Method and apparatus for signal classification of carrier phase and signal-to-noise ratio

Abstract

The application discloses a signal classification and discrimination method and device combining carrier phase and signal-to-noise ratio, relates to the global navigation satellite technology field, and comprises the following steps: S1, acquiring original observation data; S2, constructing a signal-to-noise ratio observation sequence and a carrier phase unmodeled residual sequence; S3, extracting signal strength deviation features and short-time stability features of the carrier phase; S4, constructing a comprehensive discrimination quantity; S5, analyzing an adaptive discrimination threshold; and S6, analyzing a classification and discrimination result of the original observation data. Through the construction of the carrier phase residual time sequence stability features and the signal-to-noise ratio deviation reference model statistical features in a sliding window, the signal strength deviation features and the short-time stability features of the carrier phase are used to generate the comprehensive discrimination quantity, and then the adaptive threshold is analyzed, so that high-sensitivity detection and low-false-alarm-rate discrimination of non-line-of-sight signals in a complex urban environment are realized, and the calculation accuracy and availability of precise point positioning and real-time kinematic positioning are ensured.

Description

Technical Field

[0001] This invention relates to the field of global navigation satellite technology, and in particular to a signal classification and discrimination method and apparatus that combines carrier phase and signal-to-noise ratio. Background Technology

[0002] With the continuous development and popularization of Global Navigation Satellite System (GNSS) technology, location-based services have been widely applied in many fields such as intelligent transportation, autonomous driving, precision agriculture, and disaster monitoring. Users are placing increasingly stringent demands on the accuracy and reliability of satellite navigation and positioning. In typical complex environments such as urban canyons, tree-lined roads, and densely built-up areas, satellite navigation receivers often face severe challenges from signal obstruction and multipath reflection. Due to obstacle obstruction, satellite signals cannot reach the receiver antenna via a direct path, but instead indirectly enter the receiver tracking loop through reflection and diffraction from building surfaces or the ground, forming non-line-of-sight (NLS) propagation signals. The pseudorange measurement error of NLS propagation signals can typically reach several meters to tens of meters, and carrier phase observations are accompanied by significant phase jitter and integer cycle jump risks. If these are indiscriminately used in positioning calculations, they will severely impair the accuracy of position estimation and integrity monitoring performance. Therefore, how to efficiently and accurately identify, eliminate, or downweight NLS propagation signals has become one of the core technical bottlenecks for improving satellite navigation and positioning performance in complex urban environments.

[0003] Currently, the main technical approaches for non-line-of-sight (NLS) signal detection in satellite navigation fall into three categories: methods based on external sensors, methods based on 3D environment models, and methods based on the receiver's observation of the signal's inherent characteristics. Methods based on external sensors typically require additional hardware such as lidar, visual cameras, or inertial measurement units to acquire prior environmental information, then use ray tracing or scene matching to determine the satellite signal's propagation path type. While these methods offer advantages in accuracy, they significantly increase system hardware costs and power consumption, and are highly dependent on the spatiotemporal synchronization accuracy and calibration quality between sensors. Methods based on 3D environment models require pre-constructing high-precision geometric models of buildings surrounding the station, combining satellite orbit information and the receiver's approximate location for signal obstruction analysis. However, acquiring and real-time updating 3D models is costly, and model accuracy directly impacts the reliability of NLS discrimination, facing significant obstacles to widespread adoption in practical large-scale applications.

[0004] Compared to the first two types of methods, the method based on the characteristics of the receiver-observed signal itself has better universality and ease of deployment because it does not require additional hardware and does not rely on prior environmental information. In recent years, it has received widespread attention from academia and industry.

