on the RAIM threshold adjusted in linkage with cycle slip rate, AGC value and anomaly level, it realizes the identification and removal of persistent faulty satellites.

[0069] Extended epoch joint RAIM and orbit constraint module: used to perform PVT calculation and dynamic hard constraint verification in step S5. Specifically, through least squares calculation and dynamic orbit hard constraints, it provides system integrity monitoring and assurance.

[0070] The system autonomous decision-making and orbit extrapolation module is used to perform the confidence calculation, graded handling and extrapolation mode triggering in step S6. Specifically, the system autonomously calculates the comprehensive interference confidence in real time, executes the graded handling strategy in sequence, and autonomously enters the orbit extrapolation mode after detecting severe interference. The satellite continues to perform its mission by relying on the PVT output by the orbit extrapolation and flies over the interference zone without being noticed, ensuring mission continuity.

[0071] Global baseline adaptation and telemetry module: used to perform baseline drift monitoring, reset and full-link status packaging and downlink in step S7. Specifically, it realizes the automatic calibration function of on-orbit characteristic baseline and packages and telemetry the full-link status parameters.

[0072] It should be noted that all threshold parameters and weight coefficients related to anti-interference in the core algorithm layer have reserved on-orbit remote control injection interfaces, which support the ground telemetry and control system to dynamically modify the parameters through uplink commands without recompiling and loading the software.

[0073] like Figure 10 As shown, to support the core anti-interference algorithm layer, the system defines a clear data structure relationship, as follows: (1) Core state structure SystemState: Records the current operating status, interference level, and confidence level of the system.

[0074] WelfordSat: Stores the mean, variance, M2 moments, and number of samples required for the Welford algorithm.

[0075] SlidingWindow: Manages the residual window and pointer of the RAIM sliding window.

[0076] (2) Observation data structure ObsData: Raw observations (pseudorange, carrier, Doppler).

[0077] SatInfo: Satellite ID, frequency, and valid flag.

[0078] QualityMetrics: Quality indicators (cycle slip rate, consistency indicators).

[0079] (3) Decision output structure DecisionResult: Disposal level, extrapolation flag.

[0080] Recovery: Status recovery information.

[0081] TelemetryPkt: A packaged telemetry data packet used for downloading.

[0082] Example 1: I. Hardware Platform Configuration The hardware platform in this embodiment is designed based on a typical spaceborne embedded system, with a low-Earth orbit satellite at an altitude of approximately 520 km as the application scenario. The specific configuration is as follows: Processor: A 300MHz spaceborne embedded processor, model ARM Cortex-R5F, is selected. This processor supports double-precision floating-point operations and features low power consumption and high reliability.

[0083] Storage resources: On-chip RAM capacity is no less than 128KB, used for data storage during algorithm execution; on-chip Flash capacity is no less than 1MB, used for storing algorithm code and default configuration parameters.

[0084] GNSS Receiver: A satellite-borne receiver supporting GPS (L1C / L2C) and BDS (B1I / B1C / B2A / B3I) frequencies is selected. Its output data includes pseudorange, carrier phase, Doppler shift, carrier-to-noise ratio, PDOP value, etc., with an output frequency of 1 Hz. The system employs a dual-machine redundancy configuration with primary and backup receivers.

[0085] Hardware interfaces include an SPI interface (rate ≥ 1 Mbps, used to receive observation data), an I2C interface (used to read AGC values ​​and configure the receiver), a timer interrupt (1 Hz, used to trigger the main algorithm loop), and a UART interface (rate 460800bps, used for telemetry and remote control).

[0086] Onboard time system: It uses a high-stability temperature-controlled crystal oscillator for timekeeping, and the time synchronization accuracy between the main and backup receivers is better than 100 nanoseconds.

[0087] II. Software Deployment and Debugging Process Software compilation and deployment: Compilation environment: GCC cross compiler (Code Composer Studio 12.6.0) was used, and the target architecture was ARM-Cortex-R5F.

[0088] Link configuration: Specify the code storage address (Flash starting address is 0x00004000) and data execution address (RAM starting address is 0x08000000) through a custom linker script.

[0089] Download and Deployment: Download the compiled binary file to the processor's Flash memory using a JTAG emulator (model: Segger J-Link).

