Ground radiation source multi-satellite TDOA / FDOA combined positioning method based on earth ellipsoid constraint
By introducing earth ellipsoid constraints and Newton's iterative methods in the joint positioning of multi-star TDOA/FDOA, local convergence and iterative divergence of multi-dimensional parameter optimization problems are solved, positioning accuracy and robustness are improved, and computational complexity is reduced.
Patent Information
- Application Number
- CN202411091462.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-09
- Publication Date
- 2024-11-22
- Estimated Expiration
- 2044-08-09
AI Technical Summary
The multi-star TDOA/FDOA joint positioning method is prone to local convergence and iterative divergence problems, resulting in low positioning accuracy and high computational complexity, making it difficult to effectively deal with multi-dimensional parameter optimization problems.
Using the multi-star TDOA/FDOA joint positioning method based on earth ellipsoid constraints, using polynomial root calculation and implicit function derivative theory, Newton's iterative method is proposed, and the multidimensional parameter optimization problem is transformed into the optimization problem of single Lagrangian multiplier , and construct pseudo-linear observation equations through weighted multi-dimensional scaling analysis theory to improve robustness and accuracy.
It improves the robustness and global convergence of the optimization algorithm, reduces the computational complexity, and significantly improves the joint positioning accuracy of multi-star TDOA/FDOA, and enhances the robustness of large observation errors.
Smart Images

Figure CN119001784A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of wireless signal positioning technology, and in particular to a multi-satellite TDOA / FDOA joint positioning method for ground radiation sources based on earth ellipsoid constraints. Background Art
[0002] Wireless signal positioning technology is widely used in many industrial technology fields such as smart cities, radio astronomy, seismic survey, autonomous driving, emergency search and rescue, etc. It is also an indispensable supporting technology in military fields such as battlefield environment monitoring, tactical command coordination, cluster control, missile guidance, etc. It goes without saying that whether accurate, real-time and reliable location information services can be provided has become an important symbol of a country's comprehensive strength, and it is also an important strategic support for the country's political, economic and national defense security measures.
[0003] Positioning ground radiation sources based on communication satellite platforms is a common positioning system, including multi-satellite TDOA positioning and multi-satellite TDOA / FDOA joint positioning. Compared with multi-satellite TDOA positioning, multi-satellite TDOA / FDOA joint positioning uses frequency domain observations and therefore has higher positioning accuracy. However, the multi-satellite TDOA / FDOA joint positioning problem is essentially a multi-dimensional parameter joint optimization problem, which is prone to problems such as local convergence and iterative divergence [Guo FC, Ho K C. Aquadratic constraint solution method for TDOA and FDOA localization [A]. Proceedings of the IEEE International Conference on Acoustic, Speech and Signal Processing [C]. Prague, Czech: IEEE Press, May 2011: 2588-2591.]. Summary of the invention
[0004] In order to solve the above problems, the present invention proposes a multi-satellite TDOA / FDOA joint positioning method for ground radiation sources based on the earth ellipsoid constraint for radiation sources on the earth's surface. Based on the polynomial root-finding operation and implicit function derivation theory, a Newton iteration method for explicit optimization of only a single Lagrange multiplier is proposed, which can effectively integrate the earth ellipse constraint and transform the multidimensional parameter optimization problem into the optimization problem of a single Lagrange multiplier, thereby improving the robustness and global convergence of multidimensional parameter optimization and reducing the computational complexity. On the other hand, the new method also uses the weighted multidimensional scaling analysis theory to construct a pseudo-linear observation equation, thereby improving the robustness to large observation errors, and significantly improving the multi-satellite TDOA / FDOA joint positioning accuracy in combination with the earth ellipsoid constraint.