[0005] Existing technical solutions still have two common technical defects in the classification and discrimination of non-line-of-sight and line-of-sight signals in satellite navigation: First, at the feature extraction level, the temporal stability features of the unmodeled carrier phase residuals and the deviation features of the signal-to-noise ratio relative to the elevation angle reference model are not deeply integrated and jointly analyzed, resulting in insufficient sensitivity and high false alarm rate in non-line-of-sight signal detection; Second, at the discrimination decision level, most of them use fixed thresholds or classifiers based on pure data, lacking an adaptive discrimination mechanism that quantitatively models the coupling relationship between carrier phase stability features and signal-to-noise ratio deviation features, resulting in poor robustness of classification and discrimination under environmental changes and signal quality fluctuations. Summary of the Invention

[0006] The purpose of this invention is to design a signal classification and discrimination method and device that combines carrier phase and signal-to-noise ratio to solve the above-mentioned problems.

[0007] The present invention achieves the above objectives through the following technical solutions:

[0008] Signal classification and discrimination methods that combine carrier phase and signal-to-noise ratio include:

[0009] S1. Obtain raw observation data of the target satellite in continuous epochs. The raw observation data includes carrier phase observations output by the target satellite navigation receiver and broadcast ephemeris or precise ephemeris data.

[0010] S2. Construct a signal-to-noise ratio observation sequence and a carrier phase unmodeled residual sequence based on the raw observation data of consecutive epochs;

[0011] S3, based on the signal-to-noise ratio observation sequence respectively and carrier phase unmodeled residual sequence Extract signal strength deviation characteristics and short-time stability characteristics of carrier phase;

[0012] Based on the signal-to-noise ratio observation sequence The specific steps for extracting signal intensity deviation features are as follows:

[0013] (1) Observation sequence based on signal-to-noise ratio Analysis of signal-to-noise ratio bias sequence ;

[0014] (2) Based on the signal-to-noise ratio deviation sequence Analyze signal attenuation characteristics within a sliding time window of length N. and signal fluctuation characteristics As a signal strength deviation feature, it is represented as: , ,in, This represents the i-th signal-to-noise ratio (SNR) deviation value within the window, and N represents the total number of SNR deviation value samples included within the selected sliding time window.

[0015] Based on the carrier phase unmodeled residual sequence The specific steps for extracting the short-time stability features of the carrier phase are as follows:

[0016] 1) Unmodeled residual sequence of carrier phase within a sliding time window of length N Perform first-order time difference processing to obtain the difference sequence. , represented as: ;

[0017] 2) Based on difference sequences Calculate the short-time stability characteristics within the sliding time window. , represented as: Where M represents the number of difference values ​​within the window, and M = N - 1. This represents the j-th difference value;

[0018] S4. Construct a comprehensive discrimination function based on signal strength deviation characteristics and short-time stability characteristics of carrier phase;

[0019] S5. Obtain historical observation data and adaptively determine thresholds based on the analysis of historical observation data;

[0020] S6. Analyze the classification and discrimination results of the original observation data based on the comprehensive discrimination value and the adaptive discrimination threshold. The classification and discrimination results are either non-line-of-sight propagation discrimination results or line-of-sight propagation discrimination results. When the classification and discrimination results are non-line-of-sight propagation discrimination results, output the non-line-of-sight propagation label; when the classification and discrimination results are line-of-sight propagation discrimination results, output the line-of-sight propagation label.

[0021] A signal classification and discrimination device that combines carrier phase and signal-to-noise ratio includes:

[0022] Storage; storage is used to store computer programs;

[0023] An actuator; the actuator is used to execute a computer program stored in a memory, which, when executed, implements the signal classification and discrimination method based on the combined carrier phase and signal-to-noise ratio as described above.

[0024] The beneficial effects of this invention are as follows: By constructing the carrier phase residual temporal stability characteristics and signal-to-noise ratio deviation reference model statistical characteristics within a sliding window, and generating a comprehensive discrimination quantity based on the signal strength deviation characteristics and the short-term stability characteristics of the carrier phase, and then adaptively determining the discrimination threshold according to the maximum inter-class variance criterion, the phase jitter information contained in the carrier phase observation value and the signal attenuation information contained in the signal-to-noise ratio observation value are deeply combined and coupled in the time dimension to effectively solve the technical problems of insufficient utilization of carrier phase anomaly information and the difficulty in simultaneously achieving non-line-of-sight detection sensitivity and false alarm rate due to the single dependence on signal-to-noise ratio in the prior art. Finally, it achieves high-sensitivity detection and low false alarm rate discrimination of non-line-of-sight signals in complex urban environments, ensuring the solution accuracy and availability of precise single-point positioning and real-time dynamic positioning. Attached Figure Description