[0090] Power-on initialization and debugging: Power-on self-test: After the system is powered on, it first performs memory integrity verification, communication interface initialization verification, and communication handshake verification with the receiver.

[0091] Algorithm initialization: The `leo_anti_jam_init` interface function is called to load the default threshold parameters and feature baseline, and to clear various buffers and state counters. The system state machine enters the `STATE_INIT` initialization state by default.

[0092] Pre-collection of statistical baseline: Collect at least 10 epochs of interference-free valid data (with at least 10 valid statistical samples) to establish the initial statistical baseline. After the baseline is established, the state machine switches to the STATE_NORMAL normal state and starts the main loop.

[0093] Debugging methods: Supports hardware breakpoint debugging, software log output debugging, and real-time status monitoring via ground telemetry.

[0094] III. Main Software Flow After the system powers on and initializes, the main processing function is executed, triggered by a 1 Hz timer interrupt. The main processing flow for each epoch is as follows: (1) Data synchronization: Read the observation data from the primary and backup receivers and verify whether the timestamp synchronization error is within the allowable range. If synchronization fails, the primary receiver data will be enabled by default.

[0095] (2) Dynamic threshold warning: Extract 6 core feature values, update Welford statistics, execute hard / soft rule judgment process, and output the current abnormality level.

[0096] (3) Fine anti-interference in the observation domain: Perform a multi-dimensional quality inspection process, including MW cycle slip detection, multi-frequency consistency, Doppler-pseudorange rate consistency, clock error jump detection, Doppler residual deviation detection, etc., to eliminate invalid observations.

[0097] (4) Sliding window adaptive RAIM: Calculate pseudo-range residuals and dynamically adjust the threshold according to cycle slip rate, AGC value and anomaly level.

[0098] (5) Extended epoch joint RAIM: The least squares method is used to solve the PVT and the clock error rate, innovation sequence, and orbital dynamics hard constraints are applied.

[0099] (6) System decision-making: Calculate the overall confidence level ,according to The system implements a tiered handling strategy. When a Level 3 severe interference is detected, it autonomously enters the orbit determination extrapolation mode.

[0100] (7) Baseline Adaptation: Monitors the drift of feature statistics. If the triggering conditions are met, the baseline reset process is automatically executed.

[0101] (8) Telemetry update: Package the full-link status data (system status, interference level, confidence level, extrapolation flag, interference type, etc.) and transmit it to the ground through the interface.

[0102] IV. Typical Scenario Operation Process Scenario 1: Normal Operation Without Interference: Input normal observation data, all detection steps pass, and the overall confidence level is [not specified]. The system determined the interference level to be 0 and output the positioning and orbit determination results normally.

[0103] Scenario 2: Suppression Interference: After injecting suppression interference at 1575.42 MHz with a power of -40 dBm, the AGC value drops sharply, the carrier-to-noise ratio decreases, and the number of tracked satellites decreases. A hard rule decision indicates an anomaly level of 2, and the system responds quickly. Subsequent refined detection and decision-making lead to a comprehensive confidence level... Soaring to 88%, greater than The system automatically classifies the interference level as Level 3; based on the interference type identification rules in step S6, specifically, under -40 dBm interference, the average carrier-to-noise ratio drops to approximately 27.3 dB-Hz, corresponding to... Number of satellites currently being tracked Reduced to about 3, corresponding Week jump rate Increased to approximately 6.0 times / second / star, corresponding to The AGC gain value drops to approximately 53, corresponding to... ; , , , Substitute into the formula to calculate: The overall interference confidence level J = 88% is obtained; at this time , , , All exceeded the 70% threshold for detection, therefore it was determined to be interference suppression. The system autonomously entered orbit extrapolation mode, and the satellite continued to fly and complete its mission using the PVT data provided by the orbit extrapolation. After the satellite flew out of the interference zone, it automatically resumed positioning and orbit determination functions. The total delay for interference identification and response was in the second range.

