Fast open-loop GNSS-R (Global Navigation Satellite System-Radio) height measurement method and system based on space-time frequency combination

Abstract

The invention relates to a fast open-loop GNSS-R height measurement method and system based on space-time frequency combination, and the method comprises the steps: collecting multi-frequency GNSS signals, carrying out the coarse time delay and coarse Doppler frequency offset estimation of each multi-frequency GNSS signal, and carrying out the compensation of the multi-frequency GNSS signals through coarse Doppler frequency offset; based on the coarse time delay and the compensated multi-frequency GNSS signal, carrying out forming processing on a wave beam in a time domain direction, and constructing a space-time-frequency joint vector; through a spatial domain smoothing technology, rank increasing is carried out on the autocorrelation matrix of the space-time-frequency joint vector, and the problem of rank deficiency of the autocorrelation matrix is solved; performing feature analysis on the autocorrelation matrix according to a subspace decomposition method, and estimating direct view path time delay, specular reflection path time delay and a direct view path incoming angle; and deducing a satellite elevation angle according to the direct-view path incoming angle, and inverting the height from the receiver to the ground surface through the delay inequality of the direct-view path and the mirror reflection path and the satellite elevation angle. According to the technical scheme of the invention, high-precision effective estimation of the height of the receiver can be realized quickly with low complexity.

Description

Technical Field

[0001] This application relates to the field of unmanned aerial vehicle (UAV) control technology, and in particular to a fast open-loop GNSS-R altimetry method and system based on space-time-frequency joint measurement. Background Technology

[0002] Global Navigation Satellite System Reflection Signal (GNSS-R) altimetry technology utilizes the characteristic difference between the line-of-sight path of satellite navigation signals and the reflection path of the Earth's surface to passively measure the altitude of a receiver. It offers advantages such as all-weather operation, wide coverage, and no need for active signal transmission, making it valuable for applications in low-altitude UAV flight, surface mapping, and marine monitoring. Particularly in UAV control scenarios, GNSS-R altimetry provides UAVs with interference-resistant and highly concealed altitude information, ensuring flight safety and mission effectiveness in complex environments.

[0003] In existing technologies, GNSS-R altimetry methods mostly rely on closed-loop tracking circuits or iterative optimization algorithms based on single-frequency signals. Altitude inversion is achieved by continuously tracking the time delay difference between the line-of-sight path and the reflection path. However, multipath interference is often not effectively separated during signal processing, and there is a lack of targeted solutions for the rank deficiency problem of the autocorrelation matrix. Furthermore, in the above-mentioned technical solutions, the limited bandwidth of the single-frequency signal leads to low time delay estimation resolution, and the superposition of multipath signals can easily introduce significant altimetry errors. Closed-loop tracking or iterative optimization requires multiple iterations of calculation, resulting in high computational complexity and processing time often reaching the minute level, which is difficult to meet the real-time requirements of dynamic scenarios such as UAVs. At the same time, the rank deficiency of the autocorrelation matrix reduces the accuracy of parameter estimation, and traditional methods require complex positioning calculations to obtain altitude information, further increasing processing delay and failing to achieve fast and high-precision altimetry.

[0004] It should be noted that the information disclosed in the background section above is only used to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0005] To address the aforementioned issues, this application provides a fast open-loop GNSS-R altimetry method and system based on space-time-frequency joint measurement, which can quickly and with low complexity achieve high-precision and effective estimation of receiver altitude.

[0006] To achieve the objectives of this application, the following technical solution is provided:

[0007] Firstly, this application provides a fast open-loop GNSS-R altimetry method based on space-time-frequency joint measurement, including:

[0008] Multi-frequency GNSS signals are acquired, and coarse time delay and coarse Doppler frequency offset are estimated for each multi-frequency GNSS signal. The multi-frequency GNSS signals are then compensated using the coarse Doppler frequency offset.

[0009] Based on the coarse time delay and the compensated multi-frequency GNSS signal, beamforming processing is performed on the time domain direction to construct a joint space-time-frequency vector;

[0010] The rank deficiency problem of the autocorrelation matrix is ​​solved by increasing the rank of the joint space-time-frequency vector through spatial smoothing techniques.

[0011] The autocorrelation matrix is ​​subjected to feature analysis based on the subspace decomposition method, and the direct line delay, specular reflection delay and direct line angle of arrival are estimated.

[0012] The satellite elevation angle is derived from the line-of-sight angle, and the height of the receiver above the ground is obtained by using the time delay difference between the line-of-sight angle and the mirror reflection angle and the satellite elevation angle.

[0013] Secondly, this application provides a fast open-loop GNSS-R altimetry system based on space-time-frequency joint measurement, the system being used to execute the aforementioned fast open-loop GNSS-R altimetry method based on space-time-frequency joint measurement, the system comprising:

[0014] The signal acquisition module acquires multi-frequency GNSS signals, performs coarse time delay and coarse Doppler frequency offset estimation on each multi-frequency GNSS signal, and compensates for the multi-frequency GNSS signal using the coarse Doppler frequency offset.

[0015] The vector construction module, based on the coarse time delay and the compensated multi-frequency GNSS signal, performs beamforming processing on the time domain direction to construct a joint space-time-frequency vector;

[0016] The matrix rank-increasing module increases the rank of the autocorrelation matrix of the space-time-frequency joint vector through spatial smoothing technology, thereby solving the rank deficiency problem of the autocorrelation matrix.