[0025] Figure 1 This is a schematic diagram of the signal classification and discrimination method combining carrier phase and signal-to-noise ratio of the present invention; Detailed Implementation

[0026] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0027] Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

[0028] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0029] In the description of this invention, it should be understood that the terms "upper," "lower," "inner," "outer," "left," "right," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, or the orientation or positional relationship commonly used when the product of this invention is in use, or the orientation or positional relationship commonly understood by those skilled in the art. They are only used to facilitate the description of this invention and to simplify the description, and are not intended to indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0030] Furthermore, the terms "first," "second," etc., are used only to distinguish descriptions and should not be interpreted as indicating or implying relative importance.

[0031] In the description of this invention, it should also be noted that, unless otherwise explicitly specified and limited, terms such as "set" and "connection" should be interpreted broadly. For example, "connection" can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium; it can be a connection within two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0032] The present invention will be further described below with reference to the accompanying drawings:

[0033] like Figure 1 As shown, the signal classification and discrimination method combining carrier phase and signal-to-noise ratio includes:

[0034] S1. Obtain raw observation data of the target satellite in continuous epochs. The raw observation data includes carrier phase observations output by the target satellite navigation receiver and broadcast ephemeris or precise ephemeris data.

[0035] S2. Construct a signal-to-noise ratio observation sequence and a carrier phase unmodeled residual sequence based on the raw observation data of consecutive epochs; the construction of the carrier phase unmodeled residual sequence is specifically as follows:

[0036] ① Calculate satellite clock corrections based on broadcast ephemeris or precise ephemeris data, and calculate geometric distances and tropospheric delays using a precise point positioning model based on raw observation data;

[0037] ② Construct the carrier phase observation equation, expressed as: ,in Let be the carrier phase observation value at epoch t. Let be the geometric distance between the satellite and the receiver, and c be the speed of light. and These are receiver clock bias and satellite clock bias, respectively. For tropospheric delay, Let λ be the ionospheric delay, λ be the carrier wavelength, and N0 be the integer ambiguity. This is an unmodeled error term;

[0038] ③ The extended Kalman filter method is used to measure the geometric distance. Receiver clock bias Satellite clock bias and tropospheric delay Estimate the geometric distance separately to obtain the estimated distance. Receiver clock bias Satellite clock bias and tropospheric delay ;

[0039] ④ Geometric distance after elimination of estimation Receiver clock bias Satellite clock bias and tropospheric delay Obtain the carrier phase unmodeled residual sequence For, it is represented as: .

[0040] S3, based on the signal-to-noise ratio observation sequence respectively and carrier phase unmodeled residual sequence Extract signal strength deviation characteristics and short-time stability characteristics of carrier phase;

[0041] Based on the signal-to-noise ratio observation sequence The specific steps for extracting signal intensity deviation features are as follows:

[0042] (1) Observation sequence based on signal-to-noise ratio Analysis of signal-to-noise ratio bias sequence Specifically, this refers to: observational data collected under historical unobstructed conditions and corresponding satellite elevation angle sequences. Building a reference model , represented as: Further analysis of the signal-to-noise ratio observation sequence Compared with the reference model The deviation between them is used as the signal-to-noise ratio deviation sequence. , represented as: Where a1, b1, and c1 are the model parameters of the reference model, The reference model established for historical unobstructed observation data corresponds to the satellite elevation angle θ at epoch t. t The expected signal-to-noise ratio value is the theoretical signal-to-noise ratio reference for the line-of-sight propagation signal at that elevation angle; the historical unobstructed environment refers to the environment in which this location was not covered by any obstruction in the past.

