A Beidou anti-interference method based on adaptive filtering

Abstract

The purpose of this invention is to solve the problem of mutual constraint between convergence speed and steady-state error in existing minimum mean square adaptive filtering algorithms with fixed step sizes. This invention provides a BeiDou anti-interference method based on adaptive filtering, comprising the following steps: S101: Acquire BeiDou intermediate frequency digital signals and perform preprocessing; S102: Calculate the power spectrum of the input signal and determine the interference intensity; S103: Determine whether the interference intensity exceeds a preset threshold; S104: Initialize the adaptive filter weight vector and variable step size parameters; S105: Calculate the current filter output and error signal; S106: Update the step size factor based on an improved nonlinear function; S107: Update the filter tap weight vector using the new step size; S108: Implement signal filtering and output a purified signal; S109: Determine whether the interference is continuous; S110: Output the signal to the baseband correlator. By introducing a dynamically adjustable step size factor, a direct correlation between the step size factor and the square of the error signal energy is established, realizing adaptive real-time adjustment of the step size.

Description

Technical Field

[0001] This invention relates to the field of satellite navigation signal processing technology, and in particular to a BeiDou anti-interference method based on adaptive filtering. Background Technology

[0002] The navigation signals of the BeiDou Navigation Satellite System are transmitted from an altitude of approximately 20,000 kilometers to the ground receiver. During transmission, the signal energy undergoes significant attenuation, resulting in extremely low signal power reaching the receiver antenna, typically only around -160 dBW. This makes the signal highly susceptible to external electromagnetic interference. Interference can directly cause the receiver's pseudocode acquisition and tracking functions to fail, thereby reducing navigation and positioning accuracy. In severe cases, it can even lead to loss of positioning lock, affecting the normal operation of the BeiDou Navigation Satellite System and failing to meet the reliability requirements for navigation and positioning in various fields.

[0003] To address the aforementioned interference issues, the current anti-interference technology for BeiDou receivers primarily employs a fixed-step minimum mean square adaptive filtering algorithm, such as the technical solution disclosed in the invention patent application with publication number CN116134754A. This type of algorithm has been widely used in existing BeiDou receiver anti-interference scenarios.

[0004] However, existing minimum mean square adaptive filtering algorithms with fixed step sizes have an inherent contradiction: the mutual constraint between convergence speed and steady-state error. This contradiction makes it difficult to meet the needs of different application scenarios in terms of anti-interference performance. If a larger step size is set, although the algorithm can speed up the acquisition and suppression of interference signals and respond quickly to the occurrence of interference, it will significantly increase the filtering error in steady state, thereby directly affecting the navigation and positioning accuracy of the BeiDou receiver and failing to meet the requirements of high-precision navigation. If a smaller step size is set, although it can effectively reduce steady-state error and ensure the positioning accuracy of the BeiDou receiver, the algorithm's tracking ability is insufficient in scenarios with rapidly changing interference environments such as high dynamics. It cannot track the dynamic changes of interference signals in time, resulting in a significant decrease in anti-interference effect and difficulty in adapting to complex and ever-changing interference environments. Summary of the Invention

[0005] The purpose of this invention is to solve the problem of mutual constraint between convergence speed and steady-state error caused by the use of minimum mean square adaptive filtering algorithm with fixed step size in the existing technology. It provides a BeiDou anti-interference method based on adaptive filtering, which introduces a dynamically adjusted step size factor, which not only ensures the rapid anti-interference capability of BeiDou receiver in complex interference environment, but also maintains high navigation and positioning steady-state accuracy.

[0006] To solve the above problems, the present invention adopts the following technical solution:

[0007] A BeiDou anti-interference method based on adaptive filtering, characterized by the following steps:

[0008] S101: The receiver front end down-converts the acquired radio frequency signal and performs analog-to-digital conversion to obtain the BeiDou intermediate frequency digital signal, and then performs preprocessing.

