A super-resolution fine acquisition method based on fast FRFT

By using a fast FRFT algorithm in GNSS signal processing, the Doppler rate and frequency are estimated by matching the amplitude peak of the signal, the problem of inaccurate Doppler frequency shift estimation in the prior art is solved, and the accuracy of GNSS TOA estimation is improved and reliability reduces computational complexity and enhances the performance of GNSS receivers in harsh environments.

CN118444346BActive Publication Date: 2024-12-03SHENZHEN TECH UNIV
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
CN202410548445.7
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-05-06
Publication Date
2024-12-03
Estimated Expiration
2044-05-06

AI Technical Summary

Technical Problem

When facing signal dynamics, existing GNSS signal processing methods are difficult to accurately estimate Doppler shifts and frequency, resulting in reduced accuracy and reliability of TOA estimation of GNSS arrival time, especially in multipath channels and dynamic environments.

Method used

Using a super-resolution algorithm based on fast FRFT, the operation is performed by FRFT at different rotation angles, matching the amplitude peak of the signal to estimate the Doppler rate and frequency, and using the generalized Fourier transform form of FRFT to reduce the computational complexity and improve GNSS Accuracy and reliability of TOA estimation.

Benefits of technology

The range and resolution of GNSS Doppler shift estimation are improved, the computing efficiency of signal processing is enhanced, the performance of GNSS receivers in harsh environments is improved, and the accuracy and real-time performance of TOA estimation is ensured.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118444346B_ABST
    Figure CN118444346B_ABST
Patent Text Reader

Abstract

The present invention belongs to the field of global navigation satellite system (GNSS) signal processing technology, and is a super-resolution fine acquisition method based on fast FRFT, including initial parameter setting, signal preprocessing; signal processing; simulation analysis; target detection performance verification; signal parameter estimation; FFT capture and judgment; dynamic parameter compensation: compensating the relevant results through estimated dynamic parameters; FFT capture: capturing the compensated results through the FFT algorithm; judging the acquisition success: if the maximum value is greater than the threshold, it means that the acquisition is successful; if it is less than the threshold, the local PRN replica is adjusted through the synchronization control module, and the above steps are repeated until the peak value exceeds the threshold. The present invention realizes the improvement of the range, resolution and calculation efficiency of Doppler frequency shift estimation, and improves the accuracy, reliability and real-time performance of GNSS arrival time TOA estimation.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the technical field of global navigation satellite system (GNSS) signal processing, and in particular relates to a super-resolution fine acquisition method based on fast FRFT. Background Art

[0002] The Global Navigation Satellite System (GNSS) can provide high-precision, reliable navigation and positioning services to users around the world, promoting the development and popularization of global navigation technology. Due to the development of the current international situation and the demand for civilian positioning, it is increasingly valued by people.

[0003] In the prior art, previously proposed methods for processing global navigation satellite system (GNSS) signals include long coherent integration algorithm and super resolution algorithm (SRA), which improve the performance of global navigation satellite system (GNSS) arrival time (TOA) estimation in multipath channels. However, these methods have defects:

[0004] Among them, the long coherent integration algorithm can improve the estimation accuracy, but it may cause frequency error accumulation and thus reduce the gain when facing signal dynamics. The super-resolution algorithm performs well in multipath suppression, but in actual situations, dynamic factors may cause frequency error accumulation and gain reduction.

[0005] In addition, the algorithm combining the partially matched filter (PMF) and the fast Fourier transform (FFT) improves the range and resolution of the Doppler shift estimation, but cannot compensate for the Doppler rate. Meanwhile, the FFT-based discriminator design can effectively estimate the residual of the Doppler frequency in a quasi-static environment, but cannot handle even a small Doppler rate caused by acceleration. Summary of the invention

[0006] The purpose of the embodiment of the present invention is to provide a super-resolution fine acquisition method based on fast FRFT, which can improve the range, resolution and calculation efficiency of Doppler frequency shift estimation, and improve the accuracy, reliability and real-time performance of GNSS arrival time TOA estimation.