[0043] (2) Based on the signal-to-noise ratio deviation sequence Analyzing signal attenuation characteristics within a sliding time window and signal fluctuation characteristics As a signal strength deviation feature, it is represented as: , ,in, This represents the i-th signal-to-noise ratio (SNR) deviation value within the window, where N is the total number of SNR deviation value samples included within the selected sliding time window.

[0044] Based on the carrier phase unmodeled residual sequence The specific steps for extracting the short-time stability features of the carrier phase are as follows:

[0045] 1) Unmodeled residual sequence of carrier phase within the sliding time window Perform first-order time difference processing to obtain the difference sequence. , represented as: ;

[0046] 2) Based on difference sequences Calculate the short-time stability characteristics within the sliding time window. , represented as: Where M represents the number of difference values ​​within the window, M = N-1. This represents the j-th difference value.

[0047] S4. Construct a comprehensive discriminant D based on signal strength deviation characteristics and short-time stability characteristics of carrier phase, expressed as: ,in, , and These are short-time stability characteristics. Signal attenuation characteristics and signal fluctuation characteristics The weighting coefficients are expressed as: , , , It is a stability constant that is greater than zero.

[0048] The design of the comprehensive discriminant is based on the inverse variance weighting principle, which uses the reciprocal of the variance of each feature as its weight. A larger variance indicates that the feature is more susceptible to noise and has lower reliability, thus reducing its contribution to the discrimination result. Conversely, a smaller variance indicates that the feature is more stable and reliable, and its weight is increased accordingly. Short-term stability features. Reflects short-time phase jitter and signal attenuation characteristics Reflects the overall signal attenuation and signal fluctuation characteristics. Reflecting the short-term fluctuation amplitude of the signal, these three types of features characterize different aspects of non-line-of-sight (NLS) propagation from two dimensions: phase stability and signal strength. They are complementary and, when combined, can more comprehensively identify NLS signals than a single feature. The linear weighting form is simple and computationally inexpensive, suitable for the real-time processing requirements of navigation receivers. Adding a stability constant to the denominator prevents the weights from approaching infinity when the feature values ​​are close to zero, thus ensuring numerical stability. Therefore, this comprehensive discriminant can dynamically adjust the intensity of each feature under different observation conditions, achieving adaptive fusion and improving discrimination robustness.

[0049] S5. Obtain historical observation data and analyze the adaptive discrimination threshold based on the historical observation data; specifically: analyze the adaptive discrimination threshold T based on the historical observation data using the maximum inter-class variance criterion, expressed as: Where τ is the candidate threshold, and and represent the probabilities that the classification result is divided into distance propagation result and non-distance propagation result, respectively. and These represent the mean values ​​of the classification results, categorized into range propagation and non-range propagation results. Historical observation data includes carrier phase observations and broadcast ephemeris or precise ephemeris data output by the target satellite navigation receiver over a historical period.

[0050] The calculation process of the maximum inter-class variance criterion is as follows: First, using historical observation data, samples clearly labeled as either sight-range or non-sight-range are used to extract the comprehensive discriminant D for each sample. Then, all samples are arranged in ascending order of their comprehensive discriminant D values, and the average of any two adjacent comprehensive discriminant D values ​​is taken as a candidate threshold τ. For each candidate threshold τ, the samples are divided into two groups: those with a comprehensive discriminant D less than or equal to the candidate threshold τ are assigned to the sight-range group, and those with a comprehensive discriminant D greater than the candidate threshold τ are assigned to the non-sight-range group. Next, the sample size proportion and mean of each group are calculated, and the inter-class variance is calculated. The inter-class variance is equal to the product of the proportions of the two groups multiplied by the square of the difference between the means of the two groups. After iterating through all candidate thresholds τ, the candidate threshold τ that maximizes the inter-class variance is selected as the final adaptive discriminant threshold T. This selected threshold ensures that the distribution of sight-range and non-sight-range samples on the comprehensive discriminant D is separated as much as possible, thus achieving better classification results.