[0005] In order to achieve the above object, the present invention adopts the following technical solutions:
[0006] The present invention proposes a multi-satellite TDOA / FDOA joint positioning method for ground radiation sources based on the earth ellipsoid constraint. First, TDOA / FDOA observations about the ground radiation source are obtained by using one main satellite and multiple auxiliary satellites, and they are converted into distance difference observations and distance difference change rate observations respectively by using the known signal propagation speed and signal transmission frequency. Then, two scalar product matrices and two pseudo-inverse matrices are constructed by using these two observations, and a pseudo-linear observation matrix is constructed by introducing two auxiliary variables. Then, four first-order perturbation matrices and observation error coefficient matrices are calculated by using the first-order error analysis method to quantitatively describe the influence of the distance difference observation error and the distance difference change rate observation error on the pseudo-linear observation equation, and the singularity problem of the pseudo-linear observation error covariance matrix is avoided by matrix singular value decomposition, so as to calculate the optimal weighting matrix. Then, the earth ellipsoid constraint obeyed by the ground radiation source and the algebraic characteristics of the two auxiliary variables are used to construct a positioning optimization model containing triple quadratic equality constraints. In order to avoid joint optimization of multi-dimensional parameters, two two-variable polynomial equations about the distance between the ground radiation source and the host star and the rate of change of the distance are constructed. Based on the sequential linear convolution operation, a one-variable seventh-order polynomial equation about the distance between the ground radiation source and the host star is constructed. Based on the polynomial root operation and implicit function derivative theory, a Newton iteration method for explicit optimization of only a single Lagrange multiplier is proposed. Finally, the optimal value of the Lagrange multiplier and the optimal values of the two auxiliary variables are used to obtain the position parameters of the ground radiation source. The method specifically includes:
[0007] Step 1: Select M communication satellites to locate the ground radiation source, and use the M communication satellites to obtain the TDOA observations of the ground radiation source signal reaching the mth satellite and the first satellite. and FDOA observations And use TDOA observations and FDOA observations to further obtain distance difference observations and the rate of change of the distance difference
[0008] Step 2: Exploitation Construct an M×M scalar product matrix use and Construct an M×M scalar product matrix
[0009] Step 3: Exploitation Construct an M×4-order measurement matrix use Construct an M×4-order measurement matrix
[0010] Step 4: Exploitation Construct an M×5 pseudo-inverse matrix use Construct an M×5 pseudo-inverse matrix
[0011] Step 5: Exploitation and Construct a 2M×6 order pseudo linear measurement matrix
[0012] Step 6: Based on and One or more parameters in the calculation of four first-order perturbation matrices as well as
[0013] Step 7: Set the iteration index k:=0, set the iteration threshold δ, and Calculate the initial value of the iteration The iterative initial value of the ground radiation source position vector is obtained
[0014] Step 8: Based on as well as Calculate the 2M×2(M-1) order observation error coefficient matrix
[0015] Step 9: Matrix Perform singular value decomposition to determine the 2M×2(M-1) order column orthogonal matrix And use Calculate the 2(M-1)×2(M-1) order pseudo-linear observation error covariance matrix
[0016] Step 10: Based on and Calculate the optimal weighted matrix of order 2M×2M
[0017] Step 11: Based on and Calculate the 3×3 matrix in sequence 3 3×1 rank vectors as well as 6 scalars as well as And set the initial value of the Lagrange multiplier λ3
[0018] Step 12: Based on and One or more parameters in the Lagrange multiplier vector The coefficients of the bivariate polynomial as well as And calculate as well as On the Estimates of Lagrange Multipliers The first derivative of as well as
[0019] Step 13: Based on and One or more parameters in the calculation of the two-variable polynomial coefficient vector about the ground radiation source position vector u as well as And calculate as well as about The first derivative of as well as
[0020] Step 14: Construct two simultaneous distance β and distance change rate between the ground radiation source and the first satellite The two-variable polynomial equation is based on
[0021] and One or more parameters in the above two-variable polynomial equation are used to calculate the coefficients and And calculate and about The first derivative of and
[0022] Step 15: Construct a 7th-order polynomial equation for the distance β between the ground radiation source and the first satellite, based on and The coefficients of the univariate seventh-order polynomial equation are calculated using a sequential linear convolution operation. And use the power method to solve the positive real roots of the seventh-order polynomial equation
[0023] Step 16: Based on and Calculate the iterative value of the rate of change of the distance between the ground radiation source and the first satellite β
[0024] Step 17: Calculate the vector about The first derivative of
[0025] Step 18: Based on and Calculate the iterative value of the ground radiation source position vector And calculate about The first derivative of
[0026] Step 19: Right Perform Newton iterative update to get the updated value If the update amount Then go to step 20; otherwise, And go to step 12;
[0027] Step 20: Order as well as like Then stop the calculation; otherwise update the iteration index k:=k+1 and go to step 8.
