Signal identification method and device, electronic equipment and storage medium

Abstract

The invention relates to a signal identification method and device, electronic equipment and a storage medium. The signal identification method comprises the following steps: acquiring to-be-identified satellite observation data; the to-be-recognized satellite observation data are input into a trained signal classification model, the signal confidence is obtained, the signal classification model comprises a coding module, and the coding module is used for coding in combination with historical decoding features of historical satellite observation data of a previous epoch corresponding to the to-be-recognized satellite observation data; and determining whether the to-be-identified satellite observation data is a non-line-of-sight signal or not according to a size relationship between the signal confidence and a preset threshold. By adopting the method disclosed by the invention, whether the satellite observation data is the non-line-of-sight signal or not can be accurately identified.

Description

Technical Field

[0001] The present disclosure relates to the technical field of vehicles, and in particular, to a signal recognition method, apparatus, electronic device, and storage medium. Background Art

[0002] A Global Navigation Satellite System (GNSS) is configured on a vehicle. The GNSS realizes precise time measurement and ranging through signals transmitted by navigation satellites, and then provides all-weather, all-day, and high-quality positioning, navigation, and timing services to global users. However, the positioning performance of GNSS is severely limited by the environmental conditions where the receiver is located. Especially in complex urban environments, such as typical urban scenarios like urban canyons, viaducts, and tree-lined roads, non-line-of-sight (NLOS) signals greatly affect the fast and precise positioning performance of GNSS. Therefore, how to identify NLOS signals from navigation satellites is an urgent problem to be solved. Summary of the Invention

[0003] To overcome the problems existing in the related art, the present disclosure provides a signal recognition method, apparatus, electronic device, and storage medium to accurately identify whether satellite observation data is a non-line-of-sight signal.

[0004] According to a first aspect of an embodiment of the present disclosure, a signal recognition method is provided. The method includes:

[0005] Obtain satellite observation data to be recognized;

[0006] Input the satellite observation data to be recognized into a trained signal classification model to obtain a signal confidence level, where the signal classification model includes an encoding module, and the encoding module is used to perform encoding by combining historical decoding features of historical satellite observation data corresponding to the satellite observation data to be recognized in the previous epoch;

[0007] Determine whether the satellite observation data to be recognized is a non-line-of-sight signal according to the magnitude relationship between the signal confidence level and a preset threshold.

[0008] Optionally, the signal classification model further includes a decoding module and a classification module. The step of inputting the satellite observation data to be recognized into the trained signal classification model to obtain a signal confidence level includes:

[0009] Input the satellite observation data to be recognized into a first multi-layer perceptron of the signal classification model to obtain shallow features;

[0010] The encoding module performs feature fusion processing on the historical decoding features and the shallow features through a mutual attention mechanism to obtain fused encoding features;

[0011] Input the fused encoded feature into the decoding module to obtain the decoded feature of the satellite observation data to be recognized;

[0012] Input the decoded feature of the satellite observation data to be recognized into the classification module to obtain the signal confidence level.

[0013] Optionally, the encoding module includes a first mutual attention sub-module and a second multi-layer perceptron. The encoding module performs feature fusion processing on the historical decoded feature and the shallow feature through a mutual attention mechanism to obtain a fused encoded feature, including:

[0014] Input the shallow feature, position encoding, and the historical decoded feature into the first mutual attention sub-module for feature fusion processing to obtain a fused feature;

[0015] Input the fused feature into the second multi-layer perceptron for feature mapping to obtain the fused encoded feature.

[0016] Optionally, the satellite observation data to be recognized includes observation data from multiple satellites;

[0017] The encoding module includes a first self-attention sub-module, a first mutual attention sub-module, and a second multi-layer perceptron. The encoding module performs feature fusion processing on the historical decoded feature and the shallow feature through a mutual attention mechanism to obtain a fused encoded feature, including:

[0018] Input the shallow features of different satellites combined with position encoding into the first self-attention sub-module for feature interaction processing to obtain an interaction feature;

[0019] Input the interaction feature and the historical decoded feature into the first mutual attention sub-module for feature fusion processing to obtain a fused feature;

[0020] Input the fused feature into the second multi-layer perceptron for feature mapping to obtain the fused encoded feature.

[0021] Optionally, the step of inputting the fused encoded feature into the decoding module to obtain the decoded feature of the satellite observation data to be recognized includes:

[0022] Input the output query information combined with position encoding into the second self-attention sub-module of the decoding module to obtain a first feature information;

[0023] Input the fused encoded feature and the first feature information into the second mutual attention sub-module of the decoding module to obtain a second feature information;

[0024] Input the second feature information into the third multi-layer perceptron of the decoding module for feature mapping to obtain the decoded features of the satellite observation data to be recognized.

[0025] Optionally, the satellite observation data to be recognized includes at least one of the following data:

[0026] Satellite system information;

[0027] One-frequency pseudorange consistency;

[0028] Two-frequency pseudorange consistency;

[0029] One-frequency observable;

[0030] Two-frequency observable signal-to-noise ratio;

[0031] Satellite elevation angle.

[0032] Optionally, the signal classification model is trained as follows:

[0033] Use the satellite observation data collected on a vehicle traveling in a preset environment and the label indicating whether the satellite observation data is the non-line-of-sight signal as training samples;

[0034] Train the signal classification model according to the training samples.