[0007] The embodiment of the present invention is implemented as follows:

[0008] A super-resolution fine acquisition method based on fast FRFT, comprising:

[0009] 101. Perform initial parameter settings and pre-process the signal;

[0010] The signal preprocessing process is:

[0011]

[0012]

[0013]

[0014] Where p is the FRFT order associated with the Doppler velocity estimation in the matched filtering process, and m p is the index of the FRFT order matching filter in the FRFT domain, m u is the index of the frequency matched filter in the FRFT domain, DF p {·} represents the operator of digital FRFT implementation; χ(n S ) is represented by the slow time correlation samples formed with the input signal and the local replica, n s is the index of the signal sample in the slow time domain in CPI processing, TS is the slow time interval; and are the local replica signals of the code and carrier signals associated with the LOS path, respectively; and are respectively the local replica initial carrier phase, carrier Doppler frequency and TOA estimation value of the LOS signal; R cc [·] is the operator of the autocorrelation function of the two spreading code sequences, with the time delay error Δτ l and change;

[0015] 102. Signal processing;

[0016] Signal reception: Down-convert the GNSS signal and perform analog-to-digital conversion (ADC), and sample the received signal into a discrete signal;

[0017] Local carrier stripping: The discrete signal is stripped by the local carrier NCO and multiplied with the local PRN replica;

[0018] Correlation calculation: After integration and dumping, the cross-correlation result is obtained as the LFM signal, i.e. linear frequency modulation signal;

[0019] Noise suppression: Implement the cumulative mean spectrum technique to suppress noise, and obtain the result r by Q times of cumulative averaging;

[0020] Doppler shift estimation: A rough estimate of the Doppler shift is obtained by frequency multiplication and multiplied by r Compensate for r;

[0021] Interpolation improvement: If discrete effects are found in the FRFT spectrum, linear interpolation is used to improve the L1 and L2 estimates;

[0022] 103. Simulation analysis;

[0023] Multipath channel simulation: Simulate according to the simulation parameter settings, including considering the two signal paths of LOS signal and NLOS signal, as well as TOA delay error;

[0024] Signal power analysis: Analyze the simulation results to observe the signal power overlap of LOS and NLOS signals in FRFT, as well as the signal power changes under different TOA delay errors;

[0025] 104. Target detection performance verification;

[0026] LOS and NLOS signal simulation: Verify the target detection performance of FRFT SRA for LOS and NLOS signals based on simulation results, especially when the TOA delay error is 7-16 meters;

[0027] Observation of the influence of Doppler rate: observe the influence of different input Doppler rate values ​​of LOS and NLOS signals on the detection process; different input Doppler rate values ​​of LOS and NLOS signals will have obvious peaks in the FRFT spectrum, which can distinguish LOS and NLOS signals;

[0028] 105. Signal parameter estimation;

[0029] Assuming two arbitrary angles θ1 and θ2, for a receiver moving on the ground in actual situations, β-θ1>0, θ2-β>0, then the following relationship can be obtained:

[0030] L1=L W sin(β-θ1) (3-30)

[0031] L2=L W sin(α2-θ) (3-31)

[0032] Where L1 and L2 are the normalized FRFT lengths at rotation angles α1 and α2, respectively. W , divide formula (3-31) by formula (3-30) to obtain:

[0033]

[0034] Arranging the derivation, we can get:

[0035]

[0036] Assuming there is an optimal rotation angle α0 and any two rotation angles α1 and α2, such that α0-α1>0, α2-α0>0, it can be deduced that:

[0037]

[0038] From formula (3-34), it can be seen that the estimated value of the optimal rotation angle α0 is only related to the rotation angles α1, α2 and the corresponding projection width; Defined as the optimal rotation angle The FRFT spectrum amplitude below, the u-axis coordinate of its output peak is:

[0039]

[0040] therefore, and It can be expressed as:

[0041]

[0042]

[0043] It can be seen that only three FRFT calculations are required to obtain the optimal rotation angle and position of the LFM signal in the fractional Fourier domain;

[0044] Signal parameter estimation: Based on the analysis of simulation results, the signal parameters related to the carrier frequency are estimated using formulas;

[0045] Rotation angle calculation: According to the formula Get the best rotation angle;

[0046] Doppler frequency shift and velocity estimation: According to the formula formula and Estimate Doppler frequency shift and Doppler velocity. The Doppler frequency deviation is accurately estimated as

[0047] 106. FFT capture and judgment;

[0048] Dynamic parameter compensation: The relevant results are compensated by the estimated dynamic parameters;

[0049] FFT capture: Capture the compensated results through FFT algorithm;

[0050] Determine whether the acquisition is successful: if the maximum value is greater than the threshold, it means that the acquisition is successful; if it is less than the threshold, the local PRN replica is adjusted through the synchronization control module, and the above steps are repeated until the peak value exceeds the threshold.