[0028] Furthermore, in step 1, the distance difference observation is calculated in the following manner:
[0029]
[0030] in represents the distance difference between the mth satellite and the first satellite, represents the TDOA observation of the ground radiation source signal reaching the mth satellite and the first satellite, c is the signal propagation speed, u is the ground radiation source position vector, s m is the position vector of the mth satellite, 2≤m≤M, s1 is the position vector of the first satellite, Δr m1 Represents the observation error of the distance difference between the mth satellite and the first satellite;
[0031] The distance difference change rate observation is calculated as follows
[0032]
[0033] In the formula represents the observed rate of change of the distance difference between the mth satellite and the first satellite, represents the FDOA observation of the ground radiation source signal reaching the mth satellite and the first satellite, f0 represents the signal transmission frequency, is the velocity vector of the mth satellite, is the velocity vector of the first satellite, It represents the observation error of the rate of change of the distance difference between the mth satellite and the first satellite.
[0034] Further, in step 2, the following method is used to construct and
[0035]
[0036] In the formula represents the distance between the m1th satellite and the m2th satellite, and represents the inner product of the position vector difference and the velocity vector difference between the m1th satellite and the m2th satellite;
[0037] Furthermore, in step 3, the following method is used to construct and
[0038]
[0039] Furthermore, in step 4, the following method is used to construct and
[0040]
[0041] Where 1 M represents an M×1-order all-one column vector; 0 M Represents an M×1-order all-zero column vector.
[0042] Further, in step 5, the following method is used to construct
[0043]
[0044] In the formula represents the 5th column vector in the 5×5 identity matrix I5; Representation Matrix The first column vector in ; Represented by the matrix The matrix consisting of the 2nd to 4th columns in ; Representation Matrix The fifth column vector in ; Representation Matrix The sixth column vector in .
[0045] Furthermore, in step 6, the following method is used to calculate as well as
[0046]
[0047]
[0048] Where I M-1 and I M Represents (M-1)×(M-1) and M×M unit matrices respectively; the diag function is used to create a diagonal matrix; 0 M-1 represents a (M-1)×1-order all-zero column vector; O (M-1)×(M-1) and O (3M+1)×(M-1) They represent (M-1)×(M-1)-order and (3M+1)×(M-1)-order all-0 matrices respectively;
[0049] and They represent the distance difference observation vector and the distance difference change rate observation vector respectively; Π represents the permutation matrix, which satisfies The vec function is used to convert the matrix into a column vector in the order of the matrix columns;
[0050] I4 represents a 4×4 identity matrix, and 04 represents a 4×1 all-zero column vector.
[0051] Furthermore, in step 7, the iteration initial value is calculated in the following manner:
[0052]
[0053] in express The first 4 elements of the column vector, express The last element of ;
[0054] Then we get the iterative initial value of the ground radiation source position vector I3 represents a 3×3 identity matrix; 03 represents a 3×1 all-zero column vector.
[0055] Furthermore, in step 8, the following method is used to calculate
[0056]
[0057] In the formula and O M×5M Represents M×M 2 and M×5M all-zero matrices.
[0058] Furthermore, in step 9, the matrix Perform singular value decomposition to obtain
[0059]
[0060] In the formula represents a 2M×2M order left singular matrix; represents a 2(M-1)×2(M-1) right singular matrix; represents a 2(M-1)×2(M-1) order singular value diagonal matrix; O 2×2(M-1) Represents a 2×2(M-1)-order all-zero matrix;
[0061] Then use Calculate the 2(M-1)×2(M-1) order pseudo-linear observation error covariance matrix
[0062]
[0063] Where E represents the 2(M-1)×2(M-1) order observation error covariance matrix.