[0035] According to the second aspect of the embodiments of the present disclosure, a signal recognition device is provided, including:

[0036] An acquisition module configured to acquire satellite observation data to be recognized;

[0037] An input module configured to input the satellite observation data to be recognized into a trained signal classification model to obtain a signal confidence level, where the signal classification model includes an encoding module, and the encoding module is used to encode by combining the historical decoded features of the historical satellite observation data corresponding to the satellite observation data to be recognized in the previous epoch;

[0038] An execution module configured to determine whether the satellite observation data to be recognized is a non-line-of-sight signal according to the magnitude relationship between the signal confidence level and a preset threshold.

[0039] Optionally, the signal classification model further includes a decoding module and a classification module, and the input module includes:

[0040] A first input sub-module configured to input the satellite observation data to be recognized into the first multi-layer perceptron of the signal classification model to obtain shallow features;

[0041] An encoding sub-module, configured to enable the encoding module to perform feature fusion processing on the historical decoding features and the shallow features through a mutual attention mechanism to obtain fused encoding features;

[0042] A second input sub-module, configured to input the fused encoding features into the decoding module to obtain decoding features of the satellite observation data to be recognized;

[0043] A third input sub-module, configured to input the decoding features of the satellite observation data to be recognized into the classification module to obtain the signal confidence.

[0044] Optionally, the encoding module includes a first mutual attention sub-module and a second multi-layer perceptron, and the encoding sub-module includes:

[0045] A first execution sub-module, configured to input the shallow features, positional encoding, and the historical decoding features into the first mutual attention sub-module for feature fusion processing to obtain fused features;

[0046] A second execution sub-module, configured to input the fused features into the second multi-layer perceptron for feature mapping to obtain the fused encoding features.

[0047] Optionally, the satellite observation data to be recognized includes observation data from multiple satellites;

[0048] The encoding module includes a first self-attention sub-module, a first mutual attention sub-module, and a second multi-layer perceptron, and the encoding sub-module includes:

[0049] A third execution sub-module, configured to input the shallow features of different satellites combined with positional encoding into the first self-attention sub-module for feature interaction processing to obtain interaction features;

[0050] A fourth execution sub-module, configured to input the interaction features and the historical decoding features into the first mutual attention sub-module for feature fusion processing to obtain fused features;

[0051] A fifth execution sub-module, configured to input the fused features into the second multi-layer perceptron for feature mapping to obtain the fused encoding features.

[0052] Optionally, the second input sub-module is configured to input output query information combined with positional encoding into the second self-attention sub-module of the decoding module to obtain first feature information; input the fused encoding features and the first feature information into the second mutual attention sub-module of the decoding module to obtain second feature information; input the second feature information into the third multi-layer perceptron of the decoding module for feature mapping to obtain the decoding features of the satellite observation data to be recognized.

[0053] Optionally, the satellite observation data to be recognized includes at least one of the following data:

[0054] Satellite system information;

[0055] Single-frequency pseudorange consistency;

[0056] Dual-frequency pseudorange consistency;

[0057] Single-frequency observable;

[0058] Dual-frequency observable signal-to-noise ratio;

[0059] Satellite elevation angle.

[0060] Optionally, the device further includes a training module configured to train the signal classification model in the following manner: using the satellite observation data collected on a vehicle traveling in a preset environment and the label indicating whether the satellite observation data is the NLOS signal as training samples; training the signal classification model according to the training samples.

[0061] According to a third aspect of the embodiments of the present disclosure, there is provided an electronic device including: a memory storing a computer program thereon; a processor configured to execute the computer program in the memory to implement the signal recognition method according to any one of the first aspect.

[0062] According to a fourth aspect of the embodiments of the present disclosure, there is provided a computer-readable storage medium storing computer program instructions thereon, and when the program instructions are executed by a processor, the steps of the signal recognition method provided in the first aspect of the present disclosure are implemented.

[0063] The technical solutions provided by the embodiments of the present disclosure may include the following beneficial effects:

[0064] In the case of obtaining the satellite observation data to be recognized, inputting the satellite observation data to be recognized into the trained signal classification model to obtain a signal confidence level. According to the magnitude relationship between the signal confidence level and a preset threshold, it can be determined whether the satellite observation data to be recognized is an NLOS signal. Since the encoding module in the signal classification model encodes by combining the historical decoding features of the historical satellite observation data corresponding to the satellite observation data to be recognized in the previous epoch, that is, the temporal information of the satellite observation data is introduced into the signal classification model for NLOS signal recognition, the accuracy and robustness of the signal recognition result can be improved. Therefore, by adopting the technical solution of the present disclosure, it is possible to accurately recognize whether the satellite observation data is an NLOS signal.

[0065] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and cannot limit the present disclosure. Description of the Drawings

[0066] The accompanying drawings herein are incorporated into and constitute a part of this specification, showing embodiments consistent with the present disclosure, and are used together with the specification to explain the principles of the present disclosure.

[0067] Figure 1 It is a flowchart of a signal recognition method shown according to an exemplary embodiment of the present disclosure.