[0051] The Argmax function takes the candidate threshold τ that maximizes the entire expression following it. The maximum inter-class variance formula doesn't directly concern itself with the maximum inter-class variance, but rather with the candidate threshold τ that represents the maximum inter-class variance. It iterates through all candidate thresholds τ, calculating the inter-class variance for each τ, and finds the candidate threshold τ with the largest inter-class variance. This candidate threshold τ is the final adaptive discriminant threshold T. Therefore, T equals the entire expression following Argmax, meaning T takes the threshold that best separates the two classes of samples in terms of overall discriminant strength.

[0052] For any candidate threshold τ, first count the total number of samples in the historical samples whose comprehensive discriminant D is less than or equal to the candidate threshold τ, and use this as the line-of-sight sample number. These samples are classified as line-of-sight propagation samples. Dividing the line-of-sight sample number by the total number of historical samples gives the line-of-sight sample number. Then, the number of samples with a comprehensive discriminant D greater than the candidate threshold τ is counted as the number of non-line-of-sight samples. These samples are classified as non-line-of-sight propagation samples. The result is obtained by dividing the number of non-line-of-sight samples by the total number of historical samples. ; and The sum equals 1;

[0053] For each candidate threshold τ, samples with a comprehensive discriminant D less than or equal to τ in the historical samples are classified as line-of-sight propagation, and samples with D greater than τ are classified as non-line-of-sight propagation. It is the arithmetic mean of the comprehensive discriminant D corresponding to all samples classified as line-of-sight propagation; It is the arithmetic mean of the comprehensive discriminant D corresponding to all samples classified as non-line-of-sight propagation; thus, for each change of candidate threshold τ, a pair can be recalculated based on the classification results. and Then, substitute the values ​​into the inter-class variance formula to compare the classification performance under different thresholds.

[0054] S6. Analyze the classification results of the original observation data based on the comprehensive discriminant value and the adaptive discriminant threshold. When the comprehensive discriminant value D is greater than the adaptive discriminant threshold T, the classification result is non-line-of-sight propagation, and a non-line-of-sight propagation label is output. The comprehensive discriminant value is then weighted to obtain a weighted comprehensive discriminant value D', which is output for subsequent positioning calculations. When the comprehensive discriminant value D is less than or equal to the adaptive threshold T, the classification result is line-of-sight propagation, and a line-of-sight propagation label is output. The original comprehensive discriminant value D is directly output for subsequent positioning calculations. The weighted comprehensive discriminant value is expressed as D' = 1 + ηD, where 0 < η < 1, and η is a scaling factor.

[0055] The scaling factor η is a pre-defined small positive number used to control the magnitude of the weighted composite discriminant D' = 1 + ηD. It is typically chosen as a fixed value close to zero, such as 0.01 or 0.05. This ensures that the composite discriminant of non-line-of-sight signals remains greater than 1 without introducing excessively large values ​​due to multiplication by a large factor. The specific value can be determined based on the typical range of the composite discriminant D in historical samples. Generally, it is sufficient to ensure that ηD is much less than 1. If it is desired that the original observation data corresponding to non-line-of-sight propagation has a lower weight in subsequent positioning, η can be increased appropriately; if it is desired to retain the original observation data corresponding to non-line-of-sight propagation, a smaller value should be chosen. Therefore, the scaling factor η is essentially an empirical parameter, and a suitable constant can be selected based on the test results in practical applications.

[0056] A signal classification and discrimination device that combines carrier phase and signal-to-noise ratio includes:

[0057] Storage; storage is used to store computer programs;

[0058] An actuator; the actuator is used to execute a computer program stored in a memory, which, when executed, implements the signal classification and discrimination method based on the combined carrier phase and signal-to-noise ratio as described above.