[0064] Furthermore, in step 10, the 2M×2M-order optimal weighting matrix is calculated in the following manner:
[0065]
[0066] Furthermore, in step 11, the following methods are used to calculate as well as
[0067]
[0068] Furthermore, in step 12, the following method is used to calculate as well as
[0069]
[0070]
[0071] In the formula e=0.081819790992113 represents the first eccentricity of the Earth;
[0072]
[0073] Then calculate as follows
[0074] In the formula
[0075]
[0076] Furthermore, in step 13, the following method is used to calculate as well as
[0077]
[0078]
[0079] Then calculate as follows as well as
[0080] Furthermore, in step 14, two distance β and distance change rate between the ground radiation source and the first satellite are constructed in the following manner: The two-variable polynomial equation for :
[0081]
[0082] The coefficients of the above two-variable polynomial equation are calculated as follows:
[0083]
[0084]
[0085] Where the symbols <·>1 and <·>2 represent the first and second elements in the vector respectively;
[0086] Then calculate them as follows and
[0087]
[0088] Furthermore, in step 15, a univariate seventh-order polynomial equation about the distance β between the ground radiation source and the first satellite is constructed in the following manner:
[0089]
[0090] In the formula Obtained through sequential linear convolution operations:
[0091]
[0092] Where * represents the sequential linear convolution operation.
[0093] Furthermore, in step 16, the following method is used to calculate
[0094] Furthermore, in step 17, the following method is used to calculate
[0095]
[0096] In the formula
[0097]
[0098] Furthermore, in step 18, the following method is used to calculate
[0099] Then calculate as follows
[0100]
[0101] Further, in step 19, Perform Newton iterative update:
[0102]
[0103] Where r e =6378.137km represents the equatorial radius of the earth.
[0104] Compared with the prior art, the present invention has the following beneficial effects:
[0105] The present invention proposes a multi-satellite TDOA / FDOA joint positioning method for ground radiation sources based on the earth ellipsoid constraint for radiation sources on the earth's surface. The method utilizes polynomial root-finding operations and implicit function derivation theory to transform the multi-dimensional parameter optimization problem into an optimization problem for a single Lagrange multiplier, which effectively improves the robustness and global convergence of the optimization algorithm and reduces the computational complexity. In addition, the method combines the earth ellipsoid constraint and weighted multidimensional scaling analysis theory to significantly improve the multi-satellite TDOA / FDOA joint positioning accuracy. BRIEF DESCRIPTION OF THE DRAWINGS
[0106] Figure 1 This is a flow chart of a multi-satellite TDOA / FDOA joint positioning method for ground radiation sources based on earth ellipsoid constraints according to an embodiment of the present invention;
[0107] Figure 2 It is a scatter diagram of the ground radiation source positioning result and a positioning error elliptic curve (XY plane coordinates in the ECEF coordinate system) provided by an embodiment of the present invention;
[0108] Figure 3 It is a scatter diagram of the ground radiation source positioning result and a positioning error elliptic curve (YZ plane coordinates in the ECEF coordinate system) provided by an embodiment of the present invention;
[0109] Figure 4 is a curve showing a change in the root mean square error of the ground radiation source position vector estimation as a function of the standard deviation σ of the observation error provided by an embodiment of the present invention;
[0110] Figure 5 It is a curve showing the change of the root mean square error of the ground radiation source position vector estimation along with the longitude of the ground radiation source provided by an embodiment of the present invention. DETAILED DESCRIPTION
[0111] The present invention will be further explained below with reference to the accompanying drawings and specific embodiments:
[0112] like Figure 1 As shown, a multi-satellite TDOA / FDOA joint positioning method for ground radiation sources based on earth ellipsoid constraints includes:
[0113] Step 1: Select M communication satellites to locate the ground radiation source, and use them to obtain the TDOA observations of the ground radiation source signal reaching the mth (2≤m≤M) satellite (called the auxiliary satellite) and the first satellite (called the primary satellite). and FDOA observations And use TDOA observations and FDOA observations to further obtain distance difference observations and the rate of change of the distance difference