[0068] Figure 2 It is a schematic diagram of the application of a signal classification model shown according to an exemplary embodiment.

[0069] Figure 3 It is a schematic diagram of the application of another signal classification model shown according to an exemplary embodiment.

[0070] Figure 4 It is a block diagram of a signal recognition device shown according to an exemplary embodiment of the present disclosure.

[0071] Figure 5 It is a block diagram of an electronic device for signal recognition shown according to an exemplary embodiment of the present disclosure.

[0072] Figure 6 It is a block diagram of another electronic device for signal recognition shown according to an exemplary embodiment of the present disclosure. Detailed implementation manners

[0073] Exemplary embodiments will be described in detail herein, and examples thereof are shown in the accompanying drawings. When the following description refers to the accompanying drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present disclosure. On the contrary, they are merely examples of devices and methods consistent with some aspects of the present disclosure as detailed in the appended claims.

[0074] It should be noted that all actions of obtaining signals, information, or data in the present disclosure are carried out on the premise of complying with the corresponding data protection regulations and policies of the country where it is located and obtaining authorization from the owner of the corresponding device.

[0075] In some embodiments of the present disclosure, in order to identify whether satellite observation data is a non-line-of-sight (NLOS) signal, the statistical characteristics of GNSS observables can be used to distinguish LOS / NLOS signals. For example, a rule-based classification method such as the Lau method can be used to distinguish LOS / NLOS signals, or a machine learning method such as the Hsu method can be used to distinguish LOS / NLOS signals. In other embodiments, additional information can be used to distinguish LOS / NLOS signals. For example, a fisheye camera image can be used to identify NLOS by determining whether the satellite is blocked, or an offline 3D map can be used to identify NLOS signals. However, these embodiments have the following limitations: 1) The LOS / NLOS classification method of traditional machine learning has a relatively simple algorithm, does not consider the temporal information of satellite observations, and has a low classification accuracy; 2) The method using additional information has a high cost and poor practicability.

[0076] In view of this, the present disclosure further provides a signal recognition method, apparatus, electronic device, and storage medium to obtain a more accurate LOS / NLOS signal classification result.

[0077] Figure 1 FIG. is a flowchart of a signal recognition method according to an exemplary embodiment of the present disclosure. The signal recognition method can be applied to both a terminal and a server. For example, the signal recognition method can be applied to an in-vehicle terminal, an in-vehicle GNSS system, or road traffic equipment. For another example, the signal recognition method can be applied to a cloud server communicatively connected to a vehicle. As Figure 1 shown, the signal recognition method may include the following steps.

[0078] In step S11, the satellite observation data to be recognized is acquired.

[0079] In some embodiments, any satellite observation data to be recognized can be used as the satellite observation data to be recognized in the embodiments of the present disclosure. For example, the satellite observation data received in real time can be used as the satellite observation data to be recognized. For example, the satellite observation data received at a historical time point can be used as the satellite observation data to be recognized. For example, simulation / simulation data can be used as the satellite observation data to be recognized.

[0080] In some embodiments, the satellite observation data to be recognized includes at least one of the following data: satellite system information sys, single-frequency pseudorange consistency PD1, dual-frequency pseudorange consistency PD2, single-frequency observable SNR1, dual-frequency observable signal-to-noise ratio SNR2, and satellite altitude angle AZEL.

[0081] In some embodiments, the satellite observation data to be recognized includes the observation data from one or more satellites.

[0082] In step S12, the satellite observation data to be recognized is input into the trained signal classification model to obtain the signal confidence. Among them, the signal classification model includes an encoding module, and the encoding module is used to perform encoding by combining the historical decoding features of the historical satellite observation data corresponding to the previous epoch of the satellite observation data to be recognized.

[0083] It should be explained here that an epoch is a term related to time calculation and is used to represent the starting point or reference point of a certain time system. An epoch is used to determine the starting point and chronology of time. In short, an epoch is the moment corresponding to astronomical data.

[0084] The historical satellite observation data refers to the satellite observation data of the previous epoch corresponding to the epoch of the satellite observation data to be recognized.

[0085] The historical decoding features of the historical satellite observation data refer to the information output by the decoding module in the signal classification model after the historical satellite observation data is input into the signal classification model.

[0086] In some embodiments, if there are no historical decoding features of the historical satellite observation data corresponding to the previous epoch for the satellite observation data to be recognized, then the preset decoding features or default decoding features can be used as the historical decoding features corresponding to the satellite observation data to be recognized. For example, the preset decoding features or default decoding features can be features filled with 0.

[0087] It should be noted here that when the satellite observation data to be recognized is the observation data from one satellite, the number of signal confidences output by the signal classification model is 1. When the satellite observation data to be recognized includes the observation data from multiple satellites, the number of signal confidences output by the signal classification model is multiple.