[0017] The feature analysis module performs feature analysis on the autocorrelation matrix according to the subspace decomposition method, and estimates the direct line delay, the specular reflection delay and the direct line angle of arrival.

[0018] The altitude inversion module derives the satellite elevation angle based on the line-of-sight angle and inverts the receiver's altitude above the Earth's surface using the time delay difference between the line-of-sight angle and the mirror reflection angle.

[0019] The technical solution provided in this application may include the following beneficial effects:

[0020] The fast open-loop GNSS-R altimetry method and system based on spatiotemporal-frequency joint processing provided in this application can improve the time delay estimation resolution by fusing the bandwidth expansion effect of multi-frequency GNSS signals. Furthermore, through multi-dimensional spatiotemporal-frequency joint processing, it effectively separates the line-of-sight path, specular reflection path, and multipath signals, eliminating the influence of multipath and achieving high-precision altimetry. Simultaneously, by replacing traditional iterative optimization or closed-loop tracking methods with subspace decomposition, only one feature decomposition and low-dimensional search are required, significantly reducing computational complexity and shortening processing time to meet the real-time requirements of dynamic scenarios. In addition, by combining multi-antenna estimation of the azimuth angle with the open-loop altimetry mechanism, time-consuming positioning calculations are eliminated, and altitude is directly inverted through geometric relationships, reducing processing time from minutes to milliseconds, achieving fast real-time altimetry.

[0021] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description

[0022] The accompanying drawings are provided to further understand this application and form part of the specification. They are used together with the embodiments of the invention to explain this application and do not constitute a limitation thereof. Obviously, the drawings described below are merely some embodiments of this disclosure, and those skilled in the art can obtain other drawings based on these drawings without any creative effort.

[0023] Figure 1 A flowchart illustrating a fast open-loop GN SS-R altimetry method based on space-time-frequency joint measurement provided in this application embodiment;

[0024] Figure 2 A flowchart illustrating step S100 of a fast open-loop GN SS-R altimetry method based on space-time-frequency joint measurement provided in an embodiment of this application;

[0025] Figure 3 A flowchart illustrating step S200 of a fast open-loop GNSS-R altimetry method based on space-time-frequency joint measurement provided in an embodiment of this application;

[0026] Figure 4 A flowchart illustrating step S300 of a fast open-loop GNSS-R altimetry method based on space-time-frequency joint measurement provided in an embodiment of this application;

[0027] Figure 5 A flowchart illustrating step S400 of a fast open-loop GNSS-R altimetry method based on space-time-frequency joint measurement provided in an embodiment of this application;

[0028] Figure 6A flowchart illustrating step S500 of a fast open-loop GNSS-R altimetry method based on space-time-frequency joint measurement provided in an embodiment of this application;

[0029] Figure 7 A comparison of the autocorrelation function of BeiDou multi-frequency signals and traditional single-frequency signals in a fast open-loop GN SS-R altimetry method based on space-time-frequency joint analysis provided in this application embodiment;

[0030] Figure 8 A detailed flowchart of a fast open-loop GN SS-R altimetry method based on space-time-frequency joint measurement is provided for embodiments of this application;

[0031] Figure 9 A schematic diagram of GNSS space-time-frequency joint vector formation for a fast open-loop GNSS-R altimetry method based on space-time-frequency joint measurement provided in this application embodiment;

[0032] Figure 10 A schematic diagram of the geometric delay model for a fast open-loop GNSS-R altimetry method based on space-time-frequency joint measurement, provided for an embodiment of this application;

[0033] Figure 11 A schematic diagram of the direction finding principle model of a fast multi-antenna receiver for a fast open-loop GNSS-R altimetry method based on space-time-frequency joint method provided in this application embodiment;

[0034] Figure 12 The pseudorange estimation accuracy of the traditional tracking loop and subspace method for a fast open-loop GNSS-R altimetry method based on spatiotemporal joint measurement provided in this application embodiment reaches 0.01T. c Simulation diagram of required data length;

[0035] Figure 13 Simulation diagram of pseudorange accuracy of single-frequency GNSS signal and multi-frequency GNSS signal in multipath environment based on subspace method of a fast open-loop GNSS-R altimetry method based on space-time-frequency joint method provided in the embodiments of this application;

[0036] Figure 14 This is a schematic diagram of a fast open-loop GNSS-R altimetry system based on space-time-frequency joint measurement, provided as an embodiment of this application. Detailed Implementation

[0037] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, they are provided so that this disclosure will be more comprehensive and complete, and will fully convey the concept of the exemplary embodiments to those skilled in the art. The described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.

