GPS signal management method and system based on artificial intelligence

Abstract

The present invention discloses a GPS signal management method and system based on artificial intelligence, which belongs to the field of artificial intelligence technology. The method and system include receiving original signals of multiple GPS satellites, performing preliminary screening and noise filtering preprocessing. The method builds a geographic environment model based on signal characteristics in combination with multiple sensors, and adjusts GPS parameters in real time according to environmental changes, so as to adapt to different environmental conditions and improve the adaptability of the signal. The method also performs dynamic signal optimization and enhancement through an intelligent antenna array, and can dynamically adjust the antenna according to the signal strength and the predicted multi-path signal state, so as to further enhance the stability of the signal. The optimized and enhanced GPS signal and the data of multiple sensors are fused for high-precision positioning. By fusing the GPS signal and the geographic environment model information data, the error correction of the initial position estimation can be performed, and the fused position estimation is matched and verified with the geographic environment model, so as to significantly improve the positioning accuracy.

Description

Technical Field

[0001] The present invention belongs to the field of artificial intelligence technology, and specifically refers to a GPS signal management method and system based on artificial intelligence. Background Art

[0002] With the development of modern science and technology, GPS technology has been widely used in vehicle navigation, mobile device positioning, and precision time synchronization. However, in urban canyons, high-rise buildings, or environments with severe electromagnetic interference, GPS signals are easily blocked or reflected, resulting in reduced positioning accuracy or signal loss.

[0003] However, the existing GPS signal management still has certain defects. The existing GPS signal management may not be flexible enough or respond slowly in constructing geographic environment models and adjusting GPS parameters in real time, resulting in the inability to make timely adjustments in rapidly changing environments and the inability to fully handle synchronization errors and measurement noise from different sensors, which directly affects the accuracy of the final position estimation. In complex and changeable urban environments, even optimized and enhanced GPS signals may fail due to sudden interference, affecting the stability of positioning accuracy. Three-dimensional geographic environment modeling relies on static map data and limited historical information, and is insufficient to respond to dynamic changes. Therefore, a GPS signal management method and system based on artificial intelligence are proposed. Summary of the invention

[0004] The purpose of the present invention is to provide a GPS signal management method and system based on artificial intelligence to solve the problems raised in the above background technology.

[0005] To achieve the above object, the present invention provides the following technical solution: a GPS signal management method based on artificial intelligence, comprising the following steps:

[0006] S1, receiving the original signals of multiple GPS satellites, and performing preliminary screening and noise filtering preprocessing;

[0007] S2. According to the pre-processed signal information, identify and classify different types of signals and eliminate multipath interference sources, and predict the multipath signal state for compensation;

[0008] S3, based on signal characteristics, combined with multiple sensors to build a geographic environment model, and adjust GPS parameters in real time according to environmental changes;

[0009] S4. Based on the signal recognition and prediction results, dynamic signal optimization and enhancement are performed in combination with the smart antenna array;

[0010] S5. Perform high-precision positioning based on optimized and enhanced GPS signals and multi-sensor data fusion;

[0011] S6. Monitor GPS signal quality and environmental changes in real time, automatically adjust various working parameters and provide feedback.

[0012] Among them, the S1 receives the original signals of multiple GPS satellites, performs preliminary screening and noise filtering; simultaneously receives the RF signals of multiple satellites through the GPS receiver, samples the received analog RF signals at high speed and converts them into digital signals, maps the continuous analog RF signal values ​​to a limited number of discrete values ​​for quantization, generates a reference signal matching the expected GPS carrier frequency through a local oscillator, mixes it with the received RF signal, down-converts the RF signal to baseband, synchronizes the satellite signal frequency, identifies and tracks the signal through code phase capture based on the quantized RF signal, evaluates and screens the signal-to-noise ratio of each signal, performs noise filtering on the screened signals, and compresses, packages and transmits the screened and filtered signals.