[0104] Scenario 3: Deception Interference: Forgery causes the difference in receiver clock bias estimates between adjacent epochs to be... Furthermore, the mean absolute value of the residuals between the theoretical and actual measured Doppler values ​​is 30Hz. Clock bias jumps and excessive Doppler residuals were detected in the observation domain, and the observations were marked as invalid. During extended epoch calculations, the calculations were deemed invalid due to falsified orbital altitudes exceeding the limits. System overall confidence level. The interference rate is 82%, and the system automatically classifies it as Level 3 interference. Based on the interference type identification rules in step S6, specifically, the difference in receiver clock bias estimates between adjacent epochs... According to the formula Calculated to 100%; the mean absolute value of the residuals between the theoretical Doppler value and the actual measured Doppler value. ,according to The calculation shows that the spoofing signal causes the carrier-to-noise ratio to drop to 30 dB-Hz. The number of effective satellites has dropped to 5. The AGC gain value dropped to 75. The weekly jump rate rose to 3.0 times / second / star. , , Substitute into the formula to calculate: The overall interference confidence level is J = 82%; orbit constraint violation specifically refers to the calculated satellite orbit altitude exceeding the specified range of 300km-1200km. In this example, the calculated altitude is 1500km, directly triggering an invalid solution. , All exceeded the 70% judgment threshold and the orbit constraint exceeded the limit, so it was judged as deception interference, and the system automatically entered the orbit determination extrapolation mode.

[0105] Scenario 4: Hybrid Interference: After simultaneously injecting suppression interference with a power of -60dBm and clock slip deception interference, the system's dynamic threshold early warning module triggers a hard rule and immediately outputs anomaly level 2. Further fine-tuning of the observation domain reveals that satellite observations simultaneously exhibit cycle slip anomalies and clock slip anomalies. The system's overall interference confidence level J rapidly rises to 92%, and the system automatically classifies its interference level as level 3. Based on the interference type identification rules in step S6, specifically, the various anomaly indicators at this point are as follows: Suppression of interference characteristics: The average carrier-to-noise ratio drops to approximately 30 dB-Hz, corresponding to Number of satellites currently being tracked Reduced to about 4, corresponding Week jump rate Increased to approximately 5.0 times / second / star, corresponding to The AGC gain value drops to approximately 65, corresponding to... All four of the above items are significantly higher than the normal baseline, which meets the criteria for suppressing interference.

[0106] Deceptive interference characteristics: clock bias jump ,correspond The absolute value of the mean of the residuals between the theoretical Doppler value and the actual measured Doppler value. ,correspond The number of frequency points or satellites where Doppler-pseudorange rate consistency verification failed. There are 4. Root mean square of pseudo-distance residuals ,correspond Meanwhile, the calculated orbital altitude exceeds the range of 300km-1200km. In this example, the calculated altitude is 250km, which is below the lower limit of 300km. This meets the criteria for deception interference.

[0107] Because both types of features appear simultaneously and the combined interference confidence level is calculated. According to the interference type identification rules in step S6, the system automatically determines that it is mixed interference and immediately enters the orbit determination extrapolation mode. The satellite safely flies over the interference zone by relying on the PVT data output by the orbit determination extrapolation. Therefore, the judgment conditions of both suppression interference and deception interference are met at the same time, so it is determined to be mixed interference, and the system automatically enters the orbit determination extrapolation mode.

[0108] Scenario 5: Partial Frequency Failure: When all observations of the BeiDou B3I frequency point are marked as invalid, while the data of the remaining five frequency points are normal, the dynamic threshold early warning module extracts features based on the valid frequency points and outputs an anomaly level of 0. The observation domain detection automatically skips the faulty B3I frequency point and only performs consistency checks on the remaining valid frequency point pairs, finding no anomalies. System overall confidence level With only 4% of the data collected, this is classified as Level 0, an interference-free state, and the positioning and orbit determination results are output normally. This scenario demonstrates that the software has the capability to automatically degrade to a lower level.

[0109] Scenario 6: Baseline Drift After Long-Term Operation: After several months of continuous operation, due to factors such as hardware aging, the AGC and carrier-to-noise ratio baselines drift, leading to frequent warnings even in the absence of interference. The system detects that all baseline reset trigger conditions are met. After 30 seconds of candidate baseline confirmation, the Welford statistic is automatically updated and the baseline is reset. After the reset, the false alarm rate returns to normal levels, and the system resumes stable operation.