[0038] This example implementation first provides a fast open-loop GNSS-R altimetry method based on space-time-frequency joint analysis, applied to a GNSS receiving platform equipped with a multi-antenna, multi-frequency receiver. This platform's receiver can simultaneously receive multiple frequency GNSS signals, such as the B1I and B1C signals at 1561.098MHz and 1575.42MHz respectively in the BeiDou system. Receiving multiple frequency GNSS signals can effectively expand the signal's frequency domain bandwidth, thereby improving the accuracy of time delay estimation. A comparison of the main peak of its autocorrelation function with that of the C / A code in a typical GPS system shows that… Figure 7 As shown, the C / A code of the GPS system has a bandwidth of 2.046MHz, and its autocorrelation function main lobe width is approximately ±1T. c The combined bandwidth of the BeiDou multi-frequency signals is 14.322MHz, and the main lobe width of its autocorrelation function is only about ±0.07T. c Meanwhile, the platform receiver is also equipped with a multi-antenna receiving device. The arrangement of the multi-antenna structure is not limited, but the height measurement method proposed in this patent only requires a uniform linear array. The spatial steering vector for the first GNSS signal is a(·).

[0039] refer to Figure 1 As shown, the fast open-loop GNSS-R altimetry method based on space-time-frequency joint measurement may include the following steps:

[0040] Step S100: Acquire multi-frequency GNSS signals, perform coarse time delay and coarse Doppler frequency offset estimation on each multi-frequency GNSS signal, and compensate the multi-frequency GNSS signal using the coarse Doppler frequency offset.

[0041] Step S200: Based on the coarse time delay and the compensated multi-frequency GNSS signal, the beam in the time domain is shaped to construct a joint space-time-frequency vector.

[0042] Step S300: The autocorrelation matrix of the space-time-frequency joint vector is increased in rank using spatial smoothing techniques to solve the rank deficiency problem of the autocorrelation matrix.

[0043] Step S400: Perform feature analysis on the autocorrelation matrix according to the subspace decomposition method, and estimate the direct line delay, specular reflection delay and direct line angle of arrival.

[0044] Step S500: Derive the satellite elevation angle based on the line-of-sight angle, and invert the receiver's height above the ground using the time delay difference between the line-of-sight angle and the mirror reflection angle.

[0045] In the above technical solution, firstly, the multi-antenna multi-frequency receiver performs short-time data acquisition and performs simple GNSS data acquisition and coarse time delay estimation for each GNSS signal; secondly, beamforming processing in the time domain direction is performed based on the estimated coarse time delay, extracting the spreading gain while reducing the signal processing pressure in the time domain direction, and then forming a space-time-frequency joint vector; thirdly, spatial smoothing is achieved based on the spatial angle separation characteristics to alleviate the rank deficiency phenomenon of coherent signal separation based on the subspace method; then, the angle and time delay of the line-of-sight path and the specular reflection path are separated and estimated based on the subspace method; finally, based on the estimated time delay difference between the line-of-sight path and the specular reflection path and the signal elevation angle estimation, the pseudorange estimation of the specular reflection path relative to the line-of-sight path and the satellite elevation angle are obtained, and the receiver's altitude measurement estimation above the ground is carried out.

[0046] Below, we will refer to Figures 2 to 6 The steps of the above-described fast open-loop GNSS-R altimetry method based on space-time-frequency joint measurement in this example embodiment will be described in more detail.

[0047] In step S100, multi-frequency GNSS signals are acquired, and coarse time delay and coarse Doppler frequency offset are estimated for each multi-frequency GNSS signal. The multi-frequency GNSS signals are then compensated using the coarse Doppler frequency offset.

[0048] It should be noted that the reference Figure 8 As shown, the detailed technical solution of this application is as follows: First, the multi-antenna multi-frequency receiver performs short-time data acquisition to obtain the left and right component signals of the multi-frequency GNSS signal, and each component includes the line-of-sight path and the reflection path. According to the general GNSS receiving and processing process, the coarse time delay estimate and coarse Doppler frequency offset estimate of the multi-frequency GNSS signal can be directly estimated. The Doppler frequency offset can be used to compensate and simplify the data, and finally obtain the sampled data in each code period. The signal model is established based on the assumption that the propagation delays of the left and right components of the multi-frequency GNSS signal are basically consistent, meaning that the frequency difference between the left and right components will not be too large. This ensures that the navigation satellite signal will not generate excessive ionospheric delay differences when passing through the ionosphere, guaranteeing the consistency of the transmission paths of the left and right components. Furthermore, it is necessary to first estimate the coarse delay and coarse Doppler frequency offset of the received GNSS signal. Since there are multiple paths, including the line-of-sight path, the specular reflection path, and multipath paths, the estimated coarse delay and Doppler frequency offset are composite values ​​that deviate from the true values ​​and only serve as indicators. Subsequent processing also takes into account the impact of this estimation deviation.

[0049] In one possible implementation, such as Figure 2 As shown, step S100 may further include the following sub-steps:

[0050] In step S110, multi-frequency GNSS signals are acquired by a multi-antenna multi-frequency receiver. The multi-frequency GNSS signals include line-of-sight, specular reflection path, and multipath.

[0051] It should be noted that multiple antennas must be uniform linear / area arrays; multiple frequency bands must include at least two civilian frequency points, such as the BeiDou B1I and B1C frequency bands; the collected signals include the direct line of sight + specular reflection path (strong) + diffuse reflection path (weak).

[0052] In step S120, the coarse time delay and coarse Doppler frequency offset of the multi-frequency GNSS signal are estimated according to the GNSS receiving processing procedure.

[0053] It should be noted that the coarse time delay / Doppler is estimated by the sliding correlation method. Due to the superposition of multiple paths, the result is a composite value of multiple paths (time delay error of several meters and frequency offset error of tens of Hz), which requires a two-dimensional search (time delay-Doppler plane) to solve.

