An interference classification identification system and method for Beidou multi-mode GNSS

Abstract

The application discloses a jamming type identification system and method for a Beidou multi-mode GNSS, and relates to the technical field of Beidou positioning. The system comprises a data acquisition module, a graph construction module, a base point determination module, a reconstruction module, a constraint module, a deviation analysis module and a jamming type module. The method acquires the receiving state data before and after the abnormal triggering moment, constructs a first touch node graph to determine a target disturbed frequency point; determines a reference frequency point according to the propagation correlation risk and observation stability of the same-star different-frequency frequency point; reconstructs the initial observation state of the target disturbed frequency point based on the reference frequency point, and constructs an observation constraint set; determines a standard conflict observation set through deletion test, outputs the type results of frequency point selective jamming, sweep jamming, wideband suppression jamming, deception jamming or non-propagating observation anomaly, and improves the accuracy of jamming type identification.

Description

Technical Field

[0001] This invention relates to the field of BeiDou positioning technology, specifically to an interference classification and identification system and method for BeiDou multi-mode GNSS. Background Technology

[0002] In the field of satellite navigation and positioning, with the development of BeiDou multi-mode GNSS technology, receivers can simultaneously receive navigation signals from multiple satellites and multiple frequencies, and generate multi-dimensional observation data such as positioning, pseudorange, Doppler, and carrier phase, providing basic support for high-precision positioning, timing, and navigation security. Therefore, interference identification and classification for these multi-frequency, multi-mode GNSS receivers has become an important research direction for ensuring positioning reliability.

[0003] When the receiver is affected by suppression interference, frequency sweeping interference, or spoofing signals, existing methods determine the anomaly by a single carrier-to-noise ratio drop, number of lock-outs, or changes in positioning residuals. However, this method is difficult to accurately identify different types of interference, especially when anomalies exist simultaneously on multiple frequency points and multiple satellite channels. It is impossible to clearly determine the target disturbed frequency point and the reliable reference frequency point, resulting in inaccurate interference classification. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides an interference classification and identification system and method for BeiDou multi-mode GNSS, solving the problems mentioned in the background.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a method for interference classification and identification of BeiDou multi-mode GNSS, comprising the following steps: Acquire the received data of the Beidou multi-mode GNSS receiver before and after the abnormal triggering time. The received data includes the received status data corresponding to different processing levels, different frequency points and different satellite channels. Based on the abnormal first contact times of each processing level, frequency point, and satellite channel in the received status data, a first contact node diagram is constructed; Based on the first contact node diagram, the target interference frequency point is determined, and abnormal propagation correlation analysis is performed on other frequency points of the same Beidou satellite to which the target interference frequency point belongs. Frequency points with abnormal propagation correlation with the target interference frequency point are eliminated, and the reference frequency point is determined. Based on the observation status of the reference frequency, reconstruct the initial observation status of the target's disturbed frequency under interference-free conditions; Based on the initial observation state, the actual observation state of the target disturbed frequency point, and the observation relationship between the same satellite and different frequencies, an observation constraint set is constructed. Based on the constraint deviation of each constraint condition in the observation constraint set, determine the standard conflict observation set; Based on the composition of the standard conflict observation set, the interference classification results of the BeiDou multi-mode GNSS receiver are determined.

[0006] Preferably, the specific process of constructing the first-touch node graph includes: According to the signal processing chain of the Beidou multi-mode GNSS receiver, the received status data is divided into data groups of multiple processing levels, and the graph nodes are determined based on the combination of processing level identifier, frequency point identifier, and satellite channel identifier. Establish a corresponding state time series for each graph node, and determine the first touch time of an anomaly for each graph node based on the first deviation time of the received state data from the anomaly judgment condition in the state time series. Based on the order of the first contact time of the anomaly at each graph node, the propagation direction of the anomaly from the first triggered graph node to the second triggered graph node is determined; A first-touch node graph is constructed based on the graph nodes, the first-touch time of the anomaly, and the propagation direction.

[0007] Preferably, determining the target interference frequency point based on the first-touch node diagram specifically includes: Based on the abnormal first contact time and propagation direction of each frequency point corresponding to the node in the first contact node diagram, candidate interference frequency points on the abnormal propagation path are determined. Based on the degree of deviation of the receiving state of the candidate interference frequency points after the abnormal trigger time from the reference state before the abnormal trigger time, the target interference frequency point is determined from the candidate interference frequency points.

[0008] Preferably, the logic for determining the reference frequency is as follows: Obtain other observable frequency points of the same Beidou satellite to which the target disturbed frequency point belongs, construct a set of candidate frequency points of different frequencies of the same satellite, and map the target disturbed frequency point and each candidate frequency point in the set of candidate frequency points of different frequencies of the same satellite to the first contact node graph respectively to obtain the target graph node set and the candidate graph node set; In the first-touch node graph, a directed path search is performed on the target graph node set and each candidate graph node set to obtain the path connectivity relationship, and the abnormal first-touch time sequence, processing level change sequence and frequency point change sequence are extracted. First-arrival timing consistency analysis is performed on the abnormal first-touch time sequence, cross-layer propagation analysis is performed on the processing level change sequence, and cross-frequency propagation analysis is performed on the frequency point change sequence to obtain the propagation association risk value of each candidate frequency point relative to the target disturbed frequency point; If the propagation association risk value is greater than the preset risk threshold, the corresponding candidate frequency point will be removed from the set of candidate frequency points of the same star but different frequencies, and the observation stability analysis will be performed on the remaining candidate frequency points to obtain the observation stability evaluation value. Based on the propagation association risk value and the observation stability evaluation value, the baseline confidence level of each remaining candidate frequency point is calculated, and the remaining candidate frequency points whose baseline confidence level meets the preset confidence conditions are determined as the baseline frequency points.

[0009] Preferably, reconstructing the initial observation state of the target disturbed frequency point under interference-free conditions specifically includes: Based on the change in the observation state of the reference frequency point before and after the abnormal triggering time, and the observation correlation between different frequency points of the same Beidou satellite, the change in the observation state of the target disturbed frequency point under the interference-free condition is determined; based on the actual observation state of the target disturbed frequency point before the abnormal triggering time and the change in the observation state under the interference-free condition, the initial observation state of the target disturbed frequency point is generated.

[0010] Preferably, the process of constructing the observation constraint set is as follows: Calculate the deviation between the initial observation state and the actual observation state of the target disturbed frequency point, and generate the initial observation residual; Generate initial observation constraints based on the initial observation residuals; Based on the same-satellite, different-frequency observation relationship between the target disturbed frequency point and the reference frequency point, generate same-satellite, different-frequency observation constraint conditions. Add the initial observation constraints and the same-star, different-frequency observation constraints to the observation constraint set; The initial observation constraint is used to limit the deviation between the actual observation state of the target disturbed frequency point and the initial observation state, and the same-satellite different-frequency observation constraint is used to constrain the observation consistency between different frequency points of the same Beidou satellite.

[0011] Preferably, the constraint satisfaction of the observation constraint set is determined to identify the standard conflict observation set, specifically including: For each constraint in the set of observation constraints, calculate the constraint deviation amount and mark the constraint deviation amount that exceeds the corresponding preset allowable range as conflict constraint conditions; Extract the observations involved in the calculation of conflict constraints, the conflict constraints, the processing level identifiers, frequency identifiers, satellite channel identifiers, and the first contact time of anomalies to construct a set of candidate conflict objects; The observations and conflict constraints in the candidate conflict object set are taken as the conflict objects to be tested, and the deletion test is performed on each conflict object to be tested according to the preset deletion order. The deletion test includes: deleting one conflict object to be tested from the candidate conflict object set according to a preset deletion order; recalculating the constraint deviation based on the remaining observations and remaining conflict constraints after deletion; and determining whether there are still conflict constraints whose constraint deviation exceeds the corresponding preset allowable range after deletion. If there are conflict constraints after deleting the conflict object to be tested, the conflict object to be tested is deleted from the candidate conflict object set. If there are no conflict constraints after deleting the conflict object to be tested, the conflict object to be tested is retained in the candidate conflict object set. Repeat the elimination test until any candidate conflict object in the candidate conflict object set is eliminated, and the remaining candidate conflict object set after repeated elimination tests is determined as the standard conflict observation set.

[0012] Preferably, determining the type of interference experienced by the BeiDou multi-mode GNSS receiver specifically includes: Map each observation and constraint in the standard conflict observation set to the first contact node diagram; Determine the overlap between graph nodes corresponding to the standard conflict observation set and the anomaly propagation path, and determine whether the standard conflict observation set belongs to the anomaly propagation path, specifically: When the standard conflict observation set belongs to an abnormal propagation path, the distribution structure of the standard conflict observation set in the frequency point dimension, satellite channel dimension and processing level dimension is analyzed to determine the interference classification result. The interference classification result includes frequency-selective interference, frequency sweeping interference, broadband suppression interference and deception interference. When the standard conflict observation set does not correspond to an anomaly propagation path, the corresponding anomaly is identified as a non-propagating observation anomaly.