[0013] Among them, S2, according to the preprocessed signal information, identifies and classifies different types of signals and eliminates multipath interference sources, and predicts the multipath signal state for resource scheduling; assuming that the preprocessed signal y(t) includes signal components and noise, the preprocessed signal is feature extracted, and the implementation formula is:

[0014] x(t)=Fq(y(t))-w(t)·n(t),

[0015] In the formula, x(t) represents the extracted signal feature vector, y(t) represents the preprocessed signal, Fq represents the feature extraction function, w(t) represents the adaptive weight vector, and w(t) updates the adaptive weight vector by minimizing the mean error. The implementation formula is:

[0016] w(t+1)=w(t)+μn(t)(y(t)-x(t)) T ,

[0017] In the formula, w(t+1) represents the updated weight, and μ represents the step size parameter;

[0018] According to the extracted signal feature vector, the signal feature vector is classified, the frequency domain is converted, and the energy distribution is calculated. The implementation formula is:

[0019] E in (f)=|F{x(t)}| 2 ,

[0020] In the formula, E in (f) represents the energy distribution of the signal, the signal energy intensity at frequency f, F represents the conversion of the time domain signal into the frequency domain, and x(t) represents the extracted signal feature vector;

[0021] Assume that the energy distribution model of each category c is E c (f) Calculate the distance between the energy distribution of the input signal and each distribution model. The implementation formula is:

[0022] D(c)=∫|E in (f)-E c (f)| 2 df,

[0023] In the formula, D(c) represents the difference between the energy distribution of the input signal and the model of category c, df represents a small change in frequency f, and the final classification result c * for:

[0024] c * =argmin c D(c).

[0025] Among them, S2 performs channel response modeling H(t,τ) according to the classified signal, and the modeling implementation formula is:

[0026] H(t,τ)=∑ k a k (t)δ(τ-τ k (t)),

[0027] In the formula, H(t,τ) represents the dynamic channel response model, the channel characteristics at time t and delay τ, and a k (t) represents the amplitude of the kth path, the signal strength at time t, τ k (t) represents the delay of the kth path, the signal delay at time t;

[0028] According to the channel response model, let the received signal be r(t), the transmitted signal be s(t), and the compensated signal be It is expressed as:

[0029]

[0030] In the formula, represents the compensated signal, and h(t) represents H(t,τ) converted to the frequency domain representation:

[0031]

[0032] In the formula, F represents the conversion of time domain signal into frequency domain representation, a k (t) represents the amplitude of the kth path, the signal strength at time t, represents the phase shift factor in the frequency domain, e represents the base of the natural logarithm, j represents the imaginary unit, f represents the frequency variable, τ k(t) represents the delay of the kth path, the signal delay at time t, and the inverse filter h of h(t) is calculated. -1 (t), the realization formula is:

[0033]

[0034] In the formula, h -1 (t) represents the inverse filter of the channel response h(t), s(t) represents the transmitted signal, the transmitted signal at time t, g(t) represents the filter to be optimized, E represents the expected value operator, and r(t) represents the received signal, the actual received signal at time t.

[0035] Wherein, in S2, the environmental perception data is assumed to be Es t , combined with the compensated signal To predict the multipath signal status, the implementation formula is:

[0036]

[0037] In the formula, X t+1 represents the multipath signal state prediction, represents the compensated signal, α represents the weighting factor, β(Es t ) indicates the possible multipath change trend predicted based on the current environmental characteristics;

[0038] Compensation is performed based on the predicted multipath status results, and the implementation formula is:

[0039] R i (t+1) = γP t +(1-γ)[R i (t)+λX t+1 ],

[0040] In the formula, R i (t+1) represents the demand for the i-th resource at time t+1, P t represents the predicted multipath strength at time t, γ represents the adjustment factor, and X t+1 Represents the multipath signal status prediction.

[0041] The pre-processed signals are identified and classified to distinguish between direct path signals and multipath signals. Through identification and classification, the multipath interference source can be accurately identified and corresponding measures can be taken to eliminate it. At the same time, the state of the multipath signal can be predicted and compensation can be performed based on the prediction results, further reducing the impact of the multipath effect on positioning accuracy.

[0042] Among them, S3, based on signal characteristics, combines multiple sensors to build a geographic environment model, and adjusts GPS parameters in real time according to environmental changes; obtains signal characteristics x(t), combines the data collected by the sensor in real time to identify building outlines and static obstacles, builds a three-dimensional geographic environment model based on the characteristics, updates the environment model in real time according to real-time environmental changes, and dynamically adjusts GPS receiver parameters according to the obstacle distribution in the environment model.