[0051] The embodiment of the present invention improves the accuracy and reliability of GNSS TOA estimation by using a super-resolution algorithm based on FRFT to estimate the Doppler rate and frequency of the LOS signal by matching the amplitude peak of the signal, effectively utilizes the generalized Fourier transform form of FRFT, maps the signal in the time domain to the two-dimensional space of the Fourier domain and the fractional Fourier domain, so that gains such as the carrier Doppler frequency, the carrier Doppler rate and the signal-to-noise ratio can be considered when matching the signal, thereby improving the performance of the GNSS receiver in harsh environments; the method of the present invention introduces a fast FRFT algorithm to perform FRFT operations at three different rotation angles, thereby reducing the time for searching for the optimal rotation angle within the FRFT order range, thereby avoiding unnecessary calculation steps, greatly reducing the calculation complexity, and making the operation faster. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 It is a schematic diagram of a multipath satellite signal processing method based on super-resolution of the present invention;

[0053] Figure 2 It is a flow chart of the fast FRFT algorithm implementation of the present invention. DETAILED DESCRIPTION

[0054] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0055] The present invention proposes a FRFT-based fine capture method for GNSS signals in multipath channels. The method uses amplitude peak matching to estimate the Doppler rate and frequency of the signal, and introduces an efficient fast Fourier transform algorithm (FRFT) or fractional Fourier transform to improve the range, resolution and calculation efficiency of Doppler frequency shift estimation, and improve the accuracy, reliability and real-time performance of GNSS TOA estimation.

[0056] In conjunction with the specific embodiments, see Figure 1 and Figure 2 , the specific implementation of the present invention is described in detail:

[0057] 1. Initial parameter setting: According to the proposed FRFT SRA multipath channel estimation simulation parameter setting, including signal type, RF frequency, code frequency, PRN number, etc., and FRFT SRA processing parameter setting, such as CPI, slow time interval, etc. The signal preprocessing process is

[0058]

[0059]

[0060]

[0061] Where p is the FRFT order associated with the Doppler velocity estimation in the matched filtering process, and m p is the index of the FRFT order matching filter in the FRFT domain, m u is the index of the frequency matched filter in the FRFT domain, DF p {·} represents the operator of digital FRFT implementation; χ(n S ) is represented by the slow time correlation samples formed with the input signal and the local replica, n s is the index of the signal sample in the slow time domain in CPI processing, TS is the slow time interval; and are the local replica signals of the code and carrier signals associated with the LOS path, respectively; and are respectively the local replica initial carrier phase, carrier Doppler frequency and TOA estimation value of the LOS signal; R cc [·] is the operator of the autocorrelation function of the two spreading code sequences, with the time delay error Δτ l And changes.

[0062] 2. Signal processing:

[0063] a. Signal reception: Down-convert the GNSS signal and perform analog-to-digital conversion (ADC), and sample the received signal into a discrete signal.

[0064] b. Local carrier stripping: The discrete signal is stripped by the local carrier NCO and multiplied with the local PRN replica.

[0065] c. Correlation calculation: After integration and dumping, the cross-correlation result is obtained as the LFM signal, i.e., linear frequency modulation signal;

[0066] d. Noise suppression: Implement the cumulative mean spectrum technique to suppress noise, and obtain the result r through Q times of cumulative averaging.

[0067] e. Doppler shift estimation: A rough estimate of the Doppler shift is obtained by frequency multiplication and multiplied by r Compensate for r.

[0068] f. Interpolation improvement: If discrete effects are found in the FRFT spectrum, linear interpolation is used to improve the estimates of L1 and L2.

[0069] 3. Simulation analysis:

[0070] a. Multipath channel simulation: Perform simulation according to the simulation parameter settings, including considering the two signal paths of LOS signal and NLOS signal, as well as TOA delay error, etc.

[0071] b. Signal power analysis: Based on the simulation results, we analyze and observe the signal power overlap of LOS and NLOS signals in FRFT, as well as the signal power changes under different TOA delay errors.