[0013] An interference classification and identification system for BeiDou multi-mode GNSS includes: The data acquisition module is used to acquire the received data of the Beidou multi-mode GNSS receiver before and after the abnormal triggering time. The received data includes the receiving status data corresponding to different processing levels, different frequency points and different satellite channels. The graph construction module is used to construct a first-touch node graph based on the abnormal first-touch times of each processing level, frequency point, and satellite channel in the received status data; The base point determination module is used to determine the target disturbed frequency point based on the first contact node diagram, and to perform abnormal propagation correlation analysis on other frequency points of the same Beidou satellite to which the target disturbed frequency point belongs, eliminate frequency points that have abnormal propagation correlation with the target disturbed frequency point, and determine the reference frequency point; The reconstruction module reconstructs the initial observation state of the target disturbed frequency point under interference-free conditions based on the observation state of the reference frequency point. The constraint module is used to construct a set of observation constraints based on the initial observation state, the actual observation state of the target disturbed frequency point, and the observation relationship between the same satellite and different frequencies; The deviation analysis module determines the standard conflict observation set based on the constraint deviation of each constraint in the observation constraint set; The interference classification module is used to determine the interference classification results of the BeiDou multi-mode GNSS receiver based on the composition structure of the standard conflict observation set.

[0014] This invention provides an interference pattern identification system and method for BeiDou multi-mode GNSS, which has the following beneficial effects: By recording the sequence of anomalies in the first-touch node diagram, the satellite channels, frequencies, and processing levels affected by the interference can be determined first, avoiding misjudgments based solely on carrier-to-noise ratio, number of lock-outs, or changes in positioning residuals. By screening reference frequencies, frequency points on the same satellite but at different frequencies that have been affected by the same interference are excluded, making observation comparisons more accurate. By reconstructing the expected observation state under interference-free conditions and extracting key conflict observation evidence, interference from irrelevant anomaly indicators is reduced, thereby improving the accuracy of the BeiDou multi-mode GNSS receiver in classifying different types of interference. Attached Figure Description

[0015] Figure 1 This is a flowchart of an interference classification and identification method for BeiDou multi-mode GNSS according to the present invention; Figure 2 This is a block diagram of an interference classification and identification system for BeiDou multi-mode GNSS according to the present invention. Detailed Implementation

[0016] To make the objectives, technical solutions, and beneficial effects of this invention clearer, the invention will be described in detail below with reference to specific embodiments. The time window, sampling frequency, threshold, weight, number of iterations, and duration epochs in the following embodiments are preferred values ​​for this embodiment, used to illustrate the specific implementation of the method, and do not constitute a limitation on the scope of protection of this invention. Those skilled in the art can make equivalent adjustments to the corresponding values ​​based on the receiver sampling rate, BeiDou signal system, application scenario, dynamic carrier speed, antenna type, and the types of output state variables within the receiver.

[0017] In this invention, a BeiDou multi-mode GNSS receiver refers to a receiving device that can receive BeiDou satellite navigation signals and support processing of at least two frequency points or at least two signal modes.

[0018] The receiver supports at least two of the following frequency points: B1I, B1C, B2a, and B2b. The receiver output data includes RF front-end power, automatic gain control value, acquisition correlation peak, correlation peak ratio, carrier-to-noise ratio, code loop error, carrier loop error, lock-in status, pseudorange, carrier phase, Doppler, pseudorange change rate, observation quality indicator, positioning residual, and satellite weight.

[0019] Example Please see Figure 1 and Figure 2 This invention provides an interference pattern identification method for BeiDou multi-mode GNSS, comprising the following steps: S1. Obtain the received data of the Beidou multi-mode GNSS receiver before and after the abnormal triggering time. The received data includes the received status data corresponding to different processing levels, different frequency points and different satellite channels. S101, The receiver continuously collects BeiDou multi-frequency reception status data; In this embodiment, the received status data is recorded epochally. The sampling frequency of the positioning calculation data is 1Hz, the sampling frequency of the tracking loop status data is 10Hz, and the sampling frequency of the RF front-end power and automatic gain control data is 10Hz. If the receiver can only output 1Hz status data, all status quantities are processed at 1Hz; if the receiver can output 50Hz or 100Hz loop data, it is first averaged or medianized at 0.1s time intervals before being written into the unified status table.

[0020] S102. Determine the time when the exception is triggered; When any frequency point or any satellite channel meets at least one of the following conditions, the receiver or host computer records the time of the abnormal trigger: The carrier-to-noise ratio decreases by no less than 4 dB-Hz relative to the baseline value before the anomaly over three consecutive epochs. The automatic gain control value changes by no less than 3dB within 1 second; The power of the RF front-end increases by no less than 5dB within 1 second; The capture-related peak ratio decreased by no less than 20% relative to the pre-anomaly average. The code loop error or carrier loop error exceeds three times the standard deviation before the anomaly for three consecutive epochs; Two or more consecutive unlock signs appear on the same satellite channel; The positioning residual exceeded three times the mean before the anomaly within three consecutive epochs, and involved more than two satellites.

[0021] S103, Extract data before and after the anomaly; Centered on the time T0 when the anomaly is triggered, data from 30 seconds before T0 to 60 seconds after T0 is extracted as the data for this anomaly event. The 30 seconds before T0 is used as the baseline window before the anomaly, and the 60 seconds after T0 is used as the anomaly evolution window.

[0022] A 30-second reference window can cover 30 1Hz positioning epochs or 300 10Hz loop epochs, which is sufficient to calculate the normal fluctuation range of carrier-to-noise ratio, pseudorange change rate, Doppler, and loop error. A 60-second anomalous evolution window can cover the main process of suppression interference, frequency sweeping interference, and spoofing interference propagating from the tracking layer to the observation generation layer and then to the positioning solution layer within the receiver.

[0023] S104. Establish a baseline state table before the anomaly; For each data object determined by satellite number, frequency, processing level, and status variable, read the time series within the baseline window before the anomaly and generate the following baseline fields respectively: Continuous state variables record the mean, median, standard deviation, maximum value, minimum value, and maximum change between adjacent epochs; The flag-type state variables record the proportion of normal states, the number of abnormal states, and the longest consecutive abnormal epoch. The observation status records include data integrity rate, cycle slip count, number of lockouts, and number of abnormal observation quality indicators.

[0024] S105. Output the receive status data table and the baseline status table before the abnormality; The received status data table should include at least the event number, timestamp, satellite number, frequency, processing level, status variable name, actual status variable value, and quality flag. The baseline status table before the anomaly should include at least the event number, satellite number, frequency, processing level, status variable name, baseline mean, baseline median, baseline standard deviation, normal fluctuation range, data integrity rate, and baseline reliability flag.

[0025] S2. Based on the abnormal first contact times of each processing level, frequency point, and satellite channel in the received status data, construct a first contact node diagram; S201, Generate graph nodes; First-touch node diagram nodes are generated according to satellite number, frequency, and processing level. The processing levels include the radio frequency front-end layer, automatic gain control layer, acquisition layer, tracking loop layer, observation generation layer, and positioning calculation layer.

[0026] For example, the C21-B1C tracking loop layer constitutes one graph node, the C21-B1C observation generation layer constitutes another graph node, and the C21-B2a tracking loop layer constitutes yet another graph node. Each graph node records the node number, satellite number, frequency, processing level, included state variables, node quality flag, and whether the node is valid.

[0027] S202. Determine the first contact time of an abnormal graph node; After each graph node experiences an anomaly, the state variables within the window are compared with the baseline state table before the anomaly. If at least one core state variable in a graph node continuously reaches the anomaly condition, the start time of this continuous anomaly is recorded as the first anomaly touch-out time for that graph node.

[0028] In this embodiment, the core state variables of the tracking loop layer include carrier-to-noise ratio (CNR), code loop error, carrier loop error, and lock status. A tracking layer anomaly is recorded when the CNR is lower than the pre-anomaly median minus 4 dB-Hz and remains at least 0.3 s. Similarly, a tracking layer anomaly is recorded when the code loop error or carrier loop error exceeds three times the pre-anomaly standard deviation and remains at least 0.3 s.

[0029] The core state variables of the observation generation layer include pseudorange change rate, Doppler, carrier phase continuity, and observation quality indicators. An observation generation layer anomaly is recorded when the pseudorange change rate exceeds the normal range of the same satellite at different frequencies by more than 0.5 m / s and remains so for three consecutive 1 Hz epochs; an observation generation layer anomaly is also recorded when the Doppler deviates from the normal range of the same satellite at more than 0.8 Hz and remains so for three consecutive 1 Hz epochs.

[0030] The core state variables of the positioning solution layer include positioning residuals, satellite weights, and satellite removal status. If the residual corresponding to a satellite exceeds three times the average residual before the anomaly, and the satellite weight decreases by no less than 30%, it is recorded as an anomaly in the positioning solution layer.

[0031] The process of obtaining the first contact time of the above-mentioned anomaly is as follows: first, determine the start time of the first consecutive overrun of each core state variable in the node, and then select the earliest time from these start times as the first contact time of the node's anomaly.