[0054] In step S130, the multi-frequency GNSS signal is compensated based on the coarse Doppler frequency offset.

[0055] It should be noted that Doppler compensation is achieved through general GNSS carrier loop tracking. The compensation only corrects the carrier, and code phase alignment depends on coarse time delay.

[0056] Furthermore, step S100 is illustrated by example:

[0057] Set a total of N s The nth satellite is positioned above the altimeter receiver and can provide stable and effective satellite navigation signals. s The multi-frequency signal from the satellite should be a composite signal of the line-of-sight and the specular reflection path.

[0058]

[0059] Where k = 1, ..., K, K = 2 (k = 1 represents the direct line of sight, k = 2 represents the specular reflection line of sight, and k = 3, ..., K represents the multipath), it represents the total number of direct and specular reflection lines. In the above, the subscripts "-" and "+" represent the left and right halves of the multi-frequency signal, respectively. The spatial steering vector varies with the two components of the multi-frequency signal, denoted as a... - (·) and a + (·). Furthermore, γ k , τ k and vk These represent the complex amplitude, angle of arrival, time delay, and Doppler frequency offset of the k-th path, respectively. - (t) and c + (t) represents the waveforms of two components of the satellite's multi-frequency signal, which are respectively modulated to a frequency offset f relative to the center frequency. s and -f s Up. n - (t) and n + (t) represents Gaussian white noise that is uncorrelated with the signal in both the spatial and temporal domains.

[0060] Note that, based on the received GNSS signal, the path difference between the line-of-sight path and the specular reflection path is small, and the time delay difference and Doppler frequency offset difference are also small. Based on the general GNSS receiver processing procedure, the coarse time delay estimate of this GNSS signal can be directly obtained. Coarse Doppler frequency offset estimation Furthermore, a phase-locked loop (PLL) ensures the real-time performance of the Doppler frequency offset estimation. Based on the above Doppler frequency offset estimates, the synthesized signal of the collected direct line-of-sight path and the specular reflection path can be simplified as follows:

[0061]

[0062] Currently, the left and right components are configured to contain L in each code period. - and L + Each sampling point was obtained

[0063]

[0064] Among them, T s This represents the sampling interval.

[0065] In step S200, based on the coarse time delay and the compensated multi-frequency GNSS signal, the beamforming process in the time domain direction is performed to construct a joint space-time-frequency vector.

[0066] It should be noted that, in order to reduce the signal processing complexity caused by too many signal sampling points in the time domain, the time-domain beamforming matrix B of the left and right components of the multi-frequency signal is formed based on the coarse time delay estimated in step one. - and B + This allows for the extraction of spread spectrum gain. Then, based on the correlation characteristics of the multi-frequency signals, a joint matrix Y is formed in the three dimensions of space, time, and frequency, facilitating subsequent joint signal processing.

[0067] In one possible implementation, such as Figure 3 As shown, step S200 may further include the following sub-steps:

[0068] In step S210, time-domain beamforming matrices for the left and right components of the multi-frequency signal are constructed based on coarse delay, and time-domain beamforming processing is performed on the compensated multi-frequency GNSS signal.

[0069] It should be noted that the delay set takes a coarse delay range of ±1-2 chips, uniformly samples to construct a matrix, focuses the main lobe of the signal to extract the spreading gain, and compresses the amount of time-domain data.

[0070] In step S220, the multi-frequency GNSS signal that has undergone time-domain beamforming is subjected to column vectorization to obtain the spatiotemporal joint vector of the left and right components of the multi-frequency signal.

[0071] It should be noted that column vectorization integrates the spatial array and temporal sampling point features into a one-dimensional vector, preserving the spatiotemporal correlation and providing a unified format for subsequent matrix operations.

[0072] In step S230, the space-time joint vectors of the left and right components of the multi-frequency signal are merged to form a space-time-frequency joint vector.

[0073] The time-domain beamforming matrix is ​​constructed based on a coarse time delay and a uniformly distributed set of time delays in its vicinity, and is used to extract the spread spectrum gain and reduce the complexity of time-domain signal processing.

[0074] Furthermore, the spatiotemporal joint vector is:

[0075]

[0076] Among them, y s (n) is the joint space-time-frequency vector, y s,- (n) is the spatiotemporal joint vector of the left component of the multi-frequency signal, y s,+ (n) is the space-time joint vector of the right component of the multi-frequency signal, and I2 is a 2×2 identity matrix. For spatial array manifold matrix, Let ρ be the time correlation matrix. - ρ is the compensation coefficient for the complex amplitude difference of the left component. + Φ is the compensation coefficient for the complex amplitude difference of the right component. - (τ) is the phase factor of the left subcarrier of the multi-frequency signal, Φ + (τ) is the right component subcarrier phase factor of the multi-frequency signal, γ s For the complex envelope of the satellite signal, n s It is an additive white Gaussian noise vector.

[0077] It should be noted that the identity matrix distinguishes the left and right components of a multi-frequency GNSS signal; the array manifold matrix characterizes the spatial phase response; the time correlation matrix reflects the characteristics of the spreading code; and the phase factor reflects the time delay phase caused by the subcarriers of the multi-frequency GNSS signal.