[0088] It should also be noted here that when an electromagnetic wave encounters geometries such as edges, obstacles, and buildings, it will reflect, refract, diffract, or pass through them, forming multiple transmission paths. Therefore, in the vehicle driving scenario, as the vehicle moves, there may be obstacles between the receiver on the vehicle and the navigation satellite, and the state of the existence of obstacles may last for a certain period of time. During this certain period of time, the signals received by the receiver should all be NLOS signals. Similarly, as the vehicle moves, there may be no obstacles between the receiver on the vehicle and the navigation satellite, and the state of the non-existence of obstacles may last for a certain period of time. During this certain period of time, the signals received by the receiver should all be LOS signals. That is to say, the signal type of the satellite observation data received in the previous epoch has a certain reference for the signal type of the satellite observation data received in the next epoch. In view of this, in the embodiments of the present disclosure, during the encoding process, the signal classification model encodes by combining the historical decoding features of the historical satellite observation data corresponding to the satellite observation data to be recognized in the previous epoch. Since the timing information is combined for NLOS signal recognition, the accuracy and robustness of the signal recognition result can be improved.

[0089] In step S13, according to the magnitude relationship between the signal confidence and the preset threshold, it is determined whether the satellite observation data to be recognized is a non-line-of-sight signal.

[0090] Among them, the signal classification model can be a binary classification model.

[0091] In some embodiments, when the number of signal confidences is multiple, for each signal confidence, by judging the magnitude relationship between the signal confidence and the preset threshold, it can be determined whether the observation data of the satellite corresponding to the signal confidence is a non-line-of-sight signal.

[0092] In some embodiments, the implementation manner of judging the magnitude relationship between the signal confidence and the preset threshold to determine whether the satellite observation data to be recognized is a non-line-of-sight signal may be that if it is determined that the signal confidence is greater than or equal to the preset threshold, it is determined that the satellite observation data to be recognized is a non-line-of-sight signal. If it is determined that the signal confidence is less than the preset threshold, it is determined that the satellite observation data to be recognized is not a non-line-of-sight signal.

[0093] The preset threshold is an empirical value and can be set according to requirements.

[0094] By adopting the above signal recognition method of the present disclosure, when the satellite observation data to be recognized is obtained, the satellite observation data to be recognized is input into the trained signal classification model to obtain the signal confidence level. According to the magnitude relationship between the signal confidence level and the preset threshold, it can be determined whether the satellite observation data to be recognized is a non-line-of-sight signal. Since the encoding module in the signal classification model encodes by combining the historical decoding features of the historical satellite observation data corresponding to the satellite observation data to be recognized in the previous epoch, that is, the temporal information of the satellite observation data is introduced into the signal classification model for NLOS signal recognition, the accuracy and robustness of the signal recognition result can be improved. Therefore, by adopting the above signal recognition method of the present disclosure, it can accurately recognize whether the satellite observation data is a non-line-of-sight signal.

[0095] In scenarios such as road traffic and autonomous driving, accurately identifying NLOS signals can effectively improve the positioning performance of GNSS rapid precise positioning technology in urban complex scenarios.

[0096] Optionally, the signal classification model includes a decoding module and a classification module. The implementation manner of the above step S12 of inputting the satellite observation data to be recognized into the trained signal classification model to obtain the signal confidence level may include:

[0097] Input the satellite observation data to be recognized into the first multi-layer perceptron of the signal classification model to obtain shallow features. The encoding module performs feature fusion processing on the historical decoding features and the shallow features through a mutual attention mechanism to obtain fused encoding features. Input the fused encoding features into the decoding module to obtain the decoding features of the satellite observation data to be recognized. Input the decoding features of the satellite observation data to be recognized into the classification module to obtain the signal confidence level.

[0098] In some embodiments, the first multi-layer perceptron refers to a multi-layer perceptron (MLP) connected to the encoding module.

[0099] In some embodiments, the classification module may be a classification head (MLP Head) composed of an MLP, which is used to perform binary classification processing of LOS / NLOS on the satellite observation data to be recognized to obtain the signal confidence level score.

[0100] Combine Figure 2For example, assume that the satellite observation data to be recognized includes the observation data from N satellites, and the observation data of each satellite includes 6-dimensional original features, namely, satellite system information sys, first-frequency pseudorange consistency PD1, second-frequency pseudorange consistency PD2, first-frequency observable SNR1, second-frequency observable signal-to-noise ratio SNR2, and satellite altitude angle AZEL. And if any dimension of the above 6-dimensional original features is missing in the observation data of a certain satellite, it can be supplemented by filling with 0. Based on this, the satellite observation data to be recognized can be represented as X k ∈R N×6 , where k represents the k-th epoch.

[0101] In some embodiments, the satellite observation data X to be recognized k ∈R N×6 is input into the first multi-layer perceptron MLP, and shallow features with a dimension of N×C can be extracted.

[0102] Further, the encoding module performs feature fusion processing on the historical decoded feature F k-1 and the shallow features through the mutual attention mechanism to obtain the fused encoded feature. In some embodiments, the feature fusion processing method includes but is not limited to weighting each feature component and selecting some components with a certain sampling strategy. Then, the fused encoded feature is input into the decoding module to obtain the decoded feature F k of the satellite observation data to be recognized. The decoded feature F k of the satellite observation data to be recognized is input into the classification head composed of MLP to obtain the signal confidence Y k . Y k includes N signal confidences.