[0032] S203. Establish candidate propagation edges; Candidate propagation edges are established according to the processing chain order, the relationship between different frequencies of the same satellite, the relationship between satellites with the same frequency, and the relationship between common upstream components.

[0033] When, on the same satellite and at the same frequency, the first contact time of an anomaly at the previous processing level node is earlier than that at the next processing level node, and the time difference is between 0 and 5 seconds, a candidate propagation edge for the processing chain is established. The 5-second upper limit is set because GNSS tracking anomalies usually propagate to the observation generation layer and the positioning solution layer within several epochs, and anomalies exceeding 5 seconds are more likely to be affected by other factors.

[0034] When the time difference between the first contact of an anomaly at two frequencies on the same satellite is between 0 and 10 seconds, a candidate propagation edge for the same satellite but at different frequencies is established. The 10-second upper limit is set to account for the gradual sweeping of frequency interference across different frequencies; if the time difference exceeds 10 seconds, the correlation of the same anomaly event is significantly reduced.

[0035] When multiple satellite channels exhibit anomalies at similar times on the same frequency, and the time difference between their first arrivals is no greater than 2 seconds, a candidate propagation edge across satellites on the same frequency is established. A 2-second time difference corresponds to a common manifestation of suppression-type interference or frequency-selective interference occurring approximately synchronously on multiple channels at the same frequency.

[0036] S204. Calculate the propagation correlation strength; The propagation association strength is obtained for each candidate propagation edge. The propagation association strength is jointly determined by the degree of temporal proximity, the degree of hierarchical order matching, the degree of consistency of state changes, and the common upstream relationship.

[0037] The time proximity is determined as follows: if the time difference between the first contact of two nodes is no greater than 1 second, it is considered high time proximity; if it is greater than 1 second but no greater than 5 seconds, the proximity gradually decreases as the time difference increases; processing links exceeding 5 seconds are not retained. The hierarchical order matching is determined as follows: if the anomaly propagation direction conforms to the processing order from the RF front-end layer to the automatic gain control layer, acquisition layer, tracking loop layer, observation generation layer, and positioning calculation layer, it is considered high matching; if the directions are opposite, it is not considered a valid processing link. The consistency of state variable changes is determined as follows: if both preceding and following nodes exhibit signal quality degradation, increased error, or deteriorated quality indicators, they are considered consistent; if the directions of change contradict each other, the correlation strength is reduced.

[0038] In this embodiment, the propagation association strength is between 0 and 1. Time proximity accounts for 0.35, hierarchical order matching accounts for 0.30, consistency of state changes accounts for 0.20, and common upstream relationship accounts for 0.15. When the propagation association strength is greater than or equal to 0.60, the candidate propagation edge is retained as a valid propagation edge.

[0039] A threshold of 0.60 is set because temporal sequence alone is insufficient to prove anomalous propagation; at least one of the following—hierarchy, direction of state change, or common upstream relationship—needs to provide supporting evidence. A threshold of 0.60 can eliminate most candidate edges that are only temporally adjacent but have unrelated technical mechanisms.

[0040] S205, Output the first contact node diagram; The first-touch node graph includes graph nodes, abnormal first-touch times, effective propagation edges, propagation direction, propagation time difference, propagation association strength, and common upstream node numbers.

[0041] S3. Determine the target interference frequency point based on the first contact node diagram. S301. Aggregate abnormal nodes by frequency point; Read the first-touch node graph, aggregate the abnormal nodes under the same frequency point to obtain the aggregated data of the corresponding frequency point, including the number of abnormal nodes, the number of abnormal satellite channels, the number of abnormal processing levels, the earliest abnormal first-touch time, the number of effective outgoing edges, the number of effective incoming edges, the length of the longest propagation path, the observation deviation status, and the location calculation association status.

[0042] For example, if anomalies occur in the tracking loop layers of satellite channels C21, C22 and C25 at the B1C frequency point, and anomalies also occur in the C21-B1C observation generation layer and positioning solution layer, then the number of anomaly processing layers, the number of abnormal satellite channels and the propagation path length at the B1C frequency point are all relatively high.

[0043] S302. Screening candidate disturbed frequency points; If a frequency point meets at least two of the following conditions, it will be included in the candidate disturbed frequency point set: The earliest anomaly at this frequency point is within 2 seconds after the earliest anomaly in the entire graph. This frequency anomaly covers at least two processing levels; This frequency point has at least two effective propagation edges; At this frequency point, at least one observation has deviated continuously by more than 3 epochs relative to the pre-abnormal baseline state. The anomalous node at this frequency can be connected to the observation generation layer or the location calculation layer through effective propagation edges.

[0044] S303, Obtain the first-pass dominant evaluation value; For each candidate disturbed frequency point, obtain the first-arrival dominant evaluation value A1. A1 is used to indicate whether the frequency point is at the abnormal propagation front.

[0045] The process for obtaining A1 is as follows: First, read the earliest first-contact time of an anomaly in the entire image, Tmin, and then read the earliest first-contact time of an anomaly within the candidate frequency point, Tf. If the difference between Tf and Tmin is no greater than 1 second, the first-arrival time value is 1.00; if the difference is greater than 1 second but no greater than 5 seconds, the value is linearly reduced from 1.00 to 0.20; if the difference is greater than 5 seconds, the first-arrival time value is 0.10. Then, read the number of valid outgoing edges for that frequency point. If the number of valid outgoing edges is no less than 3, the propagation output value is 1.00; if the number of valid outgoing edges is 1 to 2, the propagation output value is 0.50 to 0.80; if there are no valid outgoing edges, the value is 0.

[0046] In this embodiment, A1 is obtained by weighting the first arrival time portion and the propagation output portion, with the first arrival time portion accounting for 0.65 and the propagation output portion accounting for 0.35.

[0047] S304. Obtain the hierarchical connectivity evaluation value A2; The number of processing layers for abnormal coverage of candidate disturbed frequency points is counted, and the existence of effective propagation edges between layers is checked.

[0048] If the candidate frequency point exhibits anomalies in only one processing level, A2 should not exceed 0.30; if it covers two processing levels and has a valid propagation edge, A2 should be between 0.50 and 0.65; if it covers more than three processing levels and forms a propagation path from the tracking loop layer to the observation generation layer and then to the positioning solution layer, A2 should be between 0.80 and 1.00. If the anomaly starts from the RF front-end layer or the automatic gain control layer and propagates backward to the tracking loop layer and the observation generation layer, A2 should preferably be 0.90 or higher.

[0049] If multiple levels are all abnormal, but there is no time sequence or effective propagation edge, then A2 decreases.

[0050] S305. Obtain the observation deviation evaluation value A3; Read the actual observations of the candidate frequency points within the window after the anomaly and compare them with the baseline state before the anomaly to obtain the observation deviation evaluation value A3.

[0051] In this embodiment, carrier-to-noise ratio (CNR) deviation, pseudorange rate of change deviation, Doppler deviation, carrier phase continuity disruption, and observation quality indicator anomalies are respectively formed into sub-evaluations. When the CNR continuously decreases by at least 4 dB-Hz for more than 0.3 s, the sub-evaluation is rated 0.80 or higher; when the pseudorange rate of change continuously exceeds the normal fluctuation range before the anomaly by more than 0.5 m / s for three 1 Hz epochs, the sub-evaluation is rated 0.80 or higher; when the Doppler deviation continuously exceeds the normal fluctuation range before the anomaly by more than 0.8 Hz for three 1 Hz epochs, the sub-evaluation is rated 0.80 or higher; when the carrier phase exhibits cycle slips, phase interruptions, or continuous loss of lock, the sub-evaluation is rated 0.70 or higher.

[0052] The above sub-evaluations are then weighted and summarized. Tracking loop status accounts for 0.30, pseudorange change rate accounts for 0.25, Doppler accounts for 0.20, carrier phase continuity accounts for 0.15, and observation quality indicator accounts for 0.10. This weighting is because tracking loop status directly reflects the quality of the received signal, pseudorange change rate and Doppler directly reflect the dynamic relationship of the observed data, and carrier phase continuity and observation quality indicator are used to enhance the judgment of anomaly persistence.

[0053] S306. Obtain the impact evaluation value A4; Determine whether the anomaly of the candidate frequency point affects the positioning solution. If the positioning residual of the satellite corresponding to the frequency point exceeds three times the average value before the anomaly, and the satellite weight decreases by no less than 30%, A4 should be set to 0.80 or higher. If only the positioning residual increases but the weight does not decrease, A4 should be set to 0.40 to 0.60. If the anomaly of the frequency point has not yet been propagated to the positioning solution layer, A4 should be set to 0 to 0.30.

[0054] S307. Calculate the comprehensive disturbance evaluation value and determine the target disturbance frequency point; In this embodiment, the comprehensive disturbance evaluation value F is obtained according to the following weights: the first-reaching dominant evaluation value A1 accounts for 0.30, the hierarchical connection evaluation value A2 accounts for 0.25, the observation deviation evaluation value A3 accounts for 0.30, and the solution impact evaluation value A4 accounts for 0.15.