[0110] V. Test Verification Results A test environment consisting of a spaceborne processor simulation platform (ARM Cortex-R5F), a GNSS signal simulator (Sprent 9000), and an interference injection device (Keysight N5182B) was established to simulate the low-Earth orbit satellite operating environment at an altitude of 500km and a speed of 7600m / s. Comprehensive functional and performance tests were conducted on the proposed solution. The main test results are shown in Table 1: Table 1 VI. On-orbit verification and performance The method and system described in this invention have been verified in orbit on the XX45 and XX46 low-Earth orbit satellites. The satellite platform is equipped with an onboard GNSS receiver supporting both GPS and BDS modes and possesses dual-frequency, multi-mode real-time orbit determination capabilities. The software of this invention is integrated into the central processor of the GNSS receiver. During the months-long in-orbit testing, the satellites repeatedly flew through complex electromagnetic airspace, and the system successfully triggered automatic handling procedures for multiple effective interference events. The specific performance is as follows (the autonomous handling process of a typical event is as follows...). Figure 11 and Figure 12 (as shown) (1) Accuracy and timeliness of interference identification: Post-analysis and comparison of the integrated telemetry data of the entire satellite transmitted from the ground control station revealed that the system could accurately identify on-orbit artificial suppression interference events without any missed reports; the detection and identification of deceptive interference features met the expected objectives. The average on-board autonomous response time from the appearance of interference features to the system's decision met expectations, ensuring that when the system is severely interfered with, the system immediately and autonomously switches the orbit determination mode to the orbit determination extrapolation mode, and the satellite continues to perform its mission by relying on the PVT output by the orbit determination extrapolation and flies over the interference zone without being noticed.

[0111] (2) Effectiveness of orbit extrapolation closed-loop handling: In confirmed severe interference events, the system autonomously enters orbit extrapolation mode. During the GNSS signal failure period, the satellite relies on the high-precision orbital dynamics model of the GNSS receiver for navigation and position holding. The three-axis position error change curves during orbit extrapolation are shown in the figure below. Figure 12 As shown, the mission requirements were met. During the orbit extrapolation, the satellite's attitude remained stable, all subsystems of the platform functioned normally, and the payload mission was not interrupted. After the satellite flew out of the interference zone, the system automatically detected the recovery of the GNSS signal and smoothly switched back to normal positioning and orbit determination mode. The entire process was conducted without ground intervention, achieving a complete onboard autonomous closed loop of identification, handling, and recovery.

[0112] (3) Long-term operational stability: During the on-orbit verification period, the system operated continuously without failure for 295 days. Its built-in global baseline adaptive calibration module autonomously operated once in orbit, effectively overcoming the problem of slow drift of characteristic baselines caused by space environment and device aging, ensuring long-term stability of detection performance, and verifying the long-life adaptability of the present invention. The on-orbit test results fully demonstrate the advanced nature, reliability and engineering practicality of the present invention, and can meet the requirements of low-orbit satellites for highly autonomous, highly reliable and long-life operation in complex electromagnetic environments.

[0113] VII. On-orbit Operation and Parameter Adjustment The software of this invention supports flexible on-orbit parameter adjustment and fault handling. The ground control center can inject new threshold parameters via parameter uploading based on satellite on-orbit telemetry data. After verifying the 32-bit CRC checksum carried in the verification command is correct, the software will safely update the target parameters (using triple-modulus redundancy storage) during epoch intervals. If an anomaly occurs after on-orbit parameter adjustment, the ground can quickly inject default parameters or send a rollback command to restore the system to the previous stable version of parameters. This mechanism ensures the maintainability and adaptability of the system throughout its entire lifecycle.

[0114] The beneficial effects of this invention are: (1) This invention implements anti-interference and RAIM functions entirely based on software architecture, without requiring any physical modification to the RF front-end or baseband hardware of existing GNSS receivers. It can be directly embedded into the existing satellite receiver software framework, or upgraded by software uploading to on-orbit satellites, effectively solving the problems of long modification cycles and high costs of traditional hardware anti-interference solutions, and highly meeting the needs of commercial aerospace for low cost and rapid iteration.

[0115] (2) The traditional fixed threshold detection mechanism is abandoned, and the Welford online incremental statistical algorithm is used in combination with the coefficient of variation to dynamically calculate the adaptive threshold. This mechanism can perceive the time-varying characteristics of the high dynamic flight and electromagnetic environment of low-orbit satellites in real time and automatically adjust the detection sensitivity, fundamentally solving the problem of significantly increased false alarm rate and missed detection rate when orbital maneuvering or geometric configuration changes.