[0043] By updating the environmental model in real time and dynamically adjusting the parameters of the GPS receiver according to environmental changes, the system's adaptability in complex environments is improved, enabling the system to maintain stable positioning performance in different geographical environments. By constructing a geographical environment model, it can provide useful information for subsequent path planning and navigation, further improving the system's practicality and intelligence.

[0044] Wherein, S4, according to the signal recognition result and the prediction result, combines the intelligent antenna array to perform dynamic signal optimization and enhancement; according to the classification result c obtained in S2 * , signal response model H(t,τ) and multipath signal state prediction result X t+1 According to the initial environmental conditions, the initial parameters of the smart antenna array are set. According to the signal recognition results and the channel response model, the beam direction and shape of the antenna array are dynamically adjusted. By considering the influence of the Doppler effect, the frequency offset is corrected in real time to keep the signal locked. According to the signal strength and the predicted multipath signal status, the antenna is dynamically adjusted to optimize the signal and enhance the signal.

[0045] The intelligent antenna array can dynamically adjust the beam direction and shape of the antenna array according to the signal recognition results and channel response model to maximize the signal reception strength and reduce interference. At the same time, the system can also consider the influence of the Doppler effect, correct the frequency offset in real time, and maintain signal lock. By dynamically optimizing the enhanced signal, the system can more accurately receive and analyze satellite signals, thereby improving the accuracy and stability of positioning.

[0046] Among them, the S5 performs high-precision positioning based on the optimized and enhanced GPS signal and multi-sensor data fusion, receives the optimized and enhanced GPS signal and geographic environment model information, fuses the GPS signal and the geographic environment model information data, corrects the error of the initial position estimate according to the fused data, matches the fused position estimate with the geographic environment model, verifies whether the current position is in the expected area, and dynamically adjusts the fusion weights and filtering parameters according to real-time environmental changes.

[0047] By fusing GPS signals and geographic environment model information data, the system can correct errors in the initial position estimate and match and verify the fused position estimate with the geographic environment model. By fusing multi-source data, it can more comprehensively understand the current environment and accurately estimate the position information. By matching and verifying with the geographic environment model, it can further confirm the accuracy of the current position and eliminate potential positioning errors.

[0048] Among them, the GPS signal management system based on artificial intelligence includes:

[0049] Data acquisition preprocessing module: used to receive the original RF signals from multiple GPS satellites, and perform preliminary screening and noise filtering;

[0050] Intelligent multipath processing module: Based on the pre-processed signal information, it identifies and classifies different types of signals, eliminates multipath interference sources, and predicts the multipath signal status for resource scheduling;

[0051] Adaptive environmental perception adjustment module: It builds a geographic environment model by combining multi-sensor data and adjusts GPS parameters in real time according to environmental changes;

[0052] Dynamic signal optimization and enhancement module: Based on the signal recognition and prediction results, it combines the smart antenna array to perform dynamic signal optimization and enhancement;

[0053] High-precision positioning module: high-precision positioning is performed based on optimized and enhanced GPS signals and multi-sensor data fusion;

[0054] Real-time feedback module: used to monitor GPS signal quality and environmental changes in real time, automatically adjust various working parameters and provide feedback.

[0055] Compared with the prior art, the present invention has the following beneficial effects:

[0056] 1. The present invention constructs a geographic environment model based on signal characteristics and combines multiple sensors, and adjusts GPS parameters in real time according to environmental changes, so that the system can adapt to different environmental conditions and improve the adaptability of the signal. It also performs dynamic signal optimization and enhancement through an intelligent antenna array, and can dynamically adjust the antenna according to the signal strength and the predicted multipath signal state, further enhancing the stability of the signal.

[0057] 2. The present invention performs high-precision positioning by optimizing and enhancing the GPS signal and multi-sensor data fusion. By fusing the GPS signal with the geographic environment model information data, the error of the initial position estimate can be corrected, and the fused position estimate can be matched and verified with the geographic environment model, which significantly improves the positioning accuracy.