[0078] Furthermore, step S200 is illustrated by example:

[0079] To obtain spreading gain while reducing computational complexity, a beamforming matrix in the time domain is applied to each of the two components of the multi-frequency signal within each code period. and The time-domain beamforming matrix is ​​based on coarse time delay estimation. The value is used to select a uniformly distributed set of time delays in its vicinity. We obtain, and satisfy M τ <<min{L - ,L + The GNSS signal after time-domain beamforming is vectorized by column to obtain the spatiotemporal joint vector representation, i.e.:

[0080]

[0081] in, and The mth τ The kth and kth elements can be respectively derived from... and This indicates that the two functions take values ​​near the main peak of their autocorrelation function, which can achieve... Matrix manifolds can be represented as and Because typical multi-frequency signals generally have relatively close frequency bands, therefore... Complex amplitude γ s,- and γ s,+ There is a linear relationship between them, and they share the same sparse distribution characteristics. Furthermore, there is... and By merging the left and right parts, we can obtain the spatiotemporal-frequency multidimensional joint vector expression.

[0082]

[0083] Wherein, the coefficient ρ - and ρ + This is used to compensate for the complex amplitude difference between the left and right components, and the compensation coefficient is assumed to be known during this process. The spatiotemporal-frequency vector combines feature extensions from the spatial, temporal, and frequency domains, and their formation relationship can be described as follows: Figure 9 As shown. By combining multiple code periods, we can obtain...

[0084]

[0085] Where, Λ=[γ s1 ,...,γ sN ] and N = [n s1,...,n sN ].

[0086] In step S300, the autocorrelation matrix of the space-time-frequency joint vector is increased in rank using spatial smoothing techniques to resolve the rank deficiency problem of the autocorrelation matrix.

[0087] It should be noted that before performing joint space-time-frequency parameter estimation, due to the multipath effect in the GNSS-R altimetry environment, parameter estimation using the subspace method may suffer from rank deficiency in the autocorrelation matrix. Therefore, spatial smoothing is performed in the spatial dimension to construct the corresponding smoothing matrix. and To achieve the rank-increasing effect of the signal autocorrelation matrix.

[0088] In one possible implementation, such as Figure 4 As shown, step S300 may further include the following sub-steps:

[0089] In step S310, a sub-matrix is ​​extracted from the spatio-temporal-frequency joint vector to obtain the spatio-temporal-frequency sub-matrix.

[0090] It should be noted that submatrix extraction involves extracting consecutive rows of data, as described above. M τ OK, M is the number of spatially smooth subarrays, which is less than the number of array elements. τ The number of sampling points after time-domain beamforming is used to retain the main spatiotemporal characteristics of the signal and remove redundant components dominated by edge noise.

[0091] In step S320, a spatial forward smoothing vector is constructed based on the space-time-frequency submatrix. The spatial forward smoothing vector includes a spatial matrix and a fine time delay difference matrix.

[0092] It should be noted that the spatial matrix corresponds to the spatial distribution of antenna array elements and the direction distribution of satellite signals, the fine delay difference matrix describes the sub-chip level delay phase difference brought about by the subcarriers of multi-frequency GNSS signals, and the smoothing vector is used to reduce signal coherence.

[0093] In step S330, a forward smoothing matrix is ​​constructed using the spatial forward smoothing vector, and the autocorrelation matrix of the space-time-frequency joint vector is increased in rank using the forward smoothing matrix.

[0094] It should be noted that the forward smoothing matrix improves the rank of the autocorrelation matrix by superimposing submatrix information, restores the signal dimension, and provides a full-rank matrix basis for subspace decomposition.

[0095] Furthermore, step S300 is illustrated by example:

[0096] Rank-increasing of the signal autocorrelation matrix based on spatial smoothing. Taking the first row to the... Row data is merged to form a space-time-frequency submatrix:

[0097]

[0098] Based on the above space-time frequency submatrix, the p-th spatial forward smoothing vector of the left and right components of the multi-frequency signal is formed:

[0099]

[0100] It contains spatial matrix With fine delay difference matrix Then, construct the forward smoothing matrices respectively:

[0101]

[0102] In step S400, the autocorrelation matrix is ​​subjected to feature analysis according to the subspace decomposition method, and the direct-view path delay, specular reflection path delay, and direct-view path angle of arrival are estimated.

[0103] It should be noted that, based on the existing DOA matrix method, a spatiotemporal-frequency joint matrix R is constructed, and the eigenvalues ​​of the aforementioned spatiotemporal-frequency matrix are decomposed. Fractional time delay estimation is then performed using K large eigenvalues. Subsequently, the arrival angle minus an integer multiple of the delay is calculated using K large eigenvectors. A two-dimensional search is performed; finally, an accurate time delay estimate is synthesized.

[0104] In one possible implementation, such as Figure 5 As shown, step S400 may further include the following sub-steps:

[0105] In step S410, a spatiotemporal joint matrix is ​​constructed based on the subspace decomposition method, and the autocorrelation matrix is ​​eigenvalued by the spatiotemporal joint matrix to obtain a large eigenvector.

[0106] It should be noted that the spatiotemporal-frequency joint matrix fuses spatiotemporal-frequency features, and the large eigenvectors after eigenvalue decomposition correspond to the signal components.