[0114] Step 2: Observe the distance difference and the rate of change of the distance difference Construct an M×M scalar product matrix and the M×M scalar product matrix
[0115] Step 3: Observe the distance difference and the rate of change of the distance difference Construct an M×4-order measurement matrix and the M×4-order observation matrix
[0116] Step 4: Using the observation matrix and the observation matrix Construct an M×5 pseudo-inverse matrix and the M×5 pseudo-inverse matrix
[0117] Step 5: Using the scalar product matrix And the scalar product matrix And the pseudo-inverse matrix and the pseudo-inverse matrix Construct a 2M×6 order pseudo linear measurement matrix
[0118] Step 6: Calculate four first-order perturbation matrices in sequence as well as
[0119] Step 7: Set the iteration index k:=0, set the iteration threshold δ, and Calculate the initial value of the iteration Thus, the iterative initial value of the ground radiation source position vector is obtained
[0120] Step 8: Calculate the 2M×2(M-1) order observation error coefficient matrix
[0121] Step 9: Matrix Perform singular value decomposition to determine the 2M×2(M-1) order column orthogonal matrix And use Calculate the 2(M-1)×2(M-1) order pseudo-linear observation error covariance matrix
[0122] Step 10: Based on and Calculate the optimal weighted matrix of order 2M×2M
[0123] Step 11: Calculate the 3×3 matrix in sequence 3 3×1 rank vectors as well as 6 scalars as well as And set the initial value of the Lagrange multiplier λ3
[0124] Step 12: Calculate the Lagrange multiplier vector The coefficients of the bivariate polynomial as well as and calculate their estimates with respect to the Lagrange multipliers The first derivative of as well as
[0125] Step 13: Calculate the coefficient vector of the bivariate polynomial about the ground radiation source position vector u as well as and calculate their estimates with respect to the Lagrange multipliers The first derivative of as well as
[0126] Step 14: Construct two simultaneous distance β and distance change rate between the ground radiation source and the host star Calculate the coefficients of the above two-variable polynomial equation and And calculate and On the Estimates of Lagrange Multipliers The first derivative of and
[0127] Step 15: Construct a 7th-order polynomial equation about the distance β between the ground radiation source and the host star, based on and The coefficients of the univariate seventh-order polynomial equation are calculated using a sequential linear convolution operation. The power method is used to solve the positive real roots of the seventh-order polynomial equation (denoted as ).
[0128] Step 16: Calculate the rate of change of distance between the ground radiation source and the host star The iteration value (denoted as ).
[0129] Step 17: Calculate the vector On the Estimates of Lagrange Multipliers The first derivative of
[0130] Step 18: Calculate the iterative value of the ground radiation source position vector And calculate On the Estimates of Lagrange Multipliers The first derivative of
[0131] Step 19: Estimate the Lagrange multiplier Perform Newton iterative update to get the updated value If the update amount Then go to step 20; otherwise, And go to step 12.
[0132] Step 20: Order as well as like Then stop the calculation; otherwise update the iteration index k:=k+1 and go to step 8.
[0133] Furthermore, in step 1, M communication satellites are selected to locate the ground radiation source, wherein the first satellite is the primary satellite and the remaining satellites are auxiliary satellites. The position vector of the ground radiation source is denoted as u, and the position vector and velocity vector of the mth satellite are denoted as s respectively. m and Using them, we can obtain the TDOA observations of the ground radiation source signal reaching the mth (2≤m≤M) satellite and the main satellite. and FDOA observations The TDOA observations Multiplying by the signal propagation speed c gives the distance difference observation The corresponding expression is
[0134]
[0135] Where s1 is the position vector of the first satellite, Δr m1 represents the distance difference observation error;
[0136] Then the FDOA observation Multiply by the signal propagation speed c and then divide by the signal transmission frequency f0 to get the observed rate of change of the distance difference The corresponding expression is
[0137]
[0138] In the formula is the velocity vector of the first satellite, Represents the observation error of the rate of change of distance difference.