[0072] 4. Target detection performance verification:

[0073] a. LOS and NLOS signal simulation: Verify the target detection performance of FRFT SRA for LOS and NLOS signals based on simulation results, especially when the TOA delay error is small (generally 7-16 meters). Without any enhancement measures, the TOA estimation error of civil GPS receivers may range from a few meters to tens of meters. For example, after the US government canceled the SA (Selective Availability) policy, the published civil GPS positioning accuracy error was about 7.8 meters (horizontally) and 15.2 meters (vertically) at a 95% confidence level.

[0074] b. Observation of the influence of Doppler rate: Observe the influence of different input Doppler rate values ​​of LOS and NLOS signals on the detection process. Different input Doppler rate values ​​of LOS and NLOS signals will have obvious peaks in the FRFT spectrum, which can distinguish LOS and NLOS signals.

[0075] 5. Signal parameter estimation:

[0076] Assuming two arbitrary angles θ1 and θ2, for a receiver moving on the ground in actual situations, β-θ1>0, θ2-β>0, then the following relationship can be obtained:

[0077] L1=L W sin(β-θ1) (3-30)

[0078] L2=L W sin(α2-θ) (3-31)

[0079] Where L1 and L2 are the normalized FRFT lengths at rotation angles α1 and α2, respectively. W , divide formula (3-31) by formula (3-30) to obtain:

[0080]

[0081] Arranging the derivation, we can get:

[0082]

[0083] Assuming there is an optimal rotation angle α0 and any two rotation angles α1 and α2, such that α0-α1>0, α2-α0>0, it can be deduced that:

[0084]

[0085] From formula (3-34), it can be seen that the estimated value of the optimal rotation angle α0 is only related to the rotation angles α1, α2 and the corresponding projection width; Defined as the optimal rotation angle The FRFT spectrum amplitude below, the u-axis coordinate of its output peak is:

[0086]

[0087] therefore, and It can be expressed as:

[0088]

[0089]

[0090] It can be seen that the optimal rotation angle and position of the LFM signal in the fractional Fourier domain can be obtained by performing only three FRFT calculations, instead of the traditional two-dimensional search.

[0091] a. Signal parameter estimation: Based on the analysis of simulation results, the signal parameters related to the carrier frequency are estimated using formulas.

[0092] b. Rotation angle calculation: According to the formula Get the best rotation angle.

[0093] c. Doppler frequency shift and velocity estimation: According to the formula formula and Estimate Doppler frequency shift and Doppler velocity. The Doppler frequency deviation is accurately estimated as

[0094] 6. FFT capture and judgment:

[0095] a. Dynamic parameter compensation: The relevant results are compensated by the estimated dynamic parameters.

[0096] b. FFT capture: The compensated results are captured through the FFT algorithm.

[0097] c. Determine whether the acquisition is successful: If the maximum value is greater than the threshold, it indicates that the acquisition is successful; if it is less than the threshold, the local PRN replica is adjusted through the synchronization control module, and the above steps are repeated until the peak value exceeds the threshold.

[0098] In summary, the embodiments of the present invention improve the accuracy and reliability of GNSS TOA estimation by utilizing a super-resolution algorithm based on FRFT to estimate the Doppler rate and frequency of the LOS signal by matching the amplitude peak of the signal, and effectively utilize the generalized Fourier transform form of FRFT to map the signal in the time domain to the two-dimensional space of the Fourier domain and the fractional Fourier domain, so that gains such as carrier Doppler frequency, carrier Doppler rate and signal-to-noise ratio can be considered when matching the signal, thereby improving the performance of the GNSS receiver in harsh environments; the method of the present invention introduces a fast FRFT algorithm to perform FRFT operations at three different rotation angles, thereby reducing the time for searching for the optimal rotation angle within the FRFT order range, thereby avoiding unnecessary calculation steps, greatly reducing the calculation complexity, and making the operation faster.