[0055] When F is greater than or equal to 0.65, the candidate frequency point is determined as the target disturbed frequency point. Set 0.65 as the target threshold.

[0056] When multiple frequency points have an F value greater than or equal to 0.65, compare the F values. If the highest F value is more than 0.10 higher than the second highest F value, then only the highest frequency point is considered as the primary target interference frequency point; if the difference is less than 0.10, then all frequency points are considered as target interference frequencies points.

[0057] S308, Output the target interference frequency record; The target interference frequency record includes the event number, target frequency, satellite number, target map node set, A1, A2, A3, A4, comprehensive interference evaluation value F, trigger threshold, determination basis, and main target identifier.

[0058] S4. Perform abnormal propagation correlation analysis on other frequency points of the same Beidou satellite to which the target disturbed frequency point belongs, eliminate frequency points that have abnormal propagation correlation with the target disturbed frequency point, and determine the reference frequency point; S401. Establish a set of candidate frequency points for different frequencies on the same satellite; Based on the satellite number of the target disturbed frequency point, other observable frequency points on the same BeiDou satellite, excluding the target disturbed frequency point, are read to form a set of candidate frequency points of different frequencies on the same satellite.

[0059] For example, if the target interference frequency is C21-B1C, and the receiver simultaneously observes C21-B1I, C21-B2a, and C21-B2b, then C21-B1I, C21-B2a, and C21-B2b will all enter the candidate set.

[0060] S402. Perform a basic availability check on candidate frequency points; Candidate frequency points need to meet the following basic conditions: The data integrity rate within the baseline window prior to the anomaly is no less than 90%. The percentage of lockout statuses that are functioning normally is no less than 95%. The standard deviation of the carrier-to-noise ratio before the anomaly is no greater than 1.5 dB-Hz; The number of cycle jumps should not exceed 1 within 30 seconds prior to the abnormality. The proportion of anomalies in the observed quality indicators should not exceed 5%.

[0061] The logic for setting the conditions is as follows: the reference frequency is used to estimate the interference-free state of the target frequency. If the reference frequency itself is unstable before the anomaly, the subsequent initial state will lose its reliable basis. A 95% lock-in normal ratio ensures that the candidate frequency is in continuous tracking state within most of the reference window; a 1.5dB-Hz carrier-to-noise ratio standard deviation is used to exclude large fluctuations caused by obstruction, multipath, or low elevation angle.

[0062] Candidate frequency points that do not meet the basic availability conditions will not be used as reference frequency points; if other frequency points on the same satellite do not meet the basic conditions, a reference frequency point shortage flag will be output.

[0063] S403, Obtain the first arrival time sequence risk R1; Read the earliest abnormal first contact time Tt of the target disturbed frequency point and the earliest abnormal first contact time Tc of the candidate frequency point.

[0064] If there are no abnormal nodes among the candidate frequencies, R1 is set to 0.10. If the absolute value of the time difference between Tc and Tt is no greater than 1 second, R1 is set to 1.00, indicating that the two frequencies are almost synchronously abnormal, and the candidate frequencies are highly likely to be affected by the same abnormality. If the time difference is greater than 1 second but no greater than 5 seconds, R1 gradually decreases from 0.80 to 0.30 based on the time difference. If the time difference is greater than 5 seconds but no greater than 10 seconds, R1 is set to 0.20. If the time difference exceeds 10 seconds, R1 is set to 0.10.

[0065] The risk assessment process involves determining whether the target frequency and candidate frequency are likely within the propagation range of the same anomalous event by comparing their occurrence times. The closer the times, the higher the risk of contamination of the candidate frequency. When the candidate frequency has no anomalies, the timing risk is lowest but not directly zero, as there may still be slight contamination that has not reached the threshold.

[0066] S404, Obtain the risk of cross-layer propagation R2; If there are no abnormal nodes in the candidate frequency, R² is 0.05; if there is only a short-term abnormality in a single layer, R² is 0.20 to 0.35; if the candidate frequency covers two processing levels and there are effective propagation edges, R² is 0.50 to 0.70; if the candidate frequency covers more than three processing levels and the propagation order is consistent with the target frequency, R² is 0.80 to 1.00.

[0067] This risk is designed to avoid mistakenly using similarly disturbed frequencies as a benchmark. If a candidate frequency forms a propagation structure along the tracking layer, observation layer, and solution layer, it indicates that it has been affected by the anomaly chain and should not be used as a high-confidence benchmark.

[0068] S405, Acquire the risk of cross-frequency propagation (R3); Search for effective cross-frequency propagation edges or cross-frequency paths between target frequency point graph nodes and candidate frequency point graph nodes in the first-touch node graph.

[0069] If a valid path exists from the target frequency to the candidate frequency, and the propagation time difference is no greater than 10 seconds, then R3 is set to 0.80 to 1.00. If a valid path exists from the candidate frequency to the target frequency, it indicates that the candidate frequency may be abnormally upstream, and R3 is also set to 0.80 or higher. If there is no direct path between the two, but there is a common frequency extension chain, R3 is set to 0.40 to 0.70. If there are no cross-frequency edges or cross-frequency paths, R3 is set to 0.05 to 0.20.

[0070] S406, Obtaining the common upstream pollution risk R4; Analyze whether the target frequency and candidate frequency have the same abnormal nodes in the RF front-end, automatic gain control, or common processing module.

[0071] If the target frequency and candidate frequency share the same RF front-end anomalous node, and the first arrival of the anomalous node precedes the tracking layer anomalous nodes of both, R4 is set to 0.90 to 1.00. If they share an automatic gain control anomalous node, R4 is set to 0.70 to 0.90. If there is only a weak common upstream, such as a slight increase in noise floor during the same period but not reaching the front-end anomalous threshold, R4 is set to 0.30 to 0.50. If there is no common upstream, R4 is set to 0.05.

[0072] S407, Calculate the propagation-related risk R; In this embodiment, the propagation associated risk R is obtained as follows: the first arrival time sequence risk R1 accounts for 0.25, the cross-layer propagation risk R2 accounts for 0.25, the cross-frequency propagation risk R3 accounts for 0.25, and the common upstream pollution risk R4 accounts for 0.25.

[0073] If broadband suppression interference is common in the application scenario, the weight of R4 can be increased to 0.35; if frequency sweeping interference is common, the weight of R3 can be increased to 0.35. When the propagation association risk R is greater than 0.45, the candidate frequency point is marked as having a high risk of propagation contamination and is not used as the primary reference frequency point. The logic of setting 0.45 as the threshold is that if two of the four types of risk reach medium or above, the candidate frequency point has a significant possibility of being contaminated and is not used as a strong reference.

[0074] S408. Obtain the observation stability evaluation value S; For candidate frequency points with a propagation association risk R not greater than 0.45, obtain the observation stability evaluation value S.

[0075] S is determined by both pre-anomaly stability and post-anomaly stability. Pre-anomaly stability includes data integrity rate, normal lock ratio, carrier-to-noise ratio fluctuation, Doppler continuity, and cycle slip; post-anomaly stability includes whether the lock remains within the reconstruction window after the anomaly, whether the carrier-to-noise ratio remains stable, whether the pseudorange change rate is continuous, whether the Doppler is smooth, and whether the observation quality indicators are normal.

[0076] In this embodiment, stability before the anomaly accounts for 0.45, and stability after the anomaly accounts for 0.55. The higher weight of stability after the anomaly is because the reference frequency is actually used for the initial reconstruction after the anomaly, and whether the frequency is pure and stable after the anomaly directly affects the reconstruction result.

[0077] If the data integrity rate is not less than 95% before the candidate frequency point anomaly, the normal lock-in rate is not less than 98%, the standard deviation of the carrier-to-noise ratio is not greater than 1.0 dB-Hz, and no cycle slip or loss of lock-in occurs after the anomaly, then S is taken as 0.85 to 1.00. If the candidate frequency point is basically stable but has a small number of missing measurements or slight fluctuations, then S is taken as 0.65 to 0.85. If the fluctuations are significant after the candidate frequency point anomaly, then S is less than 0.65.

[0078] S409. Calculate the baseline confidence level C and determine the baseline frequency point; The baseline reliability C increases with the observation stability S and decreases with the propagation association risk R. In this embodiment, C is obtained by calculating 0.60×S+0.40×(1-R), where the stability weight is 0.60, and the baseline frequency used for initial reconstruction must first have a continuous, complete, and usable observation state; the propagation association risk weight is 0.40, used to exclude the effects of common contamination and cross-frequency propagation.

[0079] When C is greater than or equal to 0.70, the candidate frequency point is determined as the reference frequency point. If multiple candidate frequency points all satisfy C not less than 0.70, the one with the highest C is used as the primary reference frequency point, and the rest are used as auxiliary reference frequency points. If no candidate frequency point satisfies C not less than 0.70, an insufficient reference frequency point flag is output, and a downgraded initial reconstruction is performed in S5.