[0139] Furthermore, in step 2, the distance difference observation is used and the rate of change of the distance difference Construct an M×M scalar product matrix and the M×M scalar product matrix The elements of these two matrices are
[0140]
[0141] In the formula represents the distance between the m1th satellite and the m2th satellite; represents the inner product of the position vector difference and the velocity vector difference between the m1th satellite and the m2th satellite;
[0142] Furthermore, in step 3, the distance difference observation is used and the rate of change of the distance difference Construct an M×4-order measurement matrix and the M×4-order observation matrix The elements of these two matrices are
[0143]
[0144] Furthermore, in step 4, the observation matrix and the observation matrix Construct an M×5 pseudo-inverse matrix and the M×5 pseudo-inverse matrix The calculation formulas of these two matrices are
[0145]
[0146] Where 1 M represents an M×1-order all-one column vector; 0 M Represents an M×1-order all-zero column vector.
[0147] Furthermore, in step 5, the scalar product matrix is used And the scalar product matrix And the pseudo-inverse matrix and the pseudo-inverse matrix Construct a 2M×6 order pseudo linear measurement matrix The calculation formula of this matrix is
[0148]
[0149] In the formula represents the 5th column vector in the 5×5 identity matrix I5; Representation Matrix The first column vector in ; Represented by the matrix The matrix consisting of the 2nd to 4th columns in ; Representation Matrix The fifth column vector in ; Representation Matrix The sixth column vector in .
[0150] Furthermore, in step 6, four first-order perturbation matrices are calculated in sequence: as well as The calculation formulas for these four matrices are
[0151]
[0152] Where I M-1 and I M Represents (M-1)×(M-1) and M×M unit matrices respectively; the diag function is used to create a diagonal matrix; 0 M-1 represents a (M-1)×1-order all-zero column vector; O (M-1)×(M-1) and O (3M+1)×(M-1) They represent (M-1)×(M-1)-order and (3M+1)×(M-1)-order all-0 matrices respectively;
[0153] and They represent the distance difference observation vector and the distance difference change rate observation vector respectively; Π represents the permutation matrix, which satisfies The vec function is used to convert the matrix into a column vector in the order of the matrix columns;
[0154] I4 represents a 4×4 identity matrix, and 04 represents a 4×1 all-zero column vector.
[0155] Furthermore, in step 7, let the iteration index k:=0, set the iteration threshold value δ, and calculate the iteration initial value The corresponding calculation formula is
[0156]
[0157] in express The first 4 elements of the column vector, express The last element of .
[0158] The iterative initial value of the ground radiation source position vector is obtained Where I3 represents a 3×3 unit matrix; 03 represents a 3×1 column vector of all zeros.
[0159] Furthermore, in step 8, the 2M×2(M-1) order observation error coefficient matrix is calculated The corresponding calculation formula is
[0160]
[0161] In the formula and O M×5M Represents M×M 2 The matrix is M×5M and M×5M all-zero matrices.
[0162] Furthermore, in step 9, the matrix Performing singular value decomposition, we can get
[0163]
[0164] In the formula represents a 2M×2M order left singular matrix; represents a 2(M-1)×2(M-1) right singular matrix; represents a 2(M-1)×2(M-1) order singular value diagonal matrix; O 2×2(M-1) Represents a 2×2(M-1)-order all-zero matrix.
[0165] Then, using the 2M×2(M-1) orthogonal matrix Calculate the 2(M-1)×2(M-1) order pseudo-linear observation error covariance matrix The corresponding calculation formula is
[0166]
[0167] Where E represents the 2(M-1)×2(M-1) order observation error covariance matrix.