[0103] In some embodiments, assume that the satellite observation data to be recognized is the observation data of one satellite. Then, the encoding module may include a first mutual attention sub-module and a second multi-layer perceptron. The implementation manner in which the encoding module performs feature fusion processing on the historical decoded feature and the shallow features through the mutual attention mechanism may include:

[0104] Input the shallow features, positional encoding, and historical decoded features into the first mutual attention sub-module for feature fusion processing to obtain the fused feature. Input the fused feature into the second multi-layer perceptron for feature mapping to obtain the fused encoded feature.

[0105] Among them, the positional encoding (PE) is determined by the classification model through the Position mechanism.

[0106] The first cross-attention sub-module can be a multi-head cross-attention module.

[0107] In some other embodiments, assuming that the satellite observation data to be recognized includes observation data from multiple satellites, then the encoding module can include a first self-attention sub-module, a first cross-attention sub-module, and a second multi-layer perceptron. The implementation of the encoding module performing feature fusion processing on the historical decoded features and the shallow features through the cross-attention mechanism to obtain the fused encoded features can include:

[0108] Combining the shallow features of different satellites with the positional encoding and inputting them into the first self-attention sub-module for feature interaction processing to obtain interaction features. Inputting the interaction features and the historical decoded features into the first cross-attention sub-module for feature fusion processing to obtain fused features. Inputting the fused features into the second multi-layer perceptron for feature mapping to obtain the fused encoded features.

[0109] Among them, the first self-attention sub-module can be a multi-head self-attention module (Multi-Head Self-Attention).

[0110] In some embodiments, according to the spatial position relationship, there is a certain relationship between the observation data from different satellites received by a receiver. By performing interaction processing on the observation data from different satellites, the features of the model (network) can be enhanced, thereby improving the accuracy and robustness of the NLOS recognition result.

[0111] Adopting the encoding module of the present disclosure, on the one hand, the satellite observation data (i.e., the observation value information) of different satellites in the same epoch is fused through the self-attention mechanism to enhance the network features, thereby effectively improving the accuracy and robustness of the NLOS recognition result. On the other hand, the historical decoded features extracted in the previous epoch are fused through the cross-attention mechanism, which can also effectively improve the accuracy and robustness of the NLOS recognition result. That is to say, the encoding module of the present disclosure that simultaneously uses the self-attention mechanism and the cross-attention mechanism, on the basis of introducing temporal information to improve the signal recognition accuracy, can further improve the signal recognition accuracy and robustness due to the further introduction of the self-attention mechanism to enhance the network features.

[0112] Optionally, the decoding module includes at least one of a second self-attention sub-module, a second cross-attention sub-module, and a third multi-layer perceptron. In some embodiments, the implementation of inputting the fused encoded features into the decoding module to obtain the decoded features of the satellite observation data to be recognized can include:

[0113] The output query information is combined with the positional encoding and input into the second self-attention sub-module of the decoding module to obtain the first feature information. The fused encoded feature and the first feature information are input into the second cross-attention sub-module of the decoding module to obtain the second feature information. The second feature information is input into the third multi-layer perceptron of the decoding module for feature mapping to obtain the decoded feature of the satellite observation data to be recognized.

[0114] Among them, the output query information is a learnable N×C-dimensional output query (Queries), which is a set of randomly initialized N×C parameters and changes through learning during the model training process.

[0115] In some embodiments, the signal classification model in the embodiments of the present disclosure is trained in the following manner:

[0116] The satellite observation data collected on a vehicle traveling in a preset environment and the label indicating whether the satellite observation data is a non-line-of-sight signal are used as training samples; the signal classification model is trained according to the training samples.

[0117] Among them, the preset environment may refer to an environment that easily causes obstacles between the receiver on the vehicle and the satellite, such as typical urban scenes such as urban canyons, viaducts, and tree-lined roads.

[0118] Figure 3 It is a schematic block diagram of a signal classification model shown according to an exemplary embodiment of the present disclosure. The following combines Figure 3 to illustrate the processing process of the signal classification model:

[0119] First, the satellite observation data X to be recognized in the K-th epoch k is input into the first MPL of the signal classification model to obtain the shallow feature The shallow feature is combined with the positional encoding PE and input into the first self-attention sub-module for feature interaction between different satellites to obtain the interaction feature. Then, through the first cross-attention sub-module, the historical decoded feature F k-1 of the previous epoch is fused with the interaction feature of the current epoch to introduce temporal information and obtain the fused feature. The fused feature is subjected to feature mapping through the second MLP to obtain the fused encoded feature output by the encoding module.

[0120] Next, the output query information is combined with the positional encoding PE and input into the second self-attention sub-module of the decoding module to obtain the first feature information. The fused encoded feature output by the encoding module and the first feature information are input into the second cross-attention sub-module of the decoding module to obtain the second feature information. The second feature information is input into the third MLP of the decoding module for feature mapping to obtain the decoded feature F k .

[0121] Finally, input the decoded feature F k into the classification module for classification to obtain the signal confidence Y k .

[0122] Figure 4 is a block diagram of a signal recognition device shown according to an exemplary embodiment of the present disclosure. Referring to Figure 4 , the signal recognition device 400 includes:

[0123] An acquisition module 401, configured to acquire satellite observation data to be recognized;

[0124] An input module 402, configured to input the satellite observation data to be recognized into a trained signal classification model to obtain a signal confidence, wherein the signal classification model includes an encoding module, and the encoding module is used to encode by combining the historical decoded features of the historical satellite observation data corresponding to the satellite observation data to be recognized in the previous epoch;

[0125] An execution module 403, configured to determine whether the satellite observation data to be recognized is a non-line-of-sight signal according to the magnitude relationship between the signal confidence and a preset threshold.