[0009] S102: Use Fast Fourier Transform to perform frequency domain analysis on the preprocessed signal, calculate the power spectral density of the input signal, and determine the interference intensity;

[0010] S103: Compare the power spectral density with the preset background noise low level threshold; if it does not exceed the threshold, execute S110; if it exceeds the threshold, execute S104.

[0011] S104: Set the filter order, initialize the filter weight vector to a zero vector, and set the shape control parameters of the variable step size function. and amplitude control parameters ;

[0012] S105: Calculate the current filter output and error signal;

[0013] S106: Updating the step size factor based on an improved nonlinear function ;

[0014] S107: Based on the LMS criterion, utilize the updated step size factor. Iteratively update the filter tap weight vector;

[0015] S108: Pass the input signal through the updated adaptive filter to remove interference components and output a purified signal;

[0016] S109: Monitor the energy change of the error signal and determine whether the interference continues; if the interference continues, return to S105; if the interference disappears, execute S110.

[0017] S110: Output signal to baseband correlator.

[0018] Furthermore, the preprocessing of the BeiDou intermediate frequency digital signal in step S101 includes DC bias removal and automatic gain control regularization.

[0019] Furthermore, the automatic gain control regularization normalizes the signal amplitude to the optimal processing range of the filtering algorithm.

[0020] Furthermore, the interference intensity determination in step S102 includes identifying the interference type as narrowband interference, frequency sweep interference, or single-tone interference based on the distribution characteristics and energy value of the power spectral density.

[0021] Furthermore, the filtered output signal calculated in step S105 at the current time Represented as:

[0022]

[0023] in, Let be the conjugate transpose of the filter weight vector at time n. The input vector at time n;

[0024] Error signal Represented as:

[0025]

[0026] in, This is a reference signal.

[0027] Furthermore, the reference signal This is a delayed version of the BeiDou intermediate frequency digital signal.

[0028] Furthermore, the step size factor update formula for the improved nonlinear function in step S106 is as follows:

[0029]

[0030] in, The step size factor at time n; These are the shape control parameters for the variable step size function; For the amplitude control parameters of the variable step size function; Let n be the error signal at time n.

[0031] Furthermore, the iterative update formula for the filter tap weight vector in step S107 is as follows:

[0032]

[0033] in, Error signal The conjugate value; Let n be the filter weight vector at time n; The step size factor at time n; The filter input vector at time n; This is the filter weight vector at time n+1.

[0034] Furthermore, in step S104, the order of the filter ranges from 16 to 64.

[0035] Furthermore, the determination of whether the interference is continuous in step S109 is based on the following criteria: if the average energy of the error signal remains at a high level and fluctuates greatly, the interference is determined to be continuous; if the average energy of the error signal is stable and close to the preset thermal noise level of the Beidou receiver, the interference is determined to be eliminated.

[0036] The beneficial effects of this invention are as follows:

[0037] By introducing a nonlinear mapping mechanism based on an improved nonlinear function to dynamically adjust the step size factor, a direct correlation between the step size factor and the square of the error signal energy is established, realizing adaptive real-time adjustment of the step size. Specifically, when strong interference is detected and the error signal is large, the step size factor automatically and rapidly increases, enabling the adaptive filtering algorithm to capture and suppress the interference signal at a millisecond speed, achieving rapid convergence. When the interference is effectively suppressed and the error signal decreases and approaches steady state, the step size factor decreases smoothly, effectively reducing the algorithm's offset noise and significantly reducing steady-state filtering error. While ensuring millisecond-level rapid suppression of strong interference, the steady-state error is reduced, completely resolving the inherent contradiction between the convergence speed and steady-state error of existing fixed-step-size algorithms. This ensures both the rapid anti-interference capability of the BeiDou receiver in complex interference environments and maintains high navigation and positioning steady-state accuracy.