[0058] 3. The present invention constructs a three-dimensional geographic environment model by combining the received signal characteristics with the data collected in real time by multiple sensors. By updating the environment model in real time, the parameters of the GPS receiver can be dynamically adjusted according to environmental changes, the signal reception effect can be optimized, the adaptability in complex environments can be improved, and stable positioning performance can be maintained in different geographical environments;

[0059] 4. By real-time monitoring of signal quality and environmental changes, the system can promptly detect and respond to potential positioning errors and interference factors. At the same time, by automatically adjusting working parameters and providing feedback, it can continuously optimize its own performance and adapt to different application scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 The operation flow of the GPS signal management method based on artificial intelligence of the present invention is as follows Figure 1 ;

[0061] Figure 2 The operation flow of the GPS signal management method based on artificial intelligence of the present invention is as follows Figure 2 ;

[0062] Figure 3 The operation flow of the GPS signal management method based on artificial intelligence of the present invention is as follows Figure 3 ;

[0063] Figure 4 The figure is a schematic diagram of the structure of the GPS signal management system based on artificial intelligence of the present invention. DETAILED DESCRIPTION

[0064] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0065] Example

[0066] See also Figure 1-Figure 4 As shown, the present invention provides a technical solution: comprising the following steps:

[0067] S1, receiving the original signals of multiple GPS satellites, and performing preliminary screening and noise filtering preprocessing;

[0068] S2. According to the pre-processed signal information, identify and classify different types of signals and eliminate multipath interference sources, and predict the multipath signal state for compensation;

[0069] S3, based on signal characteristics, combined with multiple sensors to build a geographic environment model, and adjust GPS parameters in real time according to environmental changes;

[0070] S4. Based on the signal recognition and prediction results, dynamic signal optimization and enhancement are performed in combination with the smart antenna array;

[0071] S5. Perform high-precision positioning based on optimized and enhanced GPS signals and multi-sensor data fusion;

[0072] S6. Monitor GPS signal quality and environmental changes in real time, automatically adjust various working parameters and provide feedback.

[0073] Among them, the S1 receives the original signals of multiple GPS satellites, performs preliminary screening and noise filtering; simultaneously receives the RF signals of multiple satellites through the GPS receiver, samples the received analog RF signals at high speed and converts them into digital signals, maps the continuous analog RF signal values ​​to a limited number of discrete values ​​for quantization, generates a reference signal matching the expected GPS carrier frequency through a local oscillator, mixes it with the received RF signal, down-converts the RF signal to baseband, synchronizes the satellite signal frequency, identifies and tracks the signal through code phase capture based on the quantized RF signal, evaluates and screens the signal-to-noise ratio of each signal, performs noise filtering on the screened signals, and compresses, packages and transmits the screened and filtered signals.

[0074] Among them, S2, according to the preprocessed signal information, identifies and classifies different types of signals and eliminates multipath interference sources, and predicts the multipath signal state for resource scheduling; assuming that the preprocessed signal y(t) includes signal components and noise, the preprocessed signal is feature extracted, and the implementation formula is:

[0075] x(t)=Fq(y(t))-w(t)·n(t),

[0076] In the formula, x(t) represents the extracted signal feature vector, y(t) represents the preprocessed signal, Fq represents the feature extraction function, w(t) represents the adaptive weight vector, and w(t) updates the adaptive weight vector by minimizing the mean error. The implementation formula is:

[0077] w(t+1)=w(t)+μn(t)(y(t)-x(t)) T ,

[0078] In the formula, w(t+1) represents the updated weight, and μ represents the step size parameter;

[0079] According to the extracted signal feature vector, the signal feature vector is classified, the frequency domain is converted, and the energy distribution is calculated. The implementation formula is:

[0080] E in (f)=|F{x(t)}| 2 ,

[0081] In the formula, E in (f) represents the energy distribution of the signal, the signal energy intensity at frequency f, F represents the conversion of the time domain signal into the frequency domain, and x(t) represents the extracted signal feature vector;

[0082] Assume that the energy distribution model of each category c is E c (f) Calculate the distance between the energy distribution of the input signal and each distribution model. The implementation formula is:

[0083] D(c)=∫|E in (f)-E c (f)| 2 df,

[0084] In the formula, D(c) represents the difference between the energy distribution of the input signal and the model of category c, df represents a small change in frequency f, and the final classification result c * for:

[0085] c * =argmin c D(c).