[0080] S410, Output reference frequency point record; The baseline frequency record includes candidate frequency, R1, R2, R3, R4, propagation association risk R, observation stability S, baseline reliability C, whether it is used as a baseline frequency, whether it is the primary baseline frequency, and the reason for removal.

[0081] S5. Based on the observation status of the reference frequency point, reconstruct the initial observation status of the target disturbed frequency point under interference-free conditions; S501. Determine the reconstructed observations; The observations involved in the initial reconstruction include carrier-to-noise ratio, pseudorange rate of change, Doppler, carrier phase continuity, and observation quality status. If the receiver can output the acquisition correlation peak, correlation peak ratio, and AGC value, these are used as auxiliary observations.

[0082] Carrier-to-noise ratio is used to determine suppression type or frequency-selective interference; pseudorange change rate and Doppler are mainly used to determine dynamic consistency; carrier phase continuity is used to determine whether tracking has been disrupted; observation quality status is used to confirm whether the receiver has determined the observation anomaly.

[0083] S502. Establish the correlation between the same star and different frequencies before the anomaly; Read the same observation sequence of the target's disturbed frequency and the reference frequency within 30 seconds prior to the anomaly. For carrier-to-noise ratio, calculate the average difference and standard deviation of the difference between the target frequency and the reference frequency. For pseudorange change rate and Doppler, calculate the proportional relationship, average difference, and standard deviation of the difference between the two changes. For carrier phase continuity and observation quality status, calculate the proportion of times both are simultaneously normal.

[0084] For example, if the carrier-to-noise ratio of C21-B1C is on average 2.0 dB-Hz lower than that of C21-B2a within 30 seconds before the anomaly, and the standard deviation of the difference is 0.6 dB-Hz, then the normal difference in the carrier-to-noise ratio between the two is recorded as -2.0 dB-Hz, and the normal fluctuation range is ±1.8 dB-Hz. ±1.8 dB-Hz is obtained by three times the standard deviation, which can cover most normal short-term fluctuations.

[0085] For Doppler, if the average Doppler difference between B1C and B2a before the anomaly is 0.2Hz and the standard deviation is 0.15Hz, then the normal range of difference can be taken as 0.2Hz ± 0.45Hz. Using three times the standard deviation as the normal range is to avoid misjudging normal measurement noise as interference deviation.

[0086] S503, Generate the initial carrier-to-noise ratio; For each epoch after the anomaly, the current carrier-to-noise ratio (CNR) of the reference frequency is read, and the difference between the average CNR of the target frequency and the reference frequency before the anomaly is added to obtain the initial CNR of the target frequency under interference-free conditions.

[0087] For example, if the carrier-to-noise ratio (CNR) of B1C is 2.0 dB-Hz lower than that of B2a before the anomaly, and the CNR of B2a is 41.0 dB-Hz at a certain epoch after the anomaly, then the initial CNR of B1C is 39.0 dB-Hz. If the actual CNR of B1C is 33.5 dB-Hz, then the actual value is 5.5 dB-Hz lower than the initial value. The 5.5 dB-Hz value is compared with the allowable deviation range of the CNR in S6 to determine whether an initial constraint conflict has occurred.

[0088] S504. Generate the initial pseudorange rate of change; For the pseudorange change rate, first read the pseudorange change rate of the target frequency and the pseudorange change rate of the reference frequency at the reference time before the anomaly, then read the change of the reference frequency relative to the reference time at each epoch after the anomaly. Based on the ratio of the changes of the two frequency points before the anomaly, map the change of the reference frequency to the target frequency to obtain the initial pseudorange change rate of the target frequency.

[0089] For example, before the anomaly, the ratio of the pseudorange change rate between the target frequency and the reference frequency was 1.02, and the pseudorange change rate of the target frequency was -320.20 m / s. After the anomaly, the reference frequency changed by 0.40 m / s relative to the reference time at a certain epoch. Therefore, the target frequency should change by approximately 0.408 m / s under interference-free conditions, with an initial pseudorange change rate of -319.792 m / s. If the actual pseudorange change rate of the target frequency is -318.90 m / s, then the actual value deviates from the initial value by approximately 0.892 m / s.

[0090] Through the above process, the implementers gradually obtain the initial pseudorange change rate based on the pre-anomaly ratio, the post-anomaly change of the reference frequency, and the target frequency reference value.

[0091] S505, Generate initial Doppler; The initial Doppler state is generated as follows: first, obtain the Doppler difference and change ratio between the target frequency and the reference frequency before the anomaly, then read the Doppler change of the reference frequency after the anomaly and map it to the target frequency.

[0092] If the Doppler difference between the target frequency and the reference frequency is stable at 0.2Hz before the anomaly, and the reference frequency Doppler is -1680.5Hz at a certain epoch after the anomaly, then the initial Doppler of the target frequency can be taken as -1680.3Hz. If the actual Doppler of the target frequency is -1678.9Hz, then the deviation is 1.4Hz. If this deviation exceeds 0.8Hz for three consecutive epochs, then a Doppler initial constraint conflict can be formed.

[0093] S506. Generate the initial carrier phase continuity and observation quality status; If both the target frequency and the reference frequency are continuously locked before the anomaly, and the reference frequency remains continuously locked after the anomaly, then the initial carrier phase state of the target frequency is set to continuous. If the target frequency actually experiences a cycle slip, phase interruption, or loss of lock, then the actual state conflicts with the initial state.

[0094] If the observation quality indicators of both the target frequency and the reference frequency are normal before the anomaly, and the reference frequency remains normal after the anomaly, then the initial observation quality status of the target frequency is set to normal. If the actual quality indicator of the target frequency is continuously abnormal, then an initial deviation in observation quality is formed.

[0095] S507, Perform multi-reference frequency point fusion; When there are multiple reference frequency points, candidate initial states are generated based on each reference frequency point, and then weighted fusion is performed according to the reference credibility.

[0096] For example, if the confidence level of benchmark B2a is 0.86 and that of benchmark B2b is 0.74, then the weight of B2a is greater than that of B2b. First, the confidence levels of each benchmark are normalized so that the sum of all weights is 1. Then, a weighted sum is applied to each candidate initial value. The logic behind this setting is that benchmark frequencies with higher confidence levels contribute more to the final initial state, while auxiliary benchmark frequencies with lower confidence levels but still usable only provide supplementary references.

[0097] S508, Obtain the initial state confidence level Q; The initial state confidence level Q is jointly determined by the baseline confidence level, the confidence level of the correlation between the same star and the different frequencies before the anomaly, the stability of the baseline frequency point after the anomaly, and the number of participating baseline frequency points.

[0098] In this embodiment, the baseline confidence level accounts for 0.40, the pre-anomaly correlation confidence level accounts for 0.30, the post-anomaly stability level accounts for 0.20, and the consistency of the number of baseline frequency points accounts for 0.10. If Q is not less than 0.80, the initial state is marked as high confidence; if Q is greater than or equal to 0.60 and less than 0.80, it is marked as medium confidence; if Q is less than 0.60, it is marked as low confidence.

[0099] High-confidence initial states serve as strong constraint bases; medium-confidence initial states serve as general constraint bases; and low-confidence initial states serve only as auxiliary bases and do not independently support disturbance fractals.

[0100] S509. Degradation and reconstruction are performed when the reference frequency is insufficient. If S4 outputs a reference frequency point insufficient flag, it reads the target frequency point's own trend from 10 to 30 seconds before the anomaly and briefly extrapolates this trend to the window after the anomaly. The extrapolation window is preferably no more than 10 seconds. The confidence level of this initial state is fixed at no higher than 0.50.

[0101] The 10-second extrapolation upper limit is set because the trend of the target frequency can only be approximately maintained for a short period of time. After 10 seconds, carrier motion, ionospheric changes, and changes in the receiver's internal state can all amplify the error. Low-confidence markers prevent downgraded reconstruction results from being misused as strong evidence.

[0102] S510, Output the initial observation status table; The initial observation state table includes event number, target frequency, reference frequency, timestamp, initial carrier-to-noise ratio, initial pseudorange change rate, initial Doppler, initial carrier phase continuity, initial observation quality state, mapping relationship, fusion weight, and initial state confidence Q.

[0103] S6. Based on the initial observation state, the actual observation state of the target disturbed frequency point, and the observation relationship between the same star and different frequencies, construct an observation constraint set; S601. Obtain the initial observation residuals; The actual observation status of the target frequency point and the initial observation status output by S5 are read epoch by epoch, and the residuals are calculated.

[0104] The carrier-to-noise ratio (CNR) residual is the initial CNR minus the actual CNR. If the initial CNR is 39.0 dB-Hz and the actual CNR is 33.5 dB-Hz, then the CNR residual is 5.5 dB-Hz, indicating that the actual signal quality is lower than the expected level without interference.

[0105] The residual of the pseudorange change rate is the absolute value of the difference between the actual pseudorange change rate and the initial pseudorange change rate. If the initial value is -319.792 m / s and the actual value is -318.900 m / s, then the residual is 0.892 m / s.