[0038] Setting the step size factor allows for rapid adjustment based on real-time changes in the error signal. The parameters of the filtering algorithm are dynamically updated iteratively as the characteristics of the interference signal change. For non-stationary interference such as frequency sweeping interference, pulse interference, and rapidly switching narrowband interference, the algorithm can quickly respond to changes in interference and adjust the filtering parameters in a timely manner to effectively suppress new interference frequencies. This avoids the problem of slow response to time-varying interference in traditional fixed-step-size algorithms. It significantly improves the adaptability of BeiDou receivers to complex dynamic electromagnetic environments, enabling them to maintain stable operation in strong and multi-interference scenarios such as battlefields and dense urban electromagnetic environments. This effectively prevents receiver lock-up and improves the reliability of BeiDou navigation system applications in complex scenarios. Attached Figure Description

[0039] Figure 1 This is a flowchart of the method in Embodiment 1 of the present invention. Detailed Implementation

[0040] The following specific examples illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that, unless otherwise specified, the following embodiments and features described therein can be combined with each other.

[0041] It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of the present invention. Therefore, the figures only show the components related to the present invention and are not drawn according to the actual number, shape and size of the components in the actual implementation. In the actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.

[0042] Example 1:

[0043] like Figure 1 As shown, a BeiDou anti-interference method based on adaptive filtering includes the following steps:

[0044] S101: The receiver front end down-converts the radio frequency signal and samples it through an analog-to-digital converter (ADC) to obtain the BeiDou intermediate frequency digital signal. The signal is then preprocessed, including DC bias removal and automatic gain control (AGC) regularization, to ensure that the signal amplitude is within the optimal processing range of the algorithm, laying the foundation for subsequent interference detection and filtering.

[0045] S102: The frequency domain characteristics of the preprocessed signal are obtained by using Fast Fourier Transform (FFT). By calculating the energy distribution in the current frequency band, the interference components in the signal are analyzed to achieve accurate determination of the interference intensity.

[0046] S103: Compare the calculated power spectral density with the preset background noise floor level to determine whether the interference intensity exceeds the preset threshold; if it does not exceed the threshold (J / N ratio is extremely low), it means that there is no obvious interference in the current environment, and no filtering is required, so proceed directly to S110; if obvious narrowband or single-tone interference is detected, it means that the current signal is greatly affected by interference, and anti-interference filtering is required, so proceed to the adaptive filtering process and execute S104.

[0047] S104: Initialize the adaptive filter weight vector and variable step size parameters:

[0048] First, set the filter order M. In this example, the order ranges from 16 to 64.

[0049] Initial weight vector Let it be the zero vector;

[0050] Set the control parameters of the variable step size function (Control function shape) and (Control function amplitude) to prepare parameters for subsequent filtering calculations and step size updates;

[0051] S105: Calculate the current filter output and error signal;

[0052] S106: Update the step size factor based on an improved nonlinear function;

[0053] S107: According to the LMS criterion, the filter tap weight vector is updated using a new step size; through this iterative update, the amplitude-frequency response of the filter will generate a very deep zero at the interference frequency, thereby achieving accurate filtering of the interference signal.

[0054] S108: The input signal is passed through the updated filter to perform signal filtering. Then, the amplitude-frequency characteristics of the filter are used to filter out the interference components in the signal, retain the BeiDou spread spectrum signal, and output the purified signal.

[0055] S109: Continuously monitor the energy change of the error signal to determine whether the interference continues; if the error is stable and close to the thermal noise level, it indicates that the interference may have disappeared, and you can return to S105 to continue monitoring or jump directly to S110; if the error signal does not reach the thermal noise level and there are fluctuations, it indicates that the interference is still continuing, and you should return to S105 to continue filtering iteration.

[0056] S110: Output signal to baseband correlator, specifically including sending the filtered and purified signal or the original pre-processed signal without obvious interference to the subsequent pseudocode acquisition and tracking module to complete the anti-interference processing of the signal.