[0086] Among them, S2 performs channel response modeling H(t,τ) according to the classified signal, and the modeling implementation formula is:

[0087] H(t,τ)=∑ k a k (t)δ(τ-τ k (t)),

[0088] In the formula, H(t,τ) represents the dynamic channel response model, the channel characteristics at time t and delay τ, and a k (t) represents the amplitude of the kth path, the signal strength at time t, τ k (t) represents the delay of the kth path, the signal delay at time t;

[0089] According to the channel response model, let the received signal be r(t), the transmitted signal be s(t), and the compensated signal be It is expressed as:

[0090]

[0091] In the formula, represents the compensated signal, and h(t) represents H(t,τ) converted to the frequency domain representation:

[0092]

[0093] In the formula, F represents the conversion of time domain signal into frequency domain representation, a k (t) represents the amplitude of the kth path, the signal strength at time t, represents the phase shift factor in the frequency domain, e represents the base of the natural logarithm, j represents the imaginary unit, f represents the frequency variable, τ k (t) represents the delay of the kth path, the signal delay at time t, and the inverse filter h of h(t) is calculated. -1 (t), the realization formula is:

[0094]

[0095] In the formula, h -1 (t) represents the inverse filter of the channel response h(t), s(t) represents the transmitted signal, the transmitted signal at time t, g(t) represents the filter to be optimized, E represents the expected value operator, and r(t) represents the received signal, the actual received signal at time t.

[0096] Wherein, in S2, the environment perception data is assumed to be Es t , combined with the compensated signal To predict the multipath signal status, the implementation formula is:

[0097]

[0098] In the formula, X t+1 represents the multipath signal state prediction, represents the compensated signal, α represents the weighting factor, β(Es t ) indicates the possible multipath change trend predicted based on the current environmental characteristics;

[0099] Compensation is performed based on the predicted multipath status results, and the implementation formula is:

[0100] R i (t+1) = γP t +(1-γ)[R i (t)+λX t+1 ],

[0101] In the formula, R i (t+1) represents the demand for the i-th resource at time t+1, P t represents the predicted multipath strength at time t, γ represents the adjustment factor, and X t+1 Represents the multipath signal status prediction.

[0102] The pre-processed signals are identified and classified to distinguish between direct path signals and multipath signals. Through identification and classification, the multipath interference source can be accurately identified and corresponding measures can be taken to eliminate it. At the same time, the state of the multipath signal can be predicted and compensation can be performed based on the prediction results, further reducing the impact of the multipath effect on positioning accuracy.

[0103] Among them, S3, based on signal characteristics, combines multiple sensors to build a geographic environment model, and adjusts GPS parameters in real time according to environmental changes; obtains signal characteristics x(t), combines the data collected by the sensor in real time to identify building outlines and static obstacles, builds a three-dimensional geographic environment model based on the characteristics, updates the environment model in real time according to real-time environmental changes, and dynamically adjusts GPS receiver parameters according to the obstacle distribution in the environment model.

[0104] By updating the environmental model in real time and dynamically adjusting the parameters of the GPS receiver according to environmental changes, the system's adaptability in complex environments is improved, enabling the system to maintain stable positioning performance in different geographical environments. By constructing a geographical environment model, it can provide useful information for subsequent path planning and navigation, further improving the system's practicality and intelligence.

[0105] Wherein, S4, according to the signal recognition result and the prediction result, combines the intelligent antenna array to perform dynamic signal optimization and enhancement; according to the classification result c obtained in S2 * , signal response model H(t,τ) and multipath signal state prediction result X t+1 According to the initial environmental conditions, the initial parameters of the smart antenna array are set. According to the signal recognition results and the channel response model, the beam direction and shape of the antenna array are dynamically adjusted. By considering the influence of the Doppler effect, the frequency offset is corrected in real time to keep the signal locked. According to the signal strength and the predicted multipath signal status, the antenna is dynamically adjusted to optimize the signal and enhance the signal.