[0126] With the above signal recognition device, when the satellite observation data to be recognized is acquired, the satellite observation data to be recognized can be input into a trained signal classification model to obtain a signal confidence. According to the magnitude relationship between the signal confidence and a preset threshold, it can be determined whether the satellite observation data to be recognized is a non-line-of-sight signal. Since the encoding module in the signal classification model encodes by combining the historical decoded features of the historical satellite observation data corresponding to the satellite observation data to be recognized in the previous epoch, that is, the temporal information of the satellite observation data is introduced into the signal classification model for NLOS signal recognition, the accuracy and robustness of the signal recognition result can be improved. Therefore, by adopting the technical solution of the present disclosure, it is possible to accurately identify whether the satellite observation data is a non-line-of-sight signal.

[0127] Optionally, the signal classification model further includes a decoding module and a classification module, and the input module 402 includes:

[0128] A first input sub-module, configured to input the satellite observation data to be recognized into the first multi-layer perceptron of the signal classification model to obtain shallow features;

[0129] An encoding sub-module, configured to the encoding module perform feature fusion processing on the historical decoded features and the shallow features through a mutual attention mechanism to obtain fused encoded features;

[0130] A second input sub-module, configured to input the fused encoded features into the decoding module to obtain the decoded features of the satellite observation data to be recognized;

[0131] A third input sub-module, configured to input the decoded features of the satellite observation data to be recognized into the classification module to obtain the signal confidence.

[0132] Optionally, the encoding module includes a first cross-attention sub-module and a second multi-layer perceptron. The encoding sub-module includes:

[0133] A first execution sub-module, configured to input the shallow features, positional encoding, and the historical decoded features into the first cross-attention sub-module for feature fusion processing to obtain fused features;

[0134] A second execution sub-module, configured to input the fused features into the second multi-layer perceptron for feature mapping to obtain the fused encoded features.

[0135] Optionally, the satellite observation data to be recognized includes observation data from multiple satellites;

[0136] The encoding module includes a first self-attention sub-module, a first cross-attention sub-module, and a second multi-layer perceptron. The encoding sub-module includes:

[0137] A third execution sub-module, configured to input the shallow features of different satellites combined with positional encoding into the first self-attention sub-module for feature interaction processing to obtain interaction features;

[0138] A fourth execution sub-module, configured to input the interaction features and the historical decoded features into the first cross-attention sub-module for feature fusion processing to obtain fused features;

[0139] A fifth execution sub-module, configured to input the fused features into the second multi-layer perceptron for feature mapping to obtain the fused encoded features.

[0140] Optionally, the second input sub-module is configured to input the output query information combined with positional encoding into the second self-attention sub-module of the decoding module to obtain first feature information; input the fused encoded features and the first feature information into the second cross-attention sub-module of the decoding module to obtain second feature information; and input the second feature information into the third multi-layer perceptron of the decoding module for feature mapping to obtain the decoded features of the satellite observation data to be recognized.

[0141] Optionally, the satellite observation data to be recognized includes at least one of the following data:

[0142] Satellite system information;

[0143] One-frequency pseudorange consistency;

[0144] Two-frequency pseudorange consistency;

[0145] One-frequency observable;

[0146] Signal-to-noise ratio of two-frequency observables;

[0147] Satellite elevation angle.

[0148] Optionally, the signal recognition device 400 further includes a training module configured to train the signal classification model in the following manner: using the satellite observation data collected on a vehicle traveling in a preset environment and the label indicating whether the satellite observation data is the NLOS signal as training samples; training the signal classification model according to the training samples.

[0149] Regarding the device in the above embodiments, the specific manners in which each module performs operations have been described in detail in the embodiments related to the method, and will not be elaborated herein.

[0150] The present disclosure also provides a computer-readable storage medium having computer program instructions stored thereon, and when the program instructions are executed by a processor, the steps of the signal recognition method provided by the present disclosure are implemented.

[0151] Figure 5 FIG. is a block diagram of an electronic device 800 for signal recognition according to an exemplary embodiment of the present disclosure. For example, the electronic device 800 may be a vehicle, an in-vehicle device, an in-vehicle satellite navigation system, a mobile phone, a computer, a digital broadcast terminal, a messaging device, a game console, a tablet device, a medical device, a fitness device, a personal digital assistant, etc.

[0152] Refer to Figure 5 , the electronic device 800 may include one or more of the following components: a processing component 802, a memory 804, a power component 806, a multimedia component 808, an audio component 810, an input / output interface 812, a sensor component 814, and a communication component 816.

[0153] The processing component 802 generally controls the overall operation of the electronic device 800, such as operations associated with display, telephone call, data communication, camera operation, and recording operation. The processing component 802 may include one or more processors 820 to execute instructions to complete all or part of the steps of the above method. In addition, the processing component 802 may include one or more modules to facilitate the interaction between the processing component 802 and other components. For example, the processing component 802 may include a multimedia module to facilitate the interaction between the multimedia component 808 and the processing component 802.