[0116] (3) The AGC gain value is introduced as a fast response feature, and a collaborative decision-making strategy with hard rules as the priority and soft rules as the refinement is constructed. When strong interference occurs suddenly, the system can bypass complex calculations and directly trigger a severe warning. The overall interference identification response delay does not exceed 1 second, effectively ensuring the safety margin of low-orbit satellites under transient strong interference.

[0117] (4) By combining algorithms and software, using static memory allocation, circular buffer optimization and O(1) complexity recursive calculation, the static memory usage during runtime is controlled within 6.5KB, and the number of floating-point operations per epoch is ≤2100. No operating system support is required, avoiding complex matrix inversion and iterative calculations, and fully adapting to the strict resource constraints of low computing power and small memory of onboard processors.

[0118] (5) To address the statistical baseline drift caused by the aging of aerospace devices and the gradual changes in the space environment, a global baseline adaptive calibration mechanism was designed. When a continuous drift of the characteristic quantity is detected and the environment is stable, the system can automatically reset the Welford statistical baseline, maintain detection sensitivity without ground intervention, and support stable on-orbit operation for more than five years, thus solving the industry problem of long-term performance degradation of the solution.

[0119] (6) An on-board autonomous closed-loop logic was constructed, which includes interference detection, fault isolation, extrapolation triggering, and state recovery. When severe interference or satellite-to-ground communication interruption is confirmed, the system can automatically cut off abnormal GNSS data, seamlessly switch to orbit determination extrapolation mode, autonomously fly over the interference zone based on a high-precision orbit dynamics model, and smoothly restore positioning after the interference disappears. The entire process is seamless and completely eliminates the excessive dependence on ground telemetry and control.

[0120] (7) Integrating Doppler-pseudorange rate consistency verification, clock error jump detection and Doppler residual systematic deviation detection, the deception features are identified from the physical level of the signal; at the same time, hard constraints of low-orbit orbit dynamics (altitude, velocity amplitude and rate of change limits) are introduced to ensure that the solution results strictly conform to the motion law of low-orbit satellites, effectively eliminating the major risk of incorrect positioning results output by the receiver due to generative / transmitter deception interference.

[0121] (8) A hot standby and joint decision mechanism with dual receivers is adopted, and a single receiver failure will not trigger a false alarm. The master-slave switching delay is ≤1 second. Key on-orbit parameters are read using triple-modular redundancy storage and a two-out-of-three voting method. The command interaction is accompanied by a 32-bit CRC check. It follows the safety design principle of refusing to output invalid solutions rather than outputting erroneous results, and fully meets the high reliability and high security design specifications of aerospace software.

[0122] (9) All threshold parameters and weighting coefficients can be securely injected and modified via ground remote control commands without recompiling and loading the software. At the same time, based on the weighted fusion comprehensive interference confidence model, the system can automatically quantify and distinguish between suppression, deception and hybrid interference, and standardize and package the full-link status parameters for transmission, providing accurate and traceable data support for ground operation and maintenance decisions and on-orbit parameter optimization.