[0059] A quadratic polynomial reference relationship between signal-to-noise ratio (SNR) and satellite elevation angle is established based on historical unobstructed environmental data or a pre-set empirical model. The standardized deviation sequence of the SNR of the current raw observation data relative to the reference relationship is calculated. This method effectively eliminates the deterministic trend influence of satellite elevation angle changes on SNR and separates the abnormal signal attenuation caused by non-line-of-sight propagation from the normal geometric path loss. As a result, the SNR deviation features extracted subsequently can more purely reflect the degree of interference from environmental obstruction and multipath reflection, significantly improving the separability of non-line-of-sight signal detection features.

[0060] Within a sliding time window, the unmodeled residual sequence of the carrier phase undergoes first-order time difference processing, and the root mean square value of the difference sequence is calculated as a core indicator characterizing the short-term stability of the carrier phase. Simultaneously, within the same time window, the mean and standard deviation of the signal-to-noise ratio deviation sequence are calculated to quantify the degree of signal strength deviation. Furthermore, the temporal feature extraction process is strictly synchronized within the same sliding window to ensure the alignment of features in the time dimension. This approach overcomes the limitation of existing technologies that rely solely on single-epoch static statistics. By capturing the dynamic evolution of carrier phase jitter and signal-to-noise ratio fluctuations within a continuous time window, the classifier can perceive the gradual change in signal features before and after instantaneous non-line-of-sight events, thereby significantly reducing the probability of misjudgment caused by instantaneous observation noise.

[0061] The carrier phase stability index, mean signal-to-noise ratio (SNR) deviation, and standard deviation of SNR deviation are assigned weight coefficients, which are dynamically calculated in real time by introducing the inverse variance of the stability constant term, rather than using fixed empirical weights. This can automatically suppress the weight contribution of the corresponding feature terms under abnormal numerical conditions such as large variance of carrier phase residual or mean SNR deviation close to zero, and avoid the divergence of a single feature value from dominating the entire discrimination result. This improves the robustness and numerical stability of the joint discrimination factor under different observation quality and complex environments.

[0062] By constructing the carrier phase residual temporal stability features and the signal-to-noise ratio deviation from the reference model statistical features within a sliding window, and generating a comprehensive discrimination quantity based on the inverse variance weighting mechanism, a preliminary judgment is made based on an adaptive threshold. Finally, a first-order Markov state transition model is introduced to perform maximum a posteriori probability smoothing correction on the classification results between epochs. The phase jitter information contained in the carrier phase observation and the signal attenuation information contained in the signal-to-noise ratio observation are deeply coupled and modeled in the time dimension. This effectively solves the technical problems of insufficient utilization of carrier phase anomaly information, single dependence on signal-to-noise ratio, and discontinuity of classification results in the time domain in the existing technology. Ultimately, it achieves high-sensitivity detection and low false alarm rate discrimination of non-line-of-sight signals in complex urban environments, ensuring the solution accuracy and availability of precise single-point positioning and real-time dynamic positioning.

[0063] Example:

[0064] The test scenario was a canyon environment in City A, with buildings of approximately 20 to 40 meters in height surrounding the receiver antenna. Using a single GPS satellite at an elevation angle of 35 degrees and a receiver sampling frequency of 1 Hz, carrier phase and signal-to-noise ratio observations were continuously collected for 100 epochs.

[0065] The signal-to-noise ratio reference model is obtained by fitting historical unobstructed environment data as follows: ;in, Let be the satellite elevation angle at epoch t, ​​in degrees. For epoch t, ​​calculate the signal-to-noise ratio (SNR) bias sequence. ;

[0066] Within a sliding time window of 10 epochs, for epochs 51 to 60, the signal-to-noise ratio bias sequence is: The unit is dB.

[0067] The signal attenuation characteristics were calculated as follows: ;

[0068] Signal fluctuation characteristics: ;

[0069] Simultaneously, a carrier phase unmodeled residual sequence is constructed, with the residual sequences corresponding to epochs 51 to 60 being: 0.022, 0.031, 0.018, 0.026, 0.035, 0.028, 0.033, 0.024, 0.030, 0.027; in weeks.