[0057] In step S101, the process of acquiring and preprocessing the BeiDou intermediate frequency digital signal includes the following steps:

[0058] S1011: The front end of the Beidou receiver captures the radio frequency signal transmitted by the Beidou satellite through the radio frequency receiving module, and then downconverts the radio frequency signal to an intermediate frequency signal.

[0059] S1012: Discrete BeiDou intermediate frequency digital signals are obtained through analog-to-digital conversion (ADC) sampling, denoted as... ;

[0060] S1013: Preprocesses the acquired intermediate frequency digital signal. The preprocessing operations include DC bias removal and automatic gain control (AGC) regularization.

[0061] It should be noted that in step S1013, the DC bias removal process for the intermediate frequency digital signal specifically involves mean filtering of the digital signal to eliminate DC component interference. Automatic gain control regularization normalizes the signal amplitude to a preset range based on the algorithm's optimal signal amplitude range, ensuring the processing accuracy and stability of subsequent filtering algorithms and preventing algorithm saturation or failure due to excessively large or small signal amplitudes.

[0062] In step S102, the process of determining the interference intensity requires frequency domain analysis of the preprocessed BeiDou intermediate frequency digital signal. A 2048-point real-sequence Fast Fourier Transform (FFT) is used to convert the time-domain digital signal into a frequency-domain signal to obtain its frequency domain characteristics. Based on the frequency-domain signal, the periodogram method is used to calculate the power spectral density within the current BeiDou navigation signal's operating frequency band. , is represented as:

[0063]

[0064] in, The number of points in the FFT is 2048; This is the frequency domain signal after FFT transformation. By analyzing the power spectral density distribution characteristics and energy values, the intensity of electromagnetic interference currently affecting the signal is quantitatively determined, and the interference type is identified as narrowband interference, frequency sweep interference, single-tone interference, etc. In this example, interference with a power spectral density at a certain frequency point that is more than 10 dB higher than the background noise and a peak bandwidth ≤ 10 kHz is identified as narrowband interference; interference with a power spectral density at a single frequency point that is more than 20 dB higher than the background noise and a peak bandwidth ≤ 1 kHz is identified as single-tone interference; and interference with a power spectral peak that shifts linearly / nonlinearly with time at a rate ≤ 100 kHz / ms is identified as frequency sweep interference.

[0065] In step S103, it is necessary to preset the background noise floor level within the operating frequency band of the BeiDou receiver as the interference judgment threshold; background noise floor level Obtained through the thermal noise calculation formula, expressed as:

[0066]

[0067] in, This indicates the effective bandwidth of the signal, which is 2MHz in this example; The thermal noise figure is 6 dB in this example; therefore, the background noise floor level is calculated. The threshold is -105 dBm / Hz. The signal power spectral density calculated in step S102 is compared with this threshold; if the power spectral density... If the preset threshold is not exceeded, it means that there is no significant interference in the current electromagnetic environment, and no adaptive filtering is required. Proceed directly to step S110. If the power spectral density exceeds the preset threshold, it means that the current signal is subject to significant electromagnetic interference and anti-interference processing is required. Proceed to the subsequent adaptive filtering process.

[0068] In step S104, the order of the adaptive filter is set according to the hardware computing power and anti-interference requirements of the Beidou receiver. In this example, the filter order is chosen from 16th to 64th order. The order can be flexibly adjusted according to the actual interference scenario. For example, for single-tone interference or single narrowband interference, a 16th-order filter is selected to reduce hardware computational resource consumption; for the case of 2-3 superimposed narrowband interferences, a 32nd-order filter is selected to balance filtering effect and computation speed; for the case of frequency sweep interference combined with the superposition of multiple narrowband interferences, a 64th-order filter is selected to form a deeper filter zero at the interference frequency, improving the interference suppression effect. The initial weight vector of the adaptive filter is... Setting it to a zero vector ensures the filtering algorithm iterates from an initial unbiased state; simultaneously, two control parameters for the variable step size function are set, namely the shape control parameters. and amplitude control parameters , Used to control the shape of the Sigmoid function curve, determining the rate at which the step size factor changes with the error signal. Used to control the maximum range of values ​​for the step size factor, ensuring that the step size factor varies within a reasonable range.