[0168] Furthermore, in step 10, the optimal weighted matrix of order 2M×2M is calculated The corresponding calculation formula is
[0169]
[0170] Furthermore, in step 11, the 3×3 matrix is calculated in sequence 3 3×1 rank vectors as well as 6 scalars as well as The corresponding calculation formulas are
[0171]
[0172] Then set the initial value of the Lagrange multiplier λ3
[0173] Furthermore, in step 12, the Lagrange multiplier vector is calculated. The coefficients of the bivariate polynomial as well as The corresponding calculation formula is
[0174]
[0175]
[0176] In the formula Where e = 0.081819790992113 represents the first eccentricity of the earth, and the diag function is used to create a diagonal matrix;
[0177] Then calculate their estimates of the Lagrange multipliers The first derivative of as well as The corresponding calculation formula is
[0178]
[0179] In the formula
[0180]
[0181] Furthermore, in step 13, the binary polynomial coefficient vector about the ground radiation source position vector u is calculated. as well as The corresponding calculation formula is
[0182]
[0183] Then calculate their estimates of the Lagrange multipliers The first derivative of as well as The corresponding calculation formula is
[0184]
[0185] Furthermore, in step 14, two distance β and distance change rate between the ground radiation source and the primary star are constructed simultaneously. The two-variable polynomial equation calculates the two polynomial coefficients and The corresponding calculation formula is
[0186]
[0187]
[0188] In the formula
[0189]
[0190] Wherein the symbols <·>1 and <·>2 represent the first and second elements in the vector respectively.
[0191] Then calculate separately and On the Estimates of Lagrange Multipliers The first derivative of and The corresponding calculation formula is
[0192]
[0193] Furthermore, in step 15, a univariate seventh-order polynomial equation about the distance β between the ground radiation source and the host star is constructed, and its polynomial coefficient is calculated. The corresponding expression is
[0194]
[0195] The polynomial coefficients are It can be obtained through sequential linear convolution operations, and the corresponding calculation formula is
[0196]
[0197] Where * represents the sequential linear convolution operation.
[0198] The power method is then used to solve the equation for the positive real roots, which are recorded as
[0199] Furthermore, in step 16, the distance change rate between the ground radiation source and the host star is calculated. The iteration value is recorded as The corresponding calculation formula is
[0200]
[0201] Furthermore, in step 17, the vector is calculated On the Estimates of Lagrange Multipliers The first derivative of The corresponding calculation formula is
[0202]
[0203] In the formula
[0204]
[0205]
[0206] Furthermore, in step 18, the iterative value of the ground radiation source position vector is calculated The corresponding calculation formula is
[0207]
[0208] Then calculate On the Estimates of Lagrange Multipliers The first derivative of The corresponding calculation formula is
[0209]
[0210] Furthermore, in step 19, the Lagrange multiplier estimation value Perform Newton iterative update, the corresponding calculation formula is
[0211]
[0212] Where r e =6378.137km represents the radius of the earth's equator. Then go to step 20; otherwise, And go to step 12.
[0213] Further, in step 20, as well as like Then stop the calculation; otherwise update the iteration index k:=k+1 and go to step 8.
[0214] As a specific example, consider locating a radiation source on the surface of the earth, whose longitude is 133.14° and latitude is 32.20°. The position vector in the ECEF coordinate system is u=[-3694.03942.03379.2] T (km), there are 7 communication satellites that perform TDOA / FDOA joint positioning on the radiation source. The latitude, longitude and altitude of the 7 communication satellites are shown in Table 1, and the movement speeds of the 7 communication satellites are shown in Table 2.
[0215] Table 1 Latitude, longitude and altitude of 7 communication satellites
[0216]
[0217] Table 2 Movement speeds of 7 communication satellites
[0218]
[0219] The observation error covariance matrix is set to Where I6 represents the 6×6 unit matrix, 1 6×6 represents a 6×6 all-1 matrix, O 6×6 represents a 6×6 all-zero matrix, and σ represents the standard deviation of the observation error.
[0220] First, the observation error standard deviation σ is set to σ = 0.5, Figure 2 The scatter diagram of the ground radiation source positioning results and the positioning error elliptic curve (XY plane coordinates in the ECEF coordinate system) are given. Figure 3The scatter diagram of the ground radiation source positioning results and the positioning error elliptic curve (YZ plane coordinates in the ECEF coordinate system) are given. Figure 2 and Figure 3 It can be seen that the shape of the scatter plot of the positioning results of the positioning method disclosed in this patent is consistent with the shape of the positioning error ellipse, and the high probability corresponds to the large-area ellipse, and the low probability corresponds to the small-area ellipse, thereby verifying the effectiveness of the new method.