[0123] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A low-Earth orbit satellite GNSS anti-interference adaptive RAIM method, characterized by: Includes the following steps: Step S1, Data Acquisition and Dual Redundancy Synchronization: Collect raw observation data from the primary GNSS receiver and the backup GNSS receiver. The raw observation data includes at least pseudorange, carrier phase, Doppler frequency shift, carrier-to-noise ratio, and automatic gain control (AGC) gain value. Achieve synchronization of primary and backup data through time alignment. Employ a joint decision mechanism between the primary and backup receivers, confirming an interference event only when both receivers detect an anomaly simultaneously. Step S2, Multi-feature dynamic threshold fast early warning: Extract multiple core features including AGC gain value, use Welford online incremental algorithm to calculate the mean and standard deviation of each feature in real time, calculate the coefficient of variation based on the mean and standard deviation, dynamically calculate the adaptive threshold based on the coefficient of variation and continuous anomaly count, and combine hard rules and soft rules to make early warning level judgment. Step S3, Fine-grained anti-interference detection of the observation domain: Perform multi-dimensional quality detection on the original observation values, including... Combined cycle slip detection, multi-frequency consistency verification, Doppler-pseudorange rate consistency verification, clock error jump detection, and Doppler residual systematic bias detection are performed to remove observations marked as invalid and output a clean subset of observations. Step S4, Sliding window adaptive RAIM single satellite fault isolation: Calculate pseudorange residuals based on the clean observation subset, maintain the mean and standard deviation of the residuals calculated recursively by the sliding window, and dynamically adjust the RAIM detection threshold according to the carrier cycle skip rate, AGC gain value and anomaly level to identify and remove persistent faulty satellites; Step S5, Extended Epoch Combined with RAIM and Orbital Dynamics Constraints: Cache valid observation data from multiple consecutive epochs, use recursive least squares method to perform receiver state PVT solution, and introduce clock error rate of change constraint, innovation sequence statistical verification and low-Earth orbit dynamics hard constraint to ensure solution integrity; Step S6, System Decision and Closed-Loop Handling: Calculate the comprehensive interference confidence level by integrating multi-source anomaly coefficients, classify the interference level according to the comprehensive interference confidence level and execute the graded handling strategy. When it is determined to be severe interference, the system automatically stops using the raw GNSS observation data and enters the orbit determination extrapolation mode. It autonomously flies over the interference area by relying on the high-precision orbit dynamics model. After the interference disappears, it automatically restores the GNSS positioning mode. Step S7, Global Baseline Adaptation and Telemetry: Monitor the statistical baseline of each feature quantity in real time. When the preset drift condition is met, trigger the baseline adaptation reset process and package the full-link operation status, interference level, confidence level and extrapolation flag into a telemetry frame and transmit it to the ground.

2. The low-orbit satellite GNSS anti-interference adaptive RAIM method according to claim 1, characterized in that: In step S2, the dynamic adaptive threshold is... The expression is as follows: ； Among them, dynamic coefficient The expression is as follows: ； In the formula, Features standard deviation The coefficient of variation is 1. Based on the coefficient, For feature weights, For continuous anomaly counting, and These are the lower and upper limits of the coefficient, respectively. This is the amplitude limiting function.

3. The low-orbit satellite GNSS anti-interference adaptive RAIM method according to claim 1, characterized in that: The hard rule in step S2 includes any of the following conditions. When any of the conditions is met, the anomaly level 2 is directly output: the highest carrier-to-noise ratio of the GPS or BDS system is 0, and its historical baseline mean is ≥42 dB-Hz. The number of tracking satellites in the GPS or BDS system is 0, and its historical baseline average is ≥5. Position accuracy factor (PDOP) > 5.0; AGC gain value < 80.

4. The low-orbit satellite GNSS anti-interference adaptive RAIM method according to claim 1, characterized in that: In step S3, the Combined cycle slip detection includes: calculation The combined observations are expressed as follows: ； In the formula, , For the frequencies of two points, , These are carrier phase observations at two frequency points. , These are pseudorange observations at two frequency points; if the difference between epochs... If a cycle slip occurs, the satellite is determined to have experienced a cycle slip and all its frequency observations are marked as invalid. The Doppler-pseudorange rate consistency check includes: Calculate the velocity inverted from the Doppler frequency shift The velocity calculated by pseudorange difference ,like If so, then mark the observation value at that frequency point as invalid; The clock jump detection includes: If the difference in receiver clock bias estimates between adjacent epochs If so, the marking is suspected of being deceptive or misleading.

5. The low-orbit satellite GNSS anti-interference adaptive RAIM method according to claim 1, characterized in that: In step S4, the expression for dynamically adjusting the RAIM detection threshold is as follows: ； ； In the formula, As a factor linked to cycle slip rate, This is the AGC gain value linkage factor. Anomaly level linkage factor; if the absolute value of the residual mean of a certain satellite is greater than or the standard deviation of the residuals is greater than If the satellite is found to be in a persistent fault, all its frequency observations will be removed.

6. The low-orbit satellite GNSS anti-interference adaptive RAIM method according to claim 1, characterized in that: In step S5, the hard constraints of low-Earth orbit dynamics include: the track height h must satisfy: 300km≤h≤1200km; The velocity amplitude V must satisfy: 7000m / s≤V≤8000m / s; The position difference of the three axes ΔP between the two seconds before and after is ≤15km, and the velocity difference of the three axes ΔV between the two seconds before and after is ≤40m; If either the calculated position or velocity exceeds the above constraints, the calculation result for the current epoch is determined to be invalid.