[0107] In step S420, the fractional delay is estimated based on the large feature vector, and a two-dimensional search network is constructed with the direct line of sight angle and integer multiple delay. The direct line of sight angle and integer multiple delay are estimated through the peak position.

[0108] It should be noted that the fractional delay is at the sub-chip level, with precision less than the chip period. The two-dimensional search utilizes spatiotemporal correlation, and the peak position corresponds to the true angle of arrival and an integer multiple of the delay.

[0109] In step S430, the fractional delay and the integer delay are combined to form a precise delay; the precise delay includes the direct line-of-sight delay and the specular reflection delay.

[0110] Furthermore, the precise time delay is:

[0111]

[0112] in, For precise time delay, The delay is a fraction of a factor. The time delay is an integer multiple of λ, where λ is the wavelength of the multi-frequency GNSS signal.

[0113] It should be noted that after synthesis, the blurring of the entire code chip is eliminated, and the precise time delay accuracy reaches the sub-code chip level, corresponding to the actual propagation delay of the direct line of sight and the specular reflection path, respectively.

[0114] Furthermore, step S400 is illustrated by example:

[0115] Subspace decomposition enables joint estimation of time delay and angle. Based on the DOAMatrix method, the matrix is ​​constructed as follows:

[0116] R = R +,- R -,- ;

[0117] Perform eigenvalue decomposition on the above matrix and obtain:

[0118]

[0119] After eigenvalue decomposition using the above formula, a fractional time delay estimate is obtained based on the K largest eigenvalues ​​obtained from the decomposition:

[0120]

[0121] Where k = 1, 2. Subsequently, an angle-integer multiple delay network is constructed. Compensate for the fractional delay estimates obtained above, and perform a simple two-dimensional search based on the decomposed feature vectors to give the peak position with both angle and integer multiple delay estimates:

[0122]

[0123] in, Then it means The kth column, This represents the k-th column of the eigenvector obtained from the decomposition, where k = 1, 2. Finally, the estimated fractional time delay is used... With integer multiples of delay Synthetic delay estimation:

[0124]

[0125] In step S500, the satellite elevation angle is derived from the line-of-sight angle, and the height of the receiver above the ground is obtained by using the time delay difference between the line-of-sight and the mirror reflection path and the satellite elevation angle.

[0126] It should be noted that the rapid open-loop height measurement mechanism derived from the analysis utilizes the angle of arrival. and satellite elevation angle The complementary relationship is used to calculate the incident elevation angle reached by the line of sight. And using the estimated line-of-sight time delay And the time delay estimation of the mirror reflection path The inversion receiver is at a height h above the ground.

[0127] In one possible implementation, such as Figure 6 As shown, step S500 may further include the following sub-steps:

[0128] In step S510, the time delay difference is calculated based on the direct viewing path time delay and the specular reflection path time delay.

[0129] It should be noted that the time delay difference needs to eliminate weak multipath interference such as diffuse reflection, and only retain the difference between the specular reflection path and the direct viewing path to reflect the difference in their path lengths.

[0130] In step S520, the satellite elevation angle is obtained by performing a complementary calculation on the direct line of sight angle according to the first formula.

[0131] Furthermore, the first formula is:

[0132]

[0133] in, For the satellite elevation angle, The direct viewing angle is the directional angle.

[0134] It should be noted that the complementary relationship is based on geometric symmetry, that is, the angle of arrival is the angle between the signal and the horizontal plane, and the satellite elevation angle is the angle between the signal and the vertical direction, which is applicable to near-Earth scenarios.

[0135] In step S530, the height of the receiver above the ground is calculated using the time delay difference between the direct line of sight and the specular reflection path and the satellite elevation angle, according to the second formula.

[0136] Furthermore, the second formula is:

[0137]

[0138] Where h is the height of the receiver above the Earth's surface, and c is the speed of light. For the time delay of the mirror reflection path, The line-of-sight time delay is θ1, and the satellite elevation angle is θ1.

[0139] Furthermore, step S500 is illustrated by example:

[0140] Altitude inversion of the receiver by fusing time delay difference and elevation angle. In principle, altimetry based on GNS SR primarily relies on measuring the relative path code phase between the direct line-of-sight signal and the specular reflection signal. These two signals exhibit a relationship such as... Figure 10 The geometric relationship.

[0141] When the receiver is near the Earth's surface, the satellite-to-ground link and the two paths of the reflection point and the satellite can be considered approximately parallel, with their elevation angles being approximately the same. Therefore, based on the satellite elevation angle θ1 and the total path delay of the specular reflected signal relative to the direct signal, the height of the receiver above the Earth's surface can be obtained:

[0142]

[0143] By deploying multiple antennas for a GNSS receiver, it is possible to provide measurements of the signal's angle of arrival and determine the target's position relative to the receiving antennas. Taking a uniform linear array receiver as an example... Figure 11 As shown, there is a one-to-one correspondence between the angle of arrival and the satellite elevation angle. Geometrically, they exist in the same plane perpendicular to the x0y plane. Therefore, theoretically, the angle of arrival of the direct satellite signal and the satellite elevation angle are complementary.

[0144]

[0145] In other words, the satellite elevation angle can be obtained directly from the direction-of-arrival angle estimation using multi-antenna data from the radio frequency front end.