[0106] The Doppler residual is the absolute value of the difference between the actual Doppler and the initial Doppler. If the initial value is -1680.3Hz and the actual value is -1678.9Hz, then the residual is 1.4Hz.

[0107] Carrier phase continuity residuals are represented by state conflicts. A continuity conflict occurs when the initial state is continuous but the actual state is a cycle slip, loss of lock, or phase interruption.

[0108] S602, Generate initial observation constraints; In this embodiment, the allowable deviation range of the initial observation constraint is as follows: The permissible decrease in carrier-to-noise ratio is 3.5 dB-Hz; The permissible deviation range for the pseudorange rate of change is 0.5 m / s; The permissible deviation range for Doppler is 0.8 Hz; Carrier phase continuity requires that there be no cycle slip or loss of lock within three consecutive epochs; The observation quality status must not change from normal to abnormal within three consecutive epochs.

[0109] The above values ​​are set based on the following: 3.5dB-Hz is higher than the short-term carrier-to-noise ratio fluctuation of 1s to 3s under most fixed environments, while being lower than the continuous decrease caused by obvious suppression interference; the 0.5m / s pseudorange change rate deviation can cover the normal dynamic change error of low-speed and medium-speed carriers; the 0.8Hz Doppler deviation is higher than the common range of Doppler noise before the anomaly, which can reduce misjudgment caused by normal frequency estimation jitter; and three consecutive epochs are used to exclude single-epoch observation refresh anomalies.

[0110] If the actual residual exceeds the allowable deviation range and remains so for more than 3 consecutive epochs, the corresponding observation constraint is marked as a conflict. If only 1 or 2 epochs exceed the threshold, it is marked as an instantaneous deviation and is not considered a stable conflict.

[0111] S603, Generate constraints for observations of the same star at different frequencies; Read the normal difference range between the target frequency and the reference frequency before the anomaly, and then read the current difference between the two after the anomaly. If the current difference exceeds the normal difference range before the anomaly and lasts for more than 3 epochs, a co-satellite frequency constraint conflict is generated.

[0112] In this embodiment, the permissible range for different frequencies within the same satellite is preferably determined by the mean difference before the anomaly ± 3 times the standard deviation of the difference before the anomaly, and a physical lower limit is set. The permissible lower limit for the pseudorange change rate within the same satellite at different frequencies is 0.5 m / s, the permissible lower limit for Doppler within the same satellite at different frequencies is 0.8 Hz, and the permissible lower limit for the carrier-to-noise ratio within the same satellite at different frequencies is 3.5 dB-Hz.

[0113] For example, before the anomaly, the mean Doppler difference between B1C and B2a was 0.2 Hz, and the standard deviation was 0.15 Hz. Therefore, a range three times the standard deviation is 0.45 Hz. Since the lower limit of Doppler tolerance is 0.8 Hz, 0.8 Hz is used as the final tolerance range. If the difference reaches 1.4 Hz after the anomaly and persists for three epochs, then there is a conflict with the Doppler co-star frequency constraint.

[0114] The logic behind using a combination of three standard deviations and physical lower limits is that the normal differences vary between different receivers and different frequency combinations. Using only a fixed threshold is not applicable to specific equipment; using only a statistical threshold would result in an overly strict threshold due to the small fluctuations before anomalies. Combining the two can improve feasibility and stability.

[0115] S604. Generate cross-constraint consistency conditions; Generate cross-constraint consistency conditions based on the structural features required for different types of disturbance.

[0116] The structural conditions corresponding to frequency-selective interference are: minimal collisions are concentrated at a single target frequency, no propagation contamination occurs at the reference frequency, and the target frequency deviates from the reference frequency or splits at different frequencies on the same satellite.

[0117] The structural conditions corresponding to frequency sweep interference are: at least two frequency points have constraint conflicts, and the abnormal first contact times of different frequency points are progressive, with the first arrival time difference between adjacent disturbed frequency points preferably being 0.5s to 10s.

[0118] The structural conditions corresponding to broadband suppression interference are: signal quality degradation or tracking anomalies occur at least on at least two frequency points and at least three satellite channels, and there is a common upstream abnormal node in the RF front-end layer or automatic gain control layer. The logic of setting three satellite channels as the preferred lower limit is that: anomalies of one or two satellites may be caused by obstruction, multipath propagation, or low elevation angle, while anomalies of three or more satellites on the same frequency or multiple frequencies more accurately reflect the impact of external suppression.

[0119] The structural conditions corresponding to deception interference are: the RF front-end power and automatic gain control are not significantly abnormal, the carrier-to-noise ratio may not drop significantly, but there are structural conflicts in pseudorange change rate, Doppler, carrier phase continuity or dynamic consistency of different frequencies on the same satellite.

[0120] S605, Form a set of observation constraints; Each constraint record includes constraint number, constraint type, target frequency, reference frequency, observation name, actual value, reference value, allowable deviation range, constraint deviation amount, deviation direction, duration epochs, constraint confidence, constraint weight, and conflict status.

[0121] The constraint confidence level is jointly determined by the initial state confidence level Q, the baseline confidence level C, the observation data completeness rate, and the duration of the conflict. If Q is not less than 0.80, C is not less than 0.80, and the conflict lasts for more than 5 epochs, the constraint confidence level is set to high; if Q or C is between 0.60 and 0.80, the constraint confidence level is set to medium; if Q is less than 0.60, the constraint confidence level is set to low.

[0122] S6 outputs the set of observation constraints, which serves as the input for S7 to extract the standard conflict set of observations.

[0123] S7. Determine the standard conflict observation set based on the constraint deviation of each constraint condition in the observation constraint set; S701, Obtain the constraint deviation; Calculate the constraint deviation for each constraint. The constraint deviation represents the degree to which the actual deviation exceeds the allowable range.

[0124] For a carrier-to-noise ratio constraint, if the residual is 5.5 dB-Hz and the allowable range is 3.5 dB-Hz, then the excess is 2.0 dB-Hz. Dividing 2.0 dB-Hz by 3.5 dB-Hz yields a normalized deviation of 0.57. This value indicates that the deviation is approximately 57% of the allowable range.

[0125] For pseudorange rate of change constraints, if the residual is 0.892 m / s and the allowable range is 0.5 m / s, then the excess is 0.392 m / s, and the normalized deviation is 0.784.

[0126] For Doppler constraints, if the residual is 1.4Hz and the allowable range is 0.8Hz, then the excess is 0.6Hz and the normalized deviation is 0.75.

[0127] The above calculations are used to find the difference between the actual value and the benchmark value, and then compare it with the allowable range. The part exceeding the allowable range is the constraint deviation.

[0128] S702, Marking conflict constraints; When the constraint deviation is greater than 0 and the duration is not less than 3 epochs, the constraint is marked as a conflict constraint. If the normalized deviation is not less than 0.50 and the duration is not less than 5 epochs, it is marked as a strong conflict; if the normalized deviation is not less than 0.20 and less than 0.50, or the duration is 3 to 4 epochs, it is marked as a medium conflict; if it is only a slight over-limit or lasts for less than 3 epochs, it is marked as a weak conflict or momentary deviation.

[0129] Strong conflicts are given priority in entering the candidate conflict object set; medium conflicts are entered when they are related to other strong conflicts in terms of time, frequency, or processing level; weak conflicts are not entered into the standard conflict observation set candidate set on their own.

[0130] S703, Construct a set of candidate conflict objects; The observations and constraints involved in the conflict constraints are written into the candidate conflict object set. The candidate conflict objects include carrier-to-noise ratio, AGC value, correlation peak ratio, pseudorange change rate, Doppler, carrier phase continuity, observation quality indicators, co-satellite frequency difference components, positioning residuals, and satellite weights.

[0131] For each candidate conflict object, record the object number, observation name, processing level, frequency, satellite channel, associated constraint number, anomaly first contact time, constraint deviation, constraint confidence level, and duration epoch.

[0132] S704. Determine the order of deletion tests; The elimination tests are performed in order from redundancy to core. In this embodiment, positioning residuals, satellite weights, and satellite removal status are tested first, as these data are the results of previous observation anomalies. Observation quality indicators and status indicators are tested, followed by co-satellite frequency combination quantities, and finally, fundamental observations such as carrier-to-noise ratio, pseudorange change rate, Doppler, and carrier phase continuity are tested.

[0133] S705, Perform a single object deletion test; Select any observation object to be measured from the set of candidate conflict objects. The observation object corresponds to at least one observation and its associated constraints, and includes processing level identifier, frequency point identifier and satellite channel identifier information. Without changing the structure of other observation objects and constraints, remove the observation object from the set to form a pruned observation subset.

[0134] After the deletion, the constraint deviation of each constraint is recalculated based on the remaining observations, and it is determined whether there are still conflicting constraints that exceed the preset allowable range and continue for a preset number of epochs, and whether the structural relationship on which the current interference classification depends can be maintained, including the same-star inter-frequency consistency relationship and the cross-layer propagation relationship.