7. The low-Earth orbit satellite GNSS anti-interference adaptive RAIM method according to claim 1, characterized in that: In step S6, the overall interference confidence level is... The expression is as follows: ； In the formula, , , , , , , , These are the anomaly coefficients for carrier-to-noise ratio, effective number of satellites, Doppler error rate, residual, AGC gain, cycle slip rate, clock error jump, and Doppler residual, respectively, and each coefficient is normalized to the interval [0, 100].

8. The low-orbit satellite GNSS anti-interference adaptive RAIM method according to claim 1, characterized in that: In step S6, the system automatically classifies interference levels based on the J value in real time and executes corresponding strategies. The graded handling strategies include: Level 0: If the error rate is less than 10%, the system will output real-time positioning and orbit determination results normally. Level 1: 10% ≤ If the detection threshold is less than 30%, the system will automatically remove the identified faulty satellites and tighten all detection thresholds. Level 2: 30% ≤ If the result is less than 70%, the weight of the orbital dynamics constraint is increased. If the final PVT solution is still invalid, the orbital extrapolation mode is entered. Level 3: ≥70% or both primary and backup machines are in an abnormal state for more than 3 seconds simultaneously or clock difference jump variable >1×10 -4 If the mean value of the s or Doppler residuals is >30 Hz or the orbital constraints are exceeded, the system will be forced to enter the orbital extrapolation mode. The orbit determination extrapolation mode performs continuous extrapolation for 6000 seconds. When 10 consecutive epochs are detected... When the percentage of satellites is less than 30% and the number of valid satellites is no less than 4, the system will automatically revert to GNSS positioning mode.

9. The low-orbit satellite GNSS anti-interference adaptive RAIM method according to claim 1, characterized in that: In step S7, the conditions for triggering autonomous baseline drift reset are that the following conditions are met simultaneously: continuous anomaly count ≥ 1200 seconds; the highest carrier-to-noise ratio of the current GPS or BDS system ≥ 42 dB-Hz; and the coefficient of variation of all features. ; The relative drift of any feature quantity is >0.3; The reset process includes: determining whether a stable state has been reached based on the fluctuation range of the measured values ​​within the current 30-second window and the overall coefficient of variation of the feature mean; if so, recording the feature mean within the window as a candidate baseline. Then perform a baseline reset operation: the new mean after reset. The new mean after resetting The number of samples in the Welford algorithm is reset to 10, and the second-order central moments are cleared to zero.

10. A low-Earth orbit (LEO) satellite GNSS anti-jamming adaptive RAIM system, applied to the LEO satellite GNSS anti-jamming adaptive RAIM method according to any one of claims 1-9, characterized in that: It adopts a three-layer architecture, including a hardware adaptation layer, a data interaction layer, and an anti-interference core algorithm layer; The hardware adaptation layer includes a hardware interface abstraction layer, compilation environment configuration, and computing power scheduling functions, which are used to decouple the core algorithm from different spaceborne processors and GNSS receiver models. The data interaction layer includes data synchronization, telemetry management, and parameter management functions, and is responsible for the input and output of internal and external data, as well as the configuration and management of on-orbit parameters. The anti-interference core algorithm layer includes the following functional modules: a primary and backup dual receiver redundancy management module, used to perform data synchronization and joint decision-making in step S1; and a dynamic threshold interference fast early warning module, used to perform feature extraction, online statistics, and threshold early warning in step S2. The observation domain fine anti-interference module is used to perform multi-dimensional quality detection and invalid value removal in step S3; the sliding window adaptive RAIM single-satellite fault isolation module is used to perform residual calculation, linkage threshold adjustment and faulty satellite removal in step S4; the extended epoch joint RAIM and orbit constraint module is used to perform PVT solution and dynamic hard constraint verification in step S5; the system decision and orbit extrapolation module is used to perform confidence calculation, hierarchical handling and extrapolation mode triggering in step S6; and the global baseline adaptation and telemetry module is used to perform baseline drift monitoring, reset and full-link status packet downlink in step S7.

Images