[0070] Perform a first-order difference on the residual sequence to obtain the difference sequence and calculate its short-time stability characteristics. week;

[0071] Take the stability constant Calculate the weighting coefficients:

[0072] ;

[0073] ;

[0074] ;

[0075] Calculate the comprehensive discriminant:

[0076]

[0077] Using samples with known line-of-sight and non-line-of-sight labels from historical observation data, an adaptive discrimination threshold T=0.6 is calculated using the maximum inter-class variance criterion. Since D=0.6406>T, the signal within this sliding window is classified as a non-line-of-sight propagation signal, and a non-line-of-sight propagation label is output. The overall discrimination value is then weighted. Using a scaling factor η=0.01, the weighted overall discrimination value is: The weighted comprehensive discrimination value D' will be input into the subsequent navigation settlement.

[0078] In contrast, for the line-of-sight window from epoch 1 to 10, the mean signal-to-noise ratio (SNR) bias sequence is -0.1 dB, the standard deviation is 0.3 dB, and the root mean square difference of the carrier phase residual is 0.02 cycles. The weighting coefficients are calculated using the same method. , , ;

[0079] The comprehensive discriminant is:

[0080] Since D = 0.3942 < T, the window signal is identified as a line-of-sight propagation signal, and a line-of-sight propagation label is output. The comprehensive discrimination value is directly input into the subsequent navigation solution.

[0081] The above examples demonstrate that the method of the present invention can effectively distinguish between non-line-of-sight propagation signals and line-of-sight propagation signals in urban canyon environments. The comprehensive discrimination value of the non-line-of-sight window is significantly higher than that of the line-of-sight window, and accurate discrimination is achieved through adaptive thresholding.

[0082] The technical solutions of the present invention are not limited to the specific embodiments described above. Any technical modifications made in accordance with the technical solutions of the present invention fall within the protection scope of the present invention.

Claims

1. A signal classification and discrimination method combining carrier phase and signal-to-noise ratio, characterized in that, include: S1. Real-time acquisition of raw observation data of the target satellite in continuous epochs. The raw observation data includes carrier phase observations and broadcast ephemeris or precise ephemeris data output by the target satellite navigation receiver. S2. Construct a signal-to-noise ratio observation sequence and a carrier phase unmodeled residual sequence based on the raw observation data of consecutive epochs; S3, based on the signal-to-noise ratio observation sequence respectively and carrier phase unmodeled residual sequence Extract signal strength deviation characteristics and short-time stability characteristics of carrier phase; Based on the signal-to-noise ratio observation sequence The specific steps for extracting signal intensity deviation features are as follows: (1) Observation sequence based on signal-to-noise ratio Analysis of signal-to-noise ratio bias sequence ; (2) Based on the signal-to-noise ratio deviation sequence Analyzing signal attenuation characteristics within a sliding time window and signal fluctuation characteristics As a signal strength deviation feature, it is represented as: , ,in, Indicates the first in the window There are 1 signal-to-noise ratio (SNR) deviation values, where N represents the total number of SNR deviation value samples included in the selected sliding time window. Based on the carrier phase unmodeled residual sequence The specific steps for extracting the short-time stability features of the carrier phase are as follows: 1) Unmodeled residual sequence of carrier phase within the sliding time window Perform first-order time difference processing to obtain the difference sequence. , is represented as: ; 2) Based on difference sequences Calculate the short-time stability characteristics within the sliding time window. , represented as: Where M represents the number of difference values ​​within the window, This represents the j-th difference value; S4. Construct a comprehensive discrimination function based on signal strength deviation characteristics and short-time stability characteristics of carrier phase; S5. Obtain historical observation data and adaptively determine thresholds based on the analysis of historical observation data; S6. Analyze the classification and discrimination results of the original observation data based on the comprehensive discrimination value and the adaptive discrimination threshold. The classification and discrimination results are either non-line-of-sight propagation discrimination results or line-of-sight propagation discrimination results. When the classification and discrimination results are non-line-of-sight propagation discrimination results, output the non-line-of-sight propagation label; when the classification and discrimination results are line-of-sight propagation discrimination results, output the line-of-sight propagation label.