[0146] Furthermore, a comparison is made between traditional tracking loops and subspace methods, such as... Figure 12 As shown, the time-delay tracking loop of its traditional tracking loop is a second-order loop with a loop bandwidth of 10Hz. Both the traditional tracking loop and the subspace-based method use a coherent integration time of 100ms. Figure 12 The comparison of the two methods shown indicates that, with the same pseudorange accuracy, the data length used by the traditional tracking loop is 10 to 100 times that used by the subspace method, proving the feasibility of the fast height measurement method in the open-loop case of this application.

[0147] Specifically, Figure 13Simulation results of pseudorange accuracy of single-frequency GNSS signals and multi-frequency GNSS signals in a multipath environment were compared using a subspace-based method. The single-frequency GNSS signal used was the C / A code of the GPS system, while the multi-frequency GNSS signal considered the combined B1I and B1C signals from the BeiDou system. Also using a subspace-based method, with a coherence integration time of 100ms, a multipath was added in addition to the line-of-sight path and the specular reflection path. The amplitude difference between the line-of-sight path and the multipath was 4dB, and the spatial angle difference was 30°. Figure 13 The comparison of the two signals shown demonstrates that the space-time-frequency joint method based on subspace decomposition can effectively solve the multipath effect. Due to the bandwidth expansion, the multi-frequency GNSS signal can achieve decimeter-level pseudorange estimation accuracy in multipath environments, proving the effectiveness of the altimetry scheme in this application for multipath suppression.

[0148] Furthermore, in this example implementation, such as Figure 14 As shown, a fast open-loop GNSS-R altimetry system based on space-time-frequency joint measurement is also provided. This system is used to execute the aforementioned fast open-loop GNSS-R altimetry method based on space-time-frequency joint measurement. The system includes:

[0149] The signal acquisition module acquires multi-frequency GNSS signals, performs coarse time delay and coarse Doppler frequency offset estimation on each multi-frequency GNSS signal, and compensates for the multi-frequency GNSS signal using the coarse Doppler frequency offset.

[0150] The vector construction module, based on the coarse time delay and the compensated multi-frequency GNSS signal, performs beamforming processing on the time domain direction to construct a joint space-time-frequency vector;

[0151] The matrix rank-increasing module increases the rank of the autocorrelation matrix of the space-time-frequency joint vector through spatial smoothing technology, thereby solving the rank deficiency problem of the autocorrelation matrix.

[0152] The feature analysis module performs feature analysis on the autocorrelation matrix according to the subspace decomposition method, and estimates the direct line delay, the specular reflection delay and the direct line angle of arrival.

[0153] The altitude inversion module derives the satellite elevation angle based on the line-of-sight angle and inverts the receiver's altitude above the Earth's surface using the time delay difference between the line-of-sight angle and the mirror reflection angle.

[0154] In the description of this specification, references to terms such as "one possible implementation," "further," "exemplary," "specific example," or "optional," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. The illustrative expressions of the above terms in this specification do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. In addition, those skilled in the art can combine and integrate the different embodiments or examples described in this specification.

[0155] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the invention filed herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not claimed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this application are indicated by the appended claims.

Claims

1. A fast open-loop GNSS-R altimetry method based on space-time-frequency joint, characterized in that, The method comprises the following steps: Collecting multi-frequency GNSS signals, estimating coarse time delay and coarse Doppler frequency offset for each of the multi-frequency GNSS signals, and compensating the multi-frequency GNSS signals based on the coarse Doppler frequency offset; Based on the coarse time delay and the compensated multi-frequency GNSS signals, beam shaping processing is performed in the time domain direction to construct a space-time-frequency joint vector; Through a space domain smoothing technique, the rank of the autocorrelation matrix of the space-time-frequency joint vector is increased to solve the rank deficiency problem of the autocorrelation matrix; According to the subspace decomposition method, the autocorrelation matrix is analyzed, and the direct path time delay, the specular reflection path time delay and the direct path angle of arrival are estimated; According to the direct path angle of arrival, the satellite elevation angle is derived, and the height of the receiver from the ground surface is inversely calculated based on the time delay difference between the direct path and the specular reflection path and the satellite elevation angle.

2. The fast open-loop GNSS-R altimetry method based on space-time-frequency joint according to claim 1, characterized in that, The step of collecting multi-frequency GNSS signals, estimating coarse time delay and coarse Doppler frequency offset for each of the multi-frequency GNSS signals, and compensating the multi-frequency GNSS signals based on the coarse Doppler frequency offset comprises: Collecting multi-frequency GNSS signals by a multi-antenna multi-frequency receiver, wherein the multi-frequency GNSS signals include direct paths, specular reflection paths and multi-paths; Estimating the coarse time delay and the coarse Doppler frequency offset of the multi-frequency GNSS signals according to a GNSS receiving and processing flow; Compensating the multi-frequency GNSS signals based on the coarse Doppler frequency offset.