[0099] Related Names:

[0100] Digitally controlled oscillator (NCO)

[0101] Pseudo Random Noise (PRN)

[0102] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A super-resolution fine acquisition method based on fast FRFT, It is characterized in that include:

101. Perform initial parameter settings and pre-process the signal; The signal preprocessing process is: Where p is the FRFT order associated with the Doppler velocity estimation in the matched filtering process, and m p is the index of the FRFT order matching filter in the FRFT domain, m u is the index of the frequency matched filter in the FRFT domain, DF p {·} represents the operator of digital FRFT implementation; χ(n S ) is represented by the slow time correlation samples formed with the input signal and the local replica, n s is the index of the signal sample in the slow time domain in CPI processing, TS is the slow time interval; and are the local replica signals of the code and carrier signals associated with the LOS path, respectively; and are respectively the local replica initial carrier phase, carrier Doppler frequency and TOA estimation value of the LOS signal; R cc [·] is the operator of the autocorrelation function of the two spreading code sequences, with the time delay error Δτ l and change; 102. Signal processing; Signal reception: Down-convert the GNSS signal and perform analog-to-digital conversion (ADC), and sample the received signal into a discrete signal; Local carrier stripping: The discrete signal is stripped by the local carrier NCO and multiplied with the local PRN replica; Correlation calculation: After integration and dumping, the cross-correlation result is obtained as the LFM signal, i.e. linear frequency modulation signal; Noise suppression: Implement the cumulative mean spectrum technique to suppress noise, and obtain the result r by Q times of cumulative averaging; Doppler shift estimation: A rough estimate of the Doppler shift is obtained by frequency multiplication and multiplied by r Compensate for r; Interpolation improvement: If discrete effects are found in the FRFT spectrum, linear interpolation is used to improve the L1 and L2 estimates; 103. Simulation analysis; Multipath channel simulation: Simulate according to the simulation parameter settings, including considering the two signal paths of LOS signal and NLOS signal, as well as TOA delay error; Signal power analysis: Analyze the simulation results to observe the signal power overlap of LOS and NLOS signals in FRFT, as well as the signal power changes under different TOA delay errors; 104. Target detection performance verification; LOS and NLOS signal simulation: Verify the target detection performance of FRFT SRA for LOS and NLOS signals based on simulation results, especially when the TOA delay error is 7-16 meters; Observation of the influence of Doppler rate: observe the influence of different input Doppler rate values ​​of LOS and NLOS signals on the detection process; different input Doppler rate values ​​of LOS and NLOS signals will have obvious peaks in the FRFT spectrum, which can distinguish LOS and NLOS signals; 105. Signal parameter estimation; Assume two arbitrary angles θ 1 and θ 2 , for a receiver moving on the ground in actual situations, there is β-θ 1 >0,θ 2 -β>0, then we can get the following relationship: L 1 =L W sin(β-θ 1 ) (3-30) L 2 =L W sin(a 2 -i) (3-31) Among them, L 1 and L 2 They are the rotation angle α 1 and α 2 Normalized FRFT length under L W , divide formula (3-31) by formula (3-30) to obtain: Arranging the derivation, we can get: Assume that there is an optimal rotation angle α 0 and any two rotation angles α 1 and α 2 , so that α 0 -α 1 >0,α 2 -α 0 >0, we can deduce that: From formula (3-34), we can see that the optimal rotation angle α 0 The estimated value of is only related to the rotation angle α 1 , α 2 And the corresponding projection width; Defined as the optimal rotation angle The FRFT spectrum amplitude below, the u-axis coordinate of its output peak is: therefore, and It can be expressed as: It can be seen that only three FRFT calculations are required to obtain the optimal rotation angle and position of the LFM signal in the fractional Fourier domain; Signal parameter estimation: Based on the analysis of simulation results, the signal parameters related to the carrier frequency are estimated using formulas; Rotation angle calculation: According to the formula Get the best rotation angle; Doppler frequency shift and velocity estimation: According to the formula formula and Estimate Doppler frequency shift and Doppler velocity. The Doppler frequency deviation is accurately estimated as 106. FFT capture and judgment; Dynamic parameter compensation: The relevant results are compensated by the estimated dynamic parameters; FFT capture: Capture the compensated results through FFT algorithm; Determine whether the acquisition is successful: if the maximum value is greater than the threshold, it means that the acquisition is successful; if it is less than the threshold, the local PRN replica is adjusted through the synchronization control module, and the above steps are repeated until the peak value exceeds the threshold.

Citation Information

Patent Citations

  • PMF-FRFT-based BOC signal capturing algorithm under highly dynamic condition

    CN109375242A

  • High-efficiency navigation signal correlation peak fine sampling calculation method

    CN115575985A

  • Process for determining the impulse response of a band-limited high resolution radar channel in an economically favourable way

    EP0802427A2

  • Methods and Devices for Global Navigation Satellite System (GNSS) Signal Acquisition

    US20230258818A1