[0106] The intelligent antenna array can dynamically adjust the beam direction and shape of the antenna array according to the signal recognition results and channel response model to maximize the signal reception strength and reduce interference. At the same time, the system can also consider the influence of the Doppler effect, correct the frequency offset in real time, and maintain signal lock. By dynamically optimizing the enhanced signal, the system can more accurately receive and analyze satellite signals, thereby improving the accuracy and stability of positioning.

[0107] Among them, the S5 performs high-precision positioning based on the optimized and enhanced GPS signal and multi-sensor data fusion, receives the optimized and enhanced GPS signal and geographic environment model information, fuses the GPS signal and the geographic environment model information data, corrects the error of the initial position estimate according to the fused data, matches the fused position estimate with the geographic environment model, verifies whether the current position is in the expected area, and dynamically adjusts the fusion weights and filtering parameters according to real-time environmental changes.

[0108] By fusing GPS signals and geographic environment model information data, the system can correct errors in the initial position estimate and match and verify the fused position estimate with the geographic environment model. By fusing multi-source data, it can more comprehensively understand the current environment and accurately estimate the position information. By matching and verifying with the geographic environment model, it can further confirm the accuracy of the current position and eliminate potential positioning errors.

[0109] Among them, the GPS signal management system based on artificial intelligence includes:

[0110] Data acquisition preprocessing module: used to receive the original RF signals from multiple GPS satellites, and perform preliminary screening and noise filtering;

[0111] Intelligent multipath processing module: Based on the pre-processed signal information, it identifies and classifies different types of signals, eliminates multipath interference sources, and predicts the multipath signal status for resource scheduling;

[0112] Adaptive environmental perception adjustment module: It builds a geographic environment model by combining multi-sensor data and adjusts GPS parameters in real time according to environmental changes;

[0113] Dynamic signal optimization and enhancement module: Based on the signal recognition and prediction results, it combines the smart antenna array to perform dynamic signal optimization and enhancement;

[0114] High-precision positioning module: high-precision positioning is performed based on optimized and enhanced GPS signals and multi-sensor data fusion;

[0115] Real-time feedback module: used to monitor GPS signal quality and environmental changes in real time, automatically adjust various working parameters and provide feedback.

[0116] Working principle: The GPS receiver simultaneously receives radio frequency signals from multiple satellites, samples them at high speed and converts them into digital signals, then performs quantization processing to map the continuous analog signal values ​​to a limited number of discrete values, mixes the reference signal generated by the local oscillator with the received radio frequency signal, down-converts the radio frequency signal to the baseband, synchronizes the satellite signal frequency, and identifies and tracks the signal through code phase capture, evaluates and screens the signal-to-noise ratio of each signal, filters the screened signal for noise, and compresses, packages and transmits it;

[0117] By extracting features from the preprocessed signal, minimizing the mean error through adaptive weight vectors, updating weights and extracting signal feature vectors, converting the signal feature vectors to the frequency domain, calculating the energy distribution, and classifying them according to the distance between the energy distribution and each category model, channel response modeling is performed on the classified signal, and the received signal is compensated according to the channel response model to eliminate multipath interference and predict the multipath signal state. In combination with the data collected by the sensor in real time, the building outline and static obstacles are identified, and a three-dimensional geographic environment model is constructed based on these features. According to the changes in the real-time environment, the environmental model will be updated, and the parameters of the GPS receiver will be dynamically adjusted according to the obstacle distribution in the environmental model. According to the signal recognition results and the channel response model, the intelligent positioning will be dynamically adjusted. It can adjust the beam direction and shape of the antenna array to optimize signal reception. At the same time, it takes into account the influence of the Doppler effect, corrects the frequency offset in real time, maintains signal lock, and dynamically adjusts the antenna parameters according to the signal strength and predicted multipath signal status. It receives the optimized and enhanced GPS signal and geographic environment model information, fuses the information with the data of multiple sensors, corrects the error of the initial position estimate through the fused data, and matches the corrected position estimate with the geographic environment model to verify whether the current position is in the expected area. According to the real-time environmental changes, it dynamically adjusts the fusion weights and filtering parameters, monitors the GPS signal quality and environmental changes in real time, including signal strength, signal-to-noise ratio, and multipath interference, and automatically adjusts the various working parameters and provides feedback based on the monitoring results.