3. The fast open-loop GNSS-R altimetry method based on space-time-frequency joint according to claim 1, characterized in that, The step of based on the coarse time delay and the compensated multi-frequency GNSS signals, beam shaping processing is performed in the time domain direction to construct a space-time-frequency joint vector comprises: Based on the coarse time delay, time domain beam shaping matrices of left and right components of the multi-frequency signals are constructed respectively to perform time domain beam shaping processing on the compensated multi-frequency GNSS signals; Performing column vectorization operation on the multi-frequency GNSS signals subjected to time domain beam shaping to obtain space-time joint vectors of left and right components of the multi-frequency signals; Combining the space-time joint vectors of left and right components of the multi-frequency signals to form a space-time-frequency joint vector.

4. The fast open-loop GNSS-R altimetry method based on space-time-frequency joint according to claim 3, characterized in that, The space-time-frequency joint vector is: where y s (n) is the spatially and temporally joint vector, y s,- (n) is the spatially and temporally joint vector of the left component of the multi-frequency signal, y s,+ (n) is the spatially and temporally joint vector of the right component of the multi-frequency signal, I2 is a 2x2 identity matrix, is the spatial array manifold matrix, is the time correlation matrix, p - is the left component complex amplitude diversity compensation coefficient, p + is the right component complex amplitude diversity compensation coefficient, F - (τ) is the sub-carrier phase factor of the left component of the multi-frequency signal, F + (τ) is the sub-carrier phase factor of the right component of the multi-frequency signal, g s is the complex envelope of the satellite signal, n s is the additive white Gaussian noise vector.

5. The fast open-loop GNSS-R altimetry method based on space-time-frequency joint according to claim 1, characterized in that, The step of through a space domain smoothing technique, the rank of the autocorrelation matrix of the space-time-frequency joint vector is increased to solve the rank deficiency problem of the autocorrelation matrix comprises: Extracting a submatrix from the space-time-frequency joint vector to obtain a space-time-frequency submatrix; Based on the space-time-frequency submatrix, a space domain forward smoothing vector is constructed, and the space domain forward smoothing vector includes a space domain matrix and a fine time delay difference matrix; A forward smoothing matrix is constructed through the space domain forward smoothing vector, and the rank of the autocorrelation matrix of the space-time-frequency joint vector is increased through the forward smoothing matrix.

6. The fast open-loop GNSS-R altimetry method based on space-time-frequency joint according to claim 1, characterized in that, The step of according to the subspace decomposition method, the autocorrelation matrix is analyzed, and the direct path time delay, the specular reflection path time delay and the direct path angle of arrival are estimated comprises: Based on the subspace decomposition method, a space-time-frequency joint matrix is constructed, and the autocorrelation matrix is decomposed through the space-time-frequency joint matrix to obtain a large eigenvector; Estimating fractional delay based on the large eigenvector, constructing a two-dimensional search network of boresight azimuth angle and integer delay, and estimating the boresight azimuth angle and integer delay through the peak position; Combining the fractional delay and the integer delay to obtain an accurate delay; the accurate delay includes a boresight delay and a specular reflection delay.

7. The fast open-loop GNSS-R altimetry method based on space-time-frequency joint according to claim 6, characterized in that, The accurate delay is: wherein, is a fractional delay, is a fractional delay, is an integer delay, and λ is the wavelength of the multi-frequency GNSS signal.

8. The fast open-loop GNSS-R altimetry method based on space-time-frequency joint according to claim 1, characterized in that, The step of deriving a satellite elevation angle from the boresight azimuth angle and inversely calculating the height of the receiver from the ground surface through the delay difference between the boresight delay and the specular reflection delay and the satellite elevation angle includes: Calculating the delay difference based on the boresight delay and the specular reflection delay; Calculating the satellite elevation angle through mutual calculation of the boresight azimuth angle according to a first formula; Calculating the height of the receiver from the ground surface based on the delay difference between the boresight delay and the specular reflection delay and the satellite elevation angle according to a second formula.

9. The fast open-loop GNSS-R altimetry method based on space-time-frequency joint according to claim 8, characterized in that, The first formula is: wherein, is the satellite elevation angle, is the direct diameter azimuth angle; The second formula is: where h is the height of the receiver from the ground, c is the speed of light, is the specular path delay, is the direct path delay, and θ1is the satellite elevation angle.

10. A fast open-loop GNSS-R altimetry system based on space-time-frequency joint, characterized in that, The system is used to perform the fast open-loop GNSS-R height-finding method based on space-time-frequency combination according to any one of claims 1 to 9, and the system includes: A signal acquisition module that acquires multi-frequency GNSS signals, estimates coarse delay and coarse Doppler frequency offset for each of the multi-frequency GNSS signals, and compensates the multi-frequency GNSS signals through the coarse Doppler frequency offset; A vector construction module that performs shaping processing on a beam in a time domain based on the coarse delay and the compensated multi-frequency GNSS signals, and constructs a space-time-frequency combined vector; A matrix rank increasing module that increases the rank of a self-correlation matrix of the space-time-frequency combined vector through a space domain smoothing technique, and solves the rank deficiency problem of the self-correlation matrix; A feature analysis module that performs feature analysis on the self-correlation matrix according to a subspace decomposition method, and estimates a boresight delay, a specular reflection delay, and a boresight azimuth angle; A height inversion module that derives a satellite elevation angle from the boresight azimuth angle, and inversely calculates the height of the receiver from the ground surface through the delay difference between the boresight delay and the specular reflection delay and the satellite elevation angle.

Images