2. The signal classification and discrimination method based on joint carrier phase and signal-to-noise ratio according to claim 1, characterized in that, In S2, the construction of the carrier phase unmodeled residual sequence is specifically as follows: ① Calculate satellite clock corrections based on broadcast ephemeris or precise ephemeris data, and calculate geometric distances and tropospheric delays using a precise point positioning model based on raw observation data; ② Construct the carrier phase observation equation, expressed as: ,in Let be the carrier phase observation value at epoch t. Let be the geometric distance between the satellite and the receiver, and c be the speed of light. and These are receiver clock bias and satellite clock bias, respectively. For tropospheric delay, Let λ be the ionospheric delay, λ be the carrier wavelength, and N0 be the integer ambiguity. This is an unmodeled error term; ③ Geometric distance Receiver clock bias Satellite clock bias and tropospheric delay Estimate the geometric distance separately to obtain the estimated distance. Receiver clock bias Satellite clock bias and tropospheric delay ; ④ Geometric distance after elimination of estimation Receiver clock bias Satellite clock bias and tropospheric delay Obtain the carrier phase unmodeled residual sequence For, it is represented as: .

3. The signal classification and discrimination method based on joint carrier phase and signal-to-noise ratio according to claim 1, characterized in that, In (1), the observation data collected under historical unobstructed conditions and the corresponding satellite elevation angle sequence are used. Building a reference model , is represented as: Further analysis of the signal-to-noise ratio observation sequence Compared with the reference model The deviation between them is used as the signal-to-noise ratio deviation sequence. , is represented as: Where a1, b1, and c1 are the model parameters of the reference model, This represents the satellite elevation angle θ at epoch t, ​​which is the reference model built based on historical observation data collected under unobstructed conditions. t The expected signal-to-noise ratio value.

4. The signal classification and discrimination method based on joint carrier phase and signal-to-noise ratio according to claim 1, characterized in that, In S4, the comprehensive discriminant D is represented as: ,in, , and These are short-time stability characteristics. Signal attenuation characteristics and signal fluctuation characteristics The weighting coefficients are expressed as: , , , It is a stability constant that is greater than zero.

5. The signal classification and discrimination method based on joint carrier phase and signal-to-noise ratio according to claim 1, characterized in that, In S5, the adaptive discrimination threshold T is analyzed based on historical observation data using the maximum inter-class variance criterion, and is expressed as: Where τ is the candidate threshold, and These represent the probabilities of classifying the classification results into distance propagation results and non-distance propagation results, respectively. and The values ​​are the mean values ​​of the classification results, which are divided into distance propagation results and non-distance propagation results.

6. The signal classification and discrimination method based on joint carrier phase and signal-to-noise ratio according to claim 1, characterized in that, In S6, when the comprehensive discriminant is greater than the adaptive discriminant threshold T, the classification result is a non-line-of-sight propagation discriminant result; otherwise, the classification result is a non-line-of-sight propagation discriminant result.

7. The signal classification and discrimination method based on joint carrier phase and signal-to-noise ratio according to claim 1, characterized in that, When the classification result is a non-line-of-sight propagation result, the weight of the comprehensive discrimination value is reduced, and the classification result and the reduced weight of the comprehensive discrimination value are input into the input of the subsequent process; When the classification result is a line-of-sight propagation result, the classification result and the comprehensive discrimination value are directly input into the input of the subsequent process.

8. The signal classification and discrimination method based on joint carrier phase and signal-to-noise ratio according to claim 7, characterized in that, The weighted comprehensive discriminant D' is expressed as: D'=1+ηD, 0<η<1, where η is the scaling factor.

9. A signal classification and discrimination device that combines carrier phase and signal-to-noise ratio, characterized in that, include: Storage; Storage is used to store computer programs; Actuator; The actuator is used to execute a computer program stored in the memory, which, when executed, implements the signal classification and discrimination method based on the combined carrier phase and signal-to-noise ratio as described in any one of claims 1-8.

Images