[0118] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

[0119] The present invention and its embodiments are described above, and such description is not restrictive. The drawings show only one embodiment of the present invention, and the actual structure is not limited thereto. In short, if ordinary technicians in the field are inspired by it, without departing from the purpose of the invention, they can design a structure and embodiment similar to the technical solution without creativity, which should belong to the protection scope of the present invention.

Claims

1. A GPS signal management method based on artificial intelligence, characterized in that: The following steps are involved: S1, receiving the original signals of multiple GPS satellites, and performing preliminary screening and noise filtering preprocessing; S2. According to the pre-processed signal information, identify and classify different types of signals and eliminate multipath interference sources, and predict the multipath signal state for compensation; S3, based on signal characteristics, combined with multiple sensors to build a geographic environment model, and adjust GPS parameters in real time according to environmental changes; S4. Based on the signal recognition and prediction results, dynamic signal optimization and enhancement are performed in combination with the smart antenna array; S5. Perform high-precision positioning based on optimized and enhanced GPS signals and multi-sensor data fusion; S6. Monitor GPS signal quality and environmental changes in real time, automatically adjust various working parameters and provide feedback.

2. The GPS signal management method based on artificial intelligence according to claim 1, characterized in that: The S1 receives the original signals of multiple GPS satellites, performs preliminary screening and noise filtering; simultaneously receives the radio frequency signals of multiple satellites through the GPS receiver, performs high-speed sampling on the received analog radio frequency signals and converts them into digital signals, maps the continuous analog radio frequency signal values ​​to a limited number of discrete values ​​for quantization, generates a reference signal matching the expected GPS carrier frequency through a local oscillator, mixes it with the received radio frequency signal, down-converts the radio frequency signal to baseband, synchronizes the satellite signal frequency, identifies and tracks the signal through code phase capture based on the quantized radio frequency signal, evaluates and screens the signal-to-noise ratio of each signal, performs noise filtering on the screened signal, and compresses, packages and transmits the screened and filtered signal.

3. The GPS signal management method based on artificial intelligence according to claim 1, characterized in that: The S2, according to the pre-processed signal information, identifies and classifies different types of signals and eliminates multipath interference sources, and predicts the multipath signal state for resource scheduling; assuming that the pre-processed signal y(t) includes signal components and noise, the pre-processed signal is feature extracted, and the implementation formula is: x(t)=Fq(y(t))-w(t)·n(t), In the formula, x(t) represents the extracted signal feature vector, y(t) represents the preprocessed signal, Fq represents the feature extraction function, w(t) represents the adaptive weight vector, and w(t) updates the adaptive weight vector by minimizing the mean error. The implementation formula is: w(t+1)=w(t)+μn(t)(y(t)-x(t)) T , In the formula, w(t+1) represents the updated weight, and μ represents the step size parameter; According to the extracted signal feature vector, the signal feature vector is classified, the frequency domain is converted, and the energy distribution is calculated. The implementation formula is: E in (f)=|F{x(t)}| 2 , In the formula, E in (f) represents the energy distribution of the signal, the signal energy intensity at frequency f, F represents the conversion of the time domain signal into the frequency domain, and x(t) represents the extracted signal feature vector; Assume that the energy distribution model of each category c is E c (f) Calculate the distance between the energy distribution of the input signal and each distribution model. The implementation formula is: D(c)=∫|E in (f)-E c (f)| 2 df, In the formula, D(c) represents the difference between the energy distribution of the input signal and the model of category c, df represents a small change in frequency f, and the final classification result c * for: c * =arg min c D(c)。 4. The GPS signal management method based on artificial intelligence according to claim 3, characterized in that: S2 performs channel response modeling H(t,τ) based on the classified signal, and the modeling implementation formula is: H(t,τ)=∑ k a k (t)δ(τ-τ k (t)), In the formula, H(t,τ) represents the dynamic channel response model, the channel characteristics at time t and delay τ, and a k (t) represents the amplitude of the kth path, the signal strength at time t, τ k (t) represents the delay of the kth path, the signal delay at time t; According to the channel response model, let the received signal be r(t), the transmitted signal be s(t), and the compensated signal be It is expressed as: In the formula, represents the compensated signal, and h(t) represents H(t,τ) converted to the frequency domain representation: In the formula, F represents the conversion of time domain signal into frequency domain representation, a k (t) represents the amplitude of the kth path, the signal strength at time t, represents the phase shift factor in the frequency domain, e represents the base of the natural logarithm, j represents the imaginary unit, f represents the frequency variable, τ k (t) represents the delay of the kth path, the signal delay at time t, and the inverse filter h of h(t) is calculated. -1 (t), the realization formula is: In the formula, h -1 (t) represents the inverse filter of the channel response h(t), s(t) represents the transmitted signal, the transmitted signal at time t, g(t) represents the filter to be optimized, E represents the expected value operator, and r(t) represents the received signal, the actual received signal at time t.