[0154] The memory 804 is configured to store various types of data to support the operation of the electronic device 800. Examples of such data include instructions for any application or method operating on the electronic device 800, contact data, phone book data, messages, pictures, videos, and the like. The memory 804 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disks, or optical disks.

[0155] The power supply component 806 provides power to various components of the electronic device 800. The power supply component 806 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power for the electronic device 800.

[0156] The multimedia component 808 includes a screen that provides an output interface between the electronic device 800 and the user. In some embodiments, the screen may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen can be implemented as a touch screen to receive input signals from the user. The touch panel includes one or more touch sensors to sense touches, swipes, and gestures on the touch panel. The touch sensors can not only sense the boundaries of the touch or swipe actions, but also detect the duration and pressure associated with the touch or swipe operation. In some embodiments, the multimedia component 808 includes a front camera and / or a rear camera. When the electronic device 800 is in an operating mode, such as a shooting mode or a video mode, the front camera and / or the rear camera can receive external multimedia data. Each of the front camera and the rear camera can be a fixed optical lens system or have a focal length and optical zoom capabilities.

[0157] The audio component 810 is configured to output and / or input audio signals. For example, the audio component 810 includes a microphone (MIC) that is configured to receive external audio signals when the electronic device 800 is in an operating mode, such as a call mode, a recording mode, and a voice recognition mode. The received audio signals can be further stored in the memory 804 or transmitted via the communication component 816. In some embodiments, the audio component 810 further includes a speaker for outputting audio signals.

[0158] The input / output interface 812 provides an interface between the processing component 802 and a peripheral interface module, and the peripheral interface module may be a keyboard, a click wheel, buttons, etc. These buttons may include, but are not limited to: a home button, a volume button, a power button, and a lock button.

[0159] The sensor component 814 includes one or more sensors for providing an assessment of the status of the electronic device 800 in various aspects. For example, the sensor component 814 can detect the on / off state of the electronic device 800, the relative positioning of components, such as the display and keypad of the electronic device 800, the sensor component 814 can also detect a change in the position of the electronic device 800 or a component of the electronic device 800, the presence or absence of user contact with the electronic device 800, the orientation or acceleration / deceleration of the electronic device 800, and the temperature change of the electronic device 800. The sensor component 814 may include a proximity sensor configured to detect the presence of nearby objects without any physical contact. The sensor component 814 may also include a light sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, the sensor component 814 may further include an acceleration sensor, a gyro sensor, a magnetic sensor, a pressure sensor, or a temperature sensor.

[0160] The communication component 816 is configured to facilitate communication between the electronic device 800 and other devices in a wired or wireless manner. The electronic device 800 can access a wireless network based on communication standards, such as WiFi, 2G, or 3G, or a combination thereof. In an exemplary embodiment, the communication component 816 receives a broadcast signal or broadcast-related information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, the communication component 816 further includes a near field communication (NFC) module to facilitate short-range communication. For example, the NFC module may be implemented based on radio frequency identification (RFID) technology, infrared data association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology, and other technologies.

[0161] In an exemplary embodiment, the electronic device 800 may be implemented by one or more application specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components for performing the above signal recognition method.

[0162] In an exemplary embodiment, a non-transitory computer-readable storage medium including instructions is also provided, such as a memory 804 including instructions, and the above instructions can be executed by a processor 820 of an electronic device 800 to complete the above method. For example, the non-transitory computer-readable storage medium may be a ROM, a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, etc.

[0163] In addition to being an independent electronic device, the above electronic device may also be a part of an independent electronic device. For example, in one embodiment, the device may be an integrated circuit (IC) or a chip. The integrated circuit may be a single IC or a collection of multiple ICs; the chip may include, but is not limited to, the following types: GPU (Graphics Processing Unit), CPU (Central Processing Unit), FPGA (Field Programmable Gate Array), DSP (Digital Signal Processor), ASIC (Application Specific Integrated Circuit), SOC (System on Chip), etc. The above integrated circuit or chip may be used to execute executable instructions (or code) to implement the above signal recognition method. The executable instructions may be stored in the integrated circuit or chip, or may be obtained from other devices or equipment. For example, the integrated circuit or chip includes a processor, a memory, and an interface for communicating with other devices. The executable instructions may be stored in the memory, and when the executable instructions are executed by the processor, the above signal recognition method is implemented; or, the integrated circuit or chip may receive the executable instructions through the interface and transmit them to the processor for execution to implement the above signal recognition method.

[0164] In another exemplary embodiment, a computer program product is also provided. The computer program product includes a computer program that can be executed by a programmable device, and the computer program has a code portion for executing the above signal recognition method when executed by the programmable device.

[0165] Figure 6 is a block diagram of an electronic device 1900 for signal recognition shown according to an exemplary embodiment of the present disclosure. For example, the electronic device 1900 may be provided as a server. For example, the electronic device 1900 is a cloud server communicatively connected to a satellite navigation system on a vehicle. Refer to Figure 6, the electronic device 1900 includes a processing component 1922, which further includes one or more processors, and memory resources represented by a memory 1932 for storing instructions executable by the processing component 1922, such as application programs. The application programs stored in the memory 1932 may include one or more modules each corresponding to a set of instructions. In addition, the processing component 1922 is configured to execute instructions to perform the above signal recognition method.