[0135] If, after deleting the object, all key conflict constraints can still be reconstructed stably and the interfering fractal structure remains unchanged, then the object is determined to be a redundant object and removed from the candidate conflict object set; if the deletion causes any key constraint conflict to disappear or causes the propagation chain to break and become invalid, then the object is determined to be a necessary object and retained.

[0136] S706, Perform constraint-level deletion tests; After completing the reduction at the observation object level, further reduction and verification were performed at the constraint condition level. Specifically, a single constraint is selected from the set of observation constraints as the object to be tested, including initial observation constraints, same-star different-frequency observation constraints, or cross-frequency propagation constraints. First, the constraint is removed, and then the constraint satisfaction judgment system is reconstructed based on the remaining constraints.

[0137] During the recalculation process, the constraint deviations and conflict states of each observation are reassessed, and it is determined whether the current disturbance classification structure still holds, i.e., whether it still satisfies at least one of the following structural constraints: The target frequency still exhibits a stable deviation relative to the initial observation state; There is still a relationship of consistency disruption or split between different frequency points of the same star; The abnormal propagation path can still be closed in the frontier graph.

[0138] If the above structure still holds after deleting the constraint, then the constraint is considered a non-essential constraint and is removed from the set of observed constraints; if the deletion results in the failure of the interference fractal or the breakage of the key propagation structure, then the constraint is considered a core constraint and is retained.

[0139] S707. Determine the standard conflict observation set; After completing the observation object reduction test and constraint reduction test, the retained observation objects and constraints are analyzed to determine the standard conflict observation set; Specifically, a correspondence table is established between the remaining observation objects and the remaining constraints. Each row in this table records an observation object and the constraints it participates in, including the observation name, actual observation value, reference observation value, constraint deviation, duration epoch, processing level identifier, frequency identifier, satellite channel identifier, the first trigger time of the anomaly, and whether the observation participates in the initial observation constraint or the same-satellite, different-frequency observation constraint. If an observation object no longer participates in any conflict constraints, it means that the observation cannot support the interference classification, and it is removed from the remaining set. If a constraint lacks an actual observation value, a reference observation value, or a corresponding observation object, then the constraint is no longer retained as a valid constraint.

[0140] The remaining observations are mapped to the anomaly first-arrival front graph, and their corresponding graph nodes are checked to see if they can form an interpretable propagation sequence according to the anomaly's initial trigger time. For observations within the same frequency, the anomaly propagation relationship is shown from front to back in terms of processing hierarchy. For observations between different frequencies, synchronous anomalies, progressive anomalies, or same-satellite frequency splitting relationships are shown. For observations between different satellite channels, single-channel concentrated anomalies or multi-channel common anomaly relationships are shown. If an observation has numerical deviations but cannot be mapped to an anomaly propagation path, nor can it participate in same-satellite frequency constraints or initial constraints, then this observation will not be included in the standard conflicting observation set.

[0141] For each observation object and constraint in the remaining set, a non-deletability verification is performed again. Specifically, after temporarily deleting any observation object or constraint, it is re-evaluated whether the remaining content still satisfies the core evidence structure required for the current interference classification. If, after deletion, it can still prove that the target disturbance frequency point has a stable deviation relative to the initial observation state, still prove that there is an inconsistency between the target disturbance frequency point and the reference frequency point, and still maintain the corresponding propagation relationship in the anomaly first arrival front diagram, then the deleted content is not necessary and should be removed from the set. If, after deletion, the initial deviation cannot be calculated, the splitting of the same star at different frequencies cannot be proven, the propagation path is broken, the number of conflict constraints is lower than the lower limit required for classification, or the interference type judgment changes, then the observation object or constraint is determined to be necessary and retained.

[0142] S708. Perform structural verification; Verify the time structure, frequency structure, processing hierarchy structure, and satellite channel structure of the standard conflict observation set.

[0143] The time structure check is used to determine whether the first contact time of an anomaly within the standard conflict observation set conforms to the propagation direction of the first contact node diagram. The frequency point structure check is used to determine whether the standard conflict observation set is single-frequency concentrated, multi-frequency progressive, or multi-frequency synchronous. The processing hierarchy structure check is used to determine whether the standard conflict observation set is located at a processing hierarchy that matches the classification. The satellite channel structure check is used to determine whether the anomaly is concentrated at a single satellite frequency point or extends to multiple satellite channels.

[0144] If the structure check fails, return to S703 to rebuild the candidate conflict object set, or mark the classification result as a low-confidence anomaly.

[0145] S709, Output standard conflict observation set record; The standard conflict observation set record includes the core number, observation name, constraint number, frequency, satellite channel, processing level, anomaly first contact time, constraint deviation, confidence level, reason for retention, and supporting classification type; for example, the initial conflict of pseudorange change rate disappears after deletion, the same-star different-frequency Doppler split cannot be proven after deletion, and the broadband common upstream structure does not hold after deletion.

[0146] S8. Based on the composition structure of the standard conflict observation set, determine the interference classification results of the BeiDou multi-mode GNSS receiver; S801. Identify non-transmittable observational anomalies; If the standard conflicting observation set contains fewer than three core objects and cannot be mapped to an effective propagation path in the first-touch node graph, or only manifests as a short-term anomaly near a single satellite channel and a single epoch, then a non-propagational observation anomaly is output. This result corresponds to occlusion, multipath, receiver transient jitter, or occasional single-channel loss of lock.

[0147] S802, Identify frequency-selective interference; If the standard conflict observation set is concentrated at the same target frequency, the confidence level of the reference frequency is not less than 0.70, the target frequency has at least two types of constraint conflicts relative to the reference frequency, namely carrier-to-noise ratio, pseudorange change rate, Doppler or carrier phase continuity, and other co-satellite frequencies do not form a propagation contamination structure, then the output frequency selective interference occurs.

[0148] In this embodiment, the confidence level of frequency-selective interference is determined based on the number of conflicting constraints at the target frequency, the baseline confidence level, and the degree of concentration of a single frequency. If there are more than three strong conflicting constraints at the target frequency and the baseline confidence level is not less than 0.80, then the confidence level of frequency-selective interference is determined to be not less than 0.80.

[0149] S803, identify frequency sweeping interference; If the standard conflict observation set covers at least two frequency points, and the first contact times of the anomalies at different frequency points are progressive, the first arrival time difference between adjacent frequency points is between 0.5s and 10s, and there are cross-frequency propagation edges in the first contact node diagram, then frequency sweep interference is output.

[0150] Setting a lower limit of 0.5s is to distinguish between frequency sweep progression and near-synchronous broadband suppression; setting an upper limit of 10s is to exclude frequency anomalies caused by different abnormal events over long time intervals.

[0151] S804, Identify broadband suppression interference; If the standard conflict observation set covers at least two frequency points and at least three satellite channels, and there is at least one common upstream anomaly among the following: RF front-end power rise of not less than 5dB, AGC change of not less than 3dB, or noise floor sustained increase of not less than 4dB, then wideband suppression interference will be output.

[0152] Broadband suppression interference is determined based on the fact that multiple frequency points and channels are simultaneously or nearly simultaneously affected, and that the interference can be traced back to the front end or a common upstream node of the AGC. If there are only multiple abnormal observations but no common upstream node, the confidence level of broadband suppression is reduced.

[0153] S805, Identifying deceptive interference; If the RF front-end power and AGC do not reach the broadband suppression anomaly threshold, but there are at least two of the following in the standard conflict observation set: pseudorange change rate initial constraint conflict, Doppler co-star inter-frequency constraint conflict, carrier phase continuity conflict, or observation quality state conflict, and these conflicts are mainly located in the observation generation layer and the positioning solution layer, then deception interference will be output.

[0154] The determination of deception interference lies in the unreasonable dynamic relationship of the observation. If the actual pseudorange change rate and Doppler of the target frequency point continuously deviate from the initial state, while the reference frequency point remains stable, it indicates that the observation results of the target frequency point are inconsistent with the motion relationship that the same satellite should have.

[0155] Please see Figure 2 The present invention also provides an interference pattern identification system for BeiDou multi-mode GNSS, the system being used to implement the aforementioned interference pattern identification method for BeiDou multi-mode GNSS, comprising: The data acquisition module is used to acquire the received data of the Beidou multi-mode GNSS receiver before and after the abnormal triggering time. The received data includes the receiving status data corresponding to different processing levels, different frequency points and different satellite channels. The graph construction module is used to construct a first-touch node graph based on the abnormal first-touch times of each processing level, frequency point, and satellite channel in the received status data; The base point determination module is used to determine the target disturbed frequency point based on the first contact node diagram, and to perform abnormal propagation correlation analysis on other frequency points of the same Beidou satellite to which the target disturbed frequency point belongs, eliminate frequency points that have abnormal propagation correlation with the target disturbed frequency point, and determine the reference frequency point; The reconstruction module reconstructs the initial observation state of the target disturbed frequency point under interference-free conditions based on the observation state of the reference frequency point. The constraint module is used to construct a set of observation constraints based on the initial observation state, the actual observation state of the target disturbed frequency point, and the observation relationship between the same satellite and different frequencies; The deviation analysis module determines the standard conflict observation set based on the constraint deviation of each constraint in the observation constraint set; The interference classification module is used to determine the interference classification results of the BeiDou multi-mode GNSS receiver based on the composition structure of the standard conflict observation set.