[0069] In step S105, firstly, the preprocessed intermediate frequency digital signal is used as the input to the adaptive filter to construct the current time... input vector ;Calculate the filtered output signal at the current moment based on the operational relationship between the filter weight vector and the input vector. Among them, the filtered output is calculated. The formula is shown below:

[0070]

[0071] in, The input vector at time n, It is the conjugate transpose of the filter weight vector at time n.

[0072] Secondly, a delayed version of the input signal is selected as the reference signal. The delay time of the reference signal is set according to the filter order and sampling frequency. For example, a 32nd-order filter delays the signal by 16 sampling points to ensure the correlation between the reference signal and the BeiDou navigation signal; the error signal at the current moment is calculated. The formula is:

[0073]

[0074] in, The reference signal is used; the error signal reflects the deviation between the current filtering effect and the ideal state, indicating the current filtering algorithm's ability to suppress interference. The larger the error, the worse the interference suppression effect; the smaller the error, the more effectively the interference is suppressed.

[0075] In step S106, the step size factor is updated based on the improved nonlinear function, and the step size factor at the current moment is updated according to the real-time value of the error signal. The updated formula is:

[0076]

[0077] in, For shape control parameters, In this example, we take 500; For amplitude control parameters, In this example, we take 0.005; This is an error signal; This represents an exponential function with the natural constant e as its base. The formula combines the step size factor with the squared energy of the error signal. Related, when interference is strong, error signal When it is large, The value of increases significantly. Approaching 0, step size factor Rapidly increasing the value enables rapid convergence of the algorithm; when interference is effectively suppressed and error signals are minimized... When it decreases and approaches zero, Approaching 0, Approaching 1, step size factor The smooth reduction and approach to 0 effectively reduces the algorithm's offset noise and decreases the steady-state error.

[0078] In step S107, the updated step size factor from step S106 is used according to the LMS adaptive filtering criterion. The filter tap weight vector at the current moment is iteratively updated using the following formula:

[0079]

[0080] in, Error signal The conjugate value; Let n be the filter weight vector at time n; The step size factor at time n; The filter input vector at time n; Let be the filter weight vector at time n+1. Through this iterative update, the filter weight vector is gradually adjusted, and its amplitude-frequency response will form a very deep zero at the frequency of the interference signal, thereby effectively filtering out the interference signal while retaining the effective signal of BeiDou navigation.

[0081] In step S108, the BeiDou intermediate frequency digital signal preprocessed in step S101 is filtered by an adaptive filter whose tap weight vector has been updated in step S107. The filter's amplitude-frequency response characteristics are used to filter out interference components in the signal, retaining only the spread spectrum navigation signal of the BeiDou satellite, and outputting a purified digital signal to provide an interference-free signal source for subsequent baseband processing.

[0082] In step S109, the error signal from step S105 is continuously monitored. The energy change is calculated, the average energy of the error signal is smoothed, and if the average energy of the error signal remains at a high level and fluctuates greatly, it indicates that the interference still exists. The process returns to step S105 to continue filtering iteration and updates the filtering parameters in real time to achieve continuous interference suppression. If the average energy of the error signal is stable and close to the thermal noise level of the Beidou receiver, it indicates that the interference has disappeared and there is no need to continue filtering iteration. The process jumps to step S110.

[0083] In step S110, the purified signal from step S108 or the original signal deemed to have no significant interference after preprocessing in step S101 is directly output to the baseband correlator of the BeiDou receiver and sent to the subsequent pseudocode acquisition, pseudocode tracking and carrier tracking modules to complete the demodulation of the BeiDou navigation signal and navigation positioning calculation.