[0166] The electronic device 1900 may further include a power component 1926 configured to perform power management of the electronic device 1900, a wired or wireless network interface 1950 configured to connect the electronic device 1900 to a network, and an input / output interface 1958. The electronic device 1900 may operate based on an operating system stored in the memory 1932, such as Windows Server TM , Mac OS X TM , Unix TM , Linux TM , FreeBSD TM or the like.

[0167] Those skilled in the art will readily conceive of other embodiments of the present disclosure after considering the specification and practicing the present disclosure. The present disclosure is intended to cover any variations, uses, or adaptations of the present disclosure that follow the general principles of the present disclosure and include known common knowledge or conventional technical means in the technical field not disclosed by the present disclosure. The specification and examples are only considered exemplary, and the true scope and spirit of the present disclosure are pointed out by the following claims.

[0168] It should be understood that the present disclosure is not limited to the exact structures described above and shown in the drawings, and various modifications and changes can be made without departing from its scope. The scope of the present disclosure is only limited by the appended claims.

Claims

1. A signal recognition method, characterized in that, The method includes: Obtaining satellite observation data to be recognized; Inputting the satellite observation data to be recognized into a trained signal classification model to obtain a signal confidence level, where the signal classification model includes an encoding module for encoding by combining historical decoding features of historical satellite observation data corresponding to the satellite observation data to be recognized in the previous epoch; Determining whether the satellite observation data to be recognized is a non-line-of-sight signal according to the magnitude relationship between the signal confidence level and a preset threshold.

2. The method according to claim 1, wherein The signal classification model further includes a decoding module and a classification module. The step of inputting the satellite observation data to be recognized into a trained signal classification model to obtain a signal confidence level includes: Inputting the satellite observation data to be recognized into a first multi-layer perceptron of the signal classification model to obtain shallow features; The encoding module performs feature fusion processing on the historical decoding features and the shallow features through a mutual attention mechanism to obtain fused encoding features; Inputting the fused encoding features into the decoding module to obtain decoding features of the satellite observation data to be recognized; Inputting the decoding features of the satellite observation data to be recognized into the classification module to obtain the signal confidence level.

3. The method according to claim 2, wherein The encoding module includes a first mutual attention sub-module and a second multi-layer perceptron. The encoding module performs feature fusion processing on the historical decoding features and the shallow features through a mutual attention mechanism to obtain fused encoding features, including: Inputting the shallow features, position encoding, and the historical decoding features into the first mutual attention sub-module for feature fusion processing to obtain fused features; Inputting the fused features into the second multi-layer perceptron for feature mapping to obtain the fused encoding features.

4. The method according to claim 2, wherein The satellite observation data to be recognized includes observation data from multiple satellites; The encoding module includes a first self-attention sub-module, a first mutual attention sub-module, and a second multi-layer perceptron. The encoding module performs feature fusion processing on the historical decoding features and the shallow features through a mutual attention mechanism to obtain fused encoding features, including: Inputting the shallow features of different satellites combined with position encoding into the first self-attention sub-module for feature interaction processing to obtain interaction features; Inputting the interaction features and the historical decoding features into the first mutual attention sub-module for feature fusion processing to obtain fused features; Inputting the fused features into the second multi-layer perceptron for feature mapping to obtain the fused encoding features.

5. The method according to any one of claims 2-4, characterized in that, The step of inputting the fused encoding features into the decoding module to obtain decoding features of the satellite observation data to be recognized includes: Inputting output query information combined with position encoding into a second self-attention sub-module of the decoding module to obtain first feature information; Inputting the fused encoding features and the first feature information into a second mutual attention sub-module of the decoding module to obtain second feature information; Inputting the second feature information into a third multi-layer perceptron of the decoding module for feature mapping to obtain decoding features of the satellite observation data to be recognized.

6. The method according to any one of claims 1-4, characterized in that, The satellite observation data to be recognized includes at least one of the following data: Satellite system information; Single-frequency pseudorange consistency; Dual-frequency pseudorange consistency; Single-frequency observable; Dual-frequency observable signal-to-noise ratio; Satellite elevation angle.

7. The method according to claim 6, wherein The signal classification model is trained in the following manner: Using the satellite observation data collected on a vehicle traveling in a preset environment and the label indicating whether the satellite observation data is the NLOS signal as training samples; Training the signal classification model according to the training samples.

8. A signal recognition device, characterized in that, Including: An acquisition module configured to acquire satellite observation data to be recognized; An input module configured to input the satellite observation data to be recognized into the trained signal classification model to obtain a signal confidence level, where the signal classification model includes an encoding module for encoding by combining the historical decoding features of the historical satellite observation data corresponding to the satellite observation data to be recognized in the previous epoch; An execution module configured to determine whether the satellite observation data to be recognized is an NLOS signal according to the magnitude relationship between the signal confidence level and a preset threshold.

9. An electronic device, characterized in that, Including: A memory storing a computer program thereon; A processor for executing the computer program in the memory to implement the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having computer program instructions stored thereon, characterized in that, When the program instructions are executed by the processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images