[0156] The above embodiments are implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments are implemented, in whole or in part, in the form of a computer program product.

[0157] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art will use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this application.

[0158] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0159] In conclusion, the above description is only 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 method for interference pattern identification in BeiDou multi-mode GNSS, characterized in that, Includes the following steps: Acquire the received data of the Beidou multi-mode GNSS receiver before and after the abnormal triggering time. The received data includes the received status data corresponding to different processing levels, different frequency points and different satellite channels. Based on the abnormal first contact times of each processing level, frequency point, and satellite channel in the received status data, a first contact node diagram is constructed; Based on the first contact node diagram, the target interference frequency point is determined, and abnormal propagation correlation analysis is performed on other frequency points of the same Beidou satellite to which the target interference frequency point belongs. Frequency points with abnormal propagation correlation with the target interference frequency point are eliminated, and the reference frequency point is determined. Based on the observation status of the reference frequency, reconstruct the initial observation status of the target's disturbed frequency under interference-free conditions; Based on the initial observation state, the actual observation state of the target disturbed frequency point, and the observation relationship between the same star and different frequencies, an observation constraint set is constructed; Based on the constraint deviation of each constraint condition in the observation constraint set, determine the standard conflict observation set; Based on the composition of the standard conflict observation set, the interference classification results of the BeiDou multi-mode GNSS receiver are determined.

2. The interference pattern identification method for BeiDou multi-mode GNSS according to claim 1, characterized in that, The specific process of constructing the first-touch node graph includes: According to the signal processing chain of the Beidou multi-mode GNSS receiver, the received status data is divided into data groups of multiple processing levels, and the graph nodes are determined based on the combination of processing level identifier, frequency point identifier, and satellite channel identifier. Establish a corresponding state time series for each graph node, and determine the first touch time of an anomaly for each graph node based on the first deviation time of the received state data from the anomaly judgment condition in the state time series. Based on the order of the first contact time of the anomaly at each graph node, the propagation direction of the anomaly from the first triggered graph node to the second triggered graph node is determined; A first-touch node graph is constructed based on the graph nodes, the first-touch time of the anomaly, and the propagation direction.

3. The interference pattern identification method for BeiDou multi-mode GNSS according to claim 2, characterized in that, The target interference frequency point is determined based on the first contact node diagram, specifically including: Based on the abnormal first contact time and propagation direction of each frequency point corresponding to the node in the first contact node diagram, candidate interference frequency points on the abnormal propagation path are determined. Based on the degree of deviation of the receiving state of the candidate interference frequency points after the abnormal trigger time from the reference state before the abnormal trigger time, the target interference frequency point is determined from the candidate interference frequency points.

4. The interference pattern identification method for BeiDou multi-mode GNSS according to claim 3, characterized in that, The logic for determining the reference frequency is as follows: Obtain other observable frequency points of the same Beidou satellite to which the target disturbed frequency point belongs, construct a set of candidate frequency points of different frequencies of the same satellite, and map the target disturbed frequency point and each candidate frequency point in the set of candidate frequency points of different frequencies of the same satellite to the first contact node graph respectively to obtain the target graph node set and the candidate graph node set; In the first-touch node graph, a directed path search is performed on the target graph node set and each candidate graph node set to obtain the path connectivity relationship, and the abnormal first-touch time sequence, processing level change sequence and frequency point change sequence are extracted. First-arrival timing consistency analysis is performed on the abnormal first-touch time sequence, cross-layer propagation analysis is performed on the processing level change sequence, and cross-frequency propagation analysis is performed on the frequency point change sequence to obtain the propagation association risk value of each candidate frequency point relative to the target disturbed frequency point; If the propagation association risk value is greater than the preset risk threshold, the corresponding candidate frequency point will be removed from the set of candidate frequency points of the same star but different frequencies, and the observation stability analysis will be performed on the remaining candidate frequency points to obtain the observation stability evaluation value. Based on the propagation association risk value and the observation stability evaluation value, the baseline confidence level of each remaining candidate frequency point is calculated, and the remaining candidate frequency points whose baseline confidence level meets the preset confidence conditions are determined as the baseline frequency points.

5. The interference pattern identification method for BeiDou multi-mode GNSS according to claim 4, characterized in that, Reconstructing the initial observation state of the target's disturbed frequency points under interference-free conditions, specifically including: Based on the change in the observation state of the reference frequency point before and after the abnormal triggering time, and the observation correlation between different frequency points of the same Beidou satellite, the change in the observation state of the target disturbed frequency point under the interference-free condition is determined; based on the actual observation state of the target disturbed frequency point before the abnormal triggering time and the change in the observation state under the interference-free condition, the initial observation state of the target disturbed frequency point is generated.

6. The interference pattern identification method for BeiDou multi-mode GNSS according to claim 5, characterized in that, The process of constructing the set of observation constraints is as follows: Calculate the deviation between the initial observation state and the actual observation state of the target disturbed frequency point, and generate the initial observation residual; Generate initial observation constraints based on the initial observation residuals; Based on the same-satellite, different-frequency observation relationship between the target disturbed frequency point and the reference frequency point, generate same-satellite, different-frequency observation constraint conditions. The initial observation constraints and the same-star, different-frequency observation constraints are added to the observation constraint set.

7. The interference pattern identification method for BeiDou multi-mode GNSS according to claim 6, characterized in that, Perform constraint satisfaction judgment on the set of observation constraints to determine the standard conflict set of observations, specifically including: For each constraint in the set of observation constraints, calculate the constraint deviation amount and mark the constraint deviation amount that exceeds the corresponding preset allowable range as conflict constraint conditions; Extract the observations involved in the calculation of conflict constraints, the conflict constraints, the processing level identifiers, frequency identifiers, satellite channel identifiers, and the first contact time of anomalies to construct a set of candidate conflict objects; The observations and conflict constraints in the candidate conflict object set are taken as the conflict objects to be tested, and the deletion test is performed on each conflict object to be tested according to the preset deletion order. The deletion test includes: deleting one conflict object to be tested from the candidate conflict object set according to a preset deletion order; recalculating the constraint deviation based on the remaining observations and remaining conflict constraints after deletion; and determining whether there are still conflict constraints whose constraint deviation exceeds the corresponding preset allowable range after deletion. If there are conflict constraints after deleting the conflict object to be tested, the conflict object to be tested is deleted from the candidate conflict object set. If there are no conflict constraints after deleting the conflict object to be tested, the conflict object to be tested is retained in the candidate conflict object set. Repeat the elimination test until any candidate conflict object in the candidate conflict object set is eliminated, and the remaining candidate conflict object set after repeated elimination tests is determined as the standard conflict observation set.

8. The interference pattern identification method for BeiDou multi-mode GNSS according to claim 7, characterized in that, The determination of the interference classification results experienced by the BeiDou multi-mode GNSS receiver specifically includes: Map each observation and constraint in the standard conflict observation set to the first contact node graph. Based on the overlap between the corresponding graph nodes of the standard conflict observation set and the anomaly propagation path, determine whether the standard conflict observation set belongs to the anomaly propagation path. When the standard conflict observation set belongs to an abnormal propagation path, the distribution structure of the standard conflict observation set in the frequency point dimension, satellite channel dimension, and processing level dimension is analyzed to determine the interference classification result. When the standard conflict observation set does not form a correspondence with the abnormal propagation path, the corresponding anomaly is identified as a non-propagation observation anomaly. The interference classification result includes frequency-selective interference, frequency sweeping interference, broadband suppression interference, and deception interference.

9. An interference pattern identification system for BeiDou multi-mode GNSS, based on the interference pattern identification method for BeiDou multi-mode GNSS according to any one of claims 1 to 8, characterized in that, include: The data acquisition module is used to acquire the received data of the Beidou multi-mode GNSS receiver before and after the abnormal triggering time. The received data includes the receiving status data corresponding to different processing levels, different frequency points and different satellite channels. The graph construction module is used to construct a first-touch node graph based on the abnormal first-touch times of each processing level, frequency point, and satellite channel in the received status data; The base point determination module is used to determine the target disturbed frequency point based on the first contact node diagram, and to perform abnormal propagation correlation analysis on other frequency points of the same Beidou satellite to which the target disturbed frequency point belongs, eliminate frequency points that have abnormal propagation correlation with the target disturbed frequency point, and determine the reference frequency point; The reconstruction module reconstructs the initial observation state of the target disturbed frequency point under interference-free conditions based on the observation state of the reference frequency point. The constraint module is used to construct a set of observation constraints based on the initial observation state, the actual observation state of the target disturbed frequency point, and the observation relationship between the same satellite and different frequencies; The deviation analysis module determines the standard conflict observation set based on the constraint deviation of each constraint in the observation constraint set; The interference classification module is used to determine the interference classification results of the BeiDou multi-mode GNSS receiver based on the composition structure of the standard conflict observation set.

Images