[0084] The above description is merely a specific example of the present invention and does not constitute any limitation on the present invention. Obviously, those skilled in the art, after understanding the content and principles of the present invention, may make various modifications and changes in form and detail without departing from the principles and structure of the present invention; however, these modifications and changes based on the spirit of the present invention are still within the scope of protection of the claims of the present invention.

Claims

1. A BeiDou anti-interference method based on adaptive filtering, characterized in that, Includes the following steps: S101: The receiver front end down-converts the acquired radio frequency signal and performs analog-to-digital conversion to obtain the BeiDou intermediate frequency digital signal, and then performs preprocessing. S102: Use Fast Fourier Transform to perform frequency domain analysis on the preprocessed signal, calculate the power spectral density of the input signal, and determine the interference intensity; S103: Compare the power spectral density with the preset background noise low level threshold; If the threshold is not exceeded, execute S110; if the threshold is exceeded, execute S104. S104: Set the filter order, initialize the filter weight vector to a zero vector, and set the shape control parameters of the variable step size function. and amplitude control parameters ; S105: Calculate the current filter output and error signal; S106: Updating the step size factor based on an improved nonlinear function ; S107: Based on the LMS criterion, utilize the updated step size factor. Iteratively update the filter tap weight vector; S108: Pass the input signal through the updated adaptive filter to remove interference components and output a purified signal; S109: Monitor the energy change of the error signal and determine whether the interference continues; if the interference continues, return to S105; if the interference disappears, execute S110. S110: Output signal to baseband correlator.

2. The BeiDou anti-interference method based on adaptive filtering according to claim 1, characterized in that, The preprocessing of the BeiDou intermediate frequency digital signal in step S101 includes DC bias removal and automatic gain control regularization.

3. The BeiDou anti-interference method based on adaptive filtering according to claim 2, characterized in that, The automatic gain control regularization normalizes the signal amplitude to the optimal processing range of the filtering algorithm.

4. The BeiDou anti-interference method based on adaptive filtering according to claim 1, characterized in that, The interference intensity determination in step S102 includes identifying the interference type as narrowband interference, frequency sweep interference, or single-tone interference based on the distribution characteristics and energy value of the power spectral density.

5. The BeiDou anti-interference method based on adaptive filtering according to claim 1, characterized in that, The current time filtered output signal calculated in step S105 Represented as: in, Let be the conjugate transpose of the filter weight vector at time n. The input vector at time n; Error signal Represented as: in, This is a reference signal.

6. The BeiDou anti-interference method based on adaptive filtering according to claim 5, characterized in that, The reference signal This is a delayed version of the BeiDou intermediate frequency digital signal.

7. The BeiDou anti-interference method based on adaptive filtering according to claim 1, characterized in that, The step size factor update formula for the improved nonlinear function in step S106 is as follows: in, The step size factor at time n; These are the shape control parameters for the variable step size function; For the amplitude control parameters of the variable step size function; Let n be the error signal at time n.

8. The BeiDou anti-interference method based on adaptive filtering according to claim 7, characterized in that, The iterative update formula for the filter tap weight vector in step S107 is as follows: in, Error signal The conjugate value; Let n be the filter weight vector at time n; The step size factor at time n; The filter input vector at time n; This is the filter weight vector at time n+1.

9. The BeiDou anti-interference method based on adaptive filtering according to claim 1, characterized in that, In step S104, the filter order ranges from 16 to 64.

10. The BeiDou anti-interference method based on adaptive filtering according to claim 1, characterized in that, The criteria for determining whether the interference is continuous in step S109 are as follows: if the average energy of the error signal remains at a high level and fluctuates greatly, the interference is determined to be continuous; if the average energy of the error signal is stable and close to the preset thermal noise level of the Beidou receiver, the interference is determined to be eliminated.

Images