5. The GPS signal management method based on artificial intelligence according to claim 4, characterized in that: In S2, the environmental perception data is assumed to be Es t , combined with the compensated signal To predict the multipath signal status, the implementation formula is: In the formula, X t+1 represents the multipath signal state prediction, represents the compensated signal, α represents the weighting factor, β(Es t ) indicates the possible multipath change trend predicted based on the current environmental characteristics; Compensation is performed based on the predicted multipath status results, and the implementation formula is: R i (t+1)=γP t +(1-γ)[R i (t)+λX t+1 ], In the formula, R i (t+1) represents the demand for the i-th resource at time t+1, P t represents the predicted multipath strength at time t, γ represents the adjustment factor, and X t+1 Represents the multipath signal status prediction.

6. The GPS signal management method based on artificial intelligence according to claim 1, characterized in that: The S3 constructs a geographic environment model based on signal characteristics and multiple sensors, and adjusts GPS parameters in real time according to environmental changes; obtains signal characteristics x(t), identifies building outlines and static obstacles in combination with data collected by sensors in real time, constructs a three-dimensional geographic environment model based on the characteristics, updates the environment model in real time according to real-time environmental changes, and dynamically adjusts GPS receiver parameters according to the obstacle distribution in the environment model.

7. The GPS signal management method based on artificial intelligence according to claim 1, characterized in that: S4, based on the signal recognition result and the prediction result, combines the smart antenna array to perform dynamic signal optimization and enhancement; according to the classification result c obtained in S2 * , signal response model H(t,τ) and multipath signal state prediction result X t+1 According to the initial environmental conditions, the initial parameters of the smart antenna array are set. According to the signal recognition results and the channel response model, the beam direction and shape of the antenna array are dynamically adjusted. By considering the influence of the Doppler effect, the frequency offset is corrected in real time to keep the signal locked. According to the signal strength and the predicted multipath signal status, the antenna is dynamically adjusted to optimize the signal and enhance the signal.

8. The GPS signal management method based on artificial intelligence according to claim 1, characterized in that: The S5 performs high-precision positioning based on the optimized and enhanced GPS signal and multi-sensor data fusion, receives the optimized and enhanced signal and geographic environment model information, fuses the signal and geographic environment model information data, corrects the error of the initial position estimate based on the fused data, matches the fused position estimate with the geographic environment model, verifies whether the current position is in the expected area, and dynamically adjusts the fusion weights and filtering parameters according to real-time environmental changes.

9. The GPS signal management system based on artificial intelligence implemented according to the method of claim 1, comprising: Data acquisition preprocessing module: used to receive the original RF signals from multiple GPS satellites, and perform preliminary screening and noise filtering; Intelligent multipath processing module: Based on the pre-processed signal information, it identifies and classifies different types of signals, eliminates multipath interference sources, and predicts the multipath signal status for resource scheduling; Adaptive environmental perception adjustment module: It builds a geographic environment model by combining multi-sensor data and adjusts GPS parameters in real time according to environmental changes; Dynamic signal optimization and enhancement module: Based on the signal recognition and prediction results, it combines the smart antenna array to perform dynamic signal optimization and enhancement; High-precision positioning module: high-precision positioning is performed based on optimized and enhanced GPS signals and multi-sensor data fusion; Real-time feedback module: used to monitor GPS signal quality and environmental changes in real time, automatically adjust various working parameters and provide feedback.

Images