[0221] Then change the value of the observation error standard deviation σ, Figure 4 The curve of the root mean square error of the ground radiation source position vector estimation with the standard deviation of the observation error σ is given; then the standard deviation of the observation error σ is set to σ=0.5, and the longitude of the ground radiation source is changed. Figure 5 The curve of the root mean square error of the ground radiation source position vector estimation versus the longitude of the ground radiation source is given.
[0222] from Figure 4 and Figure 5 It can be seen that: (1) the multi-satellite TDOA / FDOA joint positioning method for ground radiation sources based on the earth ellipsoid constraint disclosed in this patent can asymptotically approach the corresponding Cramer-Rao bound for the root mean square error of the ground radiation source position vector estimation, thereby verifying the asymptotic statistical optimality of the new method; (2) compared with the multi-satellite TDOA / FDOA joint positioning method that does not utilize the earth ellipsoid constraint, the positioning accuracy of the new method has been significantly improved, and the performance gain obtained will increase with the increase of the observation error standard deviation and the longitude of the ground radiation source.
[0223] The above is only a preferred embodiment of the present invention. It should be pointed out that a person skilled in the art can make several improvements and modifications without departing from the principle of the present invention. These improvements and modifications should also be regarded as within the scope of protection of the present invention.
Claims
1. A multi-satellite TDOA / FDOA joint positioning method for ground radiation sources based on earth ellipsoid constraints, characterized in that: include: Step 1: Select M communication satellites to locate the ground radiation source, and use the M communication satellites to obtain the TDOA observations of the ground radiation source signal reaching the mth satellite and the first satellite. and FDOA observations And use TDOA observations and FDOA observations to further obtain distance difference observations and the rate of change of the distance difference Step 2: Exploitation Construct an M×M scalar product matrix use and Construct an M×M scalar product matrix Step 3: Exploitation Construct an M×4-order measurement matrix use Construct an M×4-order measurement matrix Step 4: Exploitation Construct an M×5 pseudo-inverse matrix use Construct an M×5 pseudo-inverse matrix Step 5: Exploitation and Construct a 2M×6 order pseudo linear measurement matrix Step 6: Based on and One or more parameters in the calculation of four first-order perturbation matrices as well as Step 7: Set the iteration index k:=0, set the iteration threshold δ, and Calculate the initial value of the iteration Thus, the iterative initial value of the ground radiation source position vector is obtained Step 8: Based on as well as Calculate the 2M×2(M-1) order observation error coefficient matrix Step 9: Matrix Perform singular value decomposition to determine the 2M×2(M-1) order column orthogonal matrix And use Calculate the 2(M-1)×2(M-1) order pseudo-linear observation error covariance matrix Step 10: Based on and Calculate the optimal weighted matrix of order 2M×2M Step 11: Based on and Calculate the 3×3 matrix in sequence 3 3×1 rank vectors as well as 6 scalars as well as And set the initial value of the Lagrange multiplier λ3 Step 12: Based on and One or more parameters in the Lagrange multiplier vector The coefficients of the bivariate polynomial as well as And calculate as well as On the Estimates of Lagrange Multipliers The first derivative of as well as Step 13: Based on and One or more parameters in the calculation of the two-variable polynomial coefficient vector about the ground radiation source position vector u as well as And calculate as well as about The first derivative of as well as Step 14: Construct two simultaneous distance β and distance change rate between the ground radiation source and the first satellite The two-variable polynomial equation is based on and One or more parameters in the above two-variable polynomial equation are used to calculate the coefficients and And calculate and about The first derivative of and Step 15: Construct a 7th-order polynomial equation for the distance β between the ground radiation source and the first satellite, based on and The coefficients of the univariate seventh-order polynomial equation are calculated using a sequential linear convolution operation. And use the power method to solve the positive real roots of the seventh-order polynomial equation Step 16: Based on and Calculate the rate of change of the distance between the ground radiation source and the first satellite Iteration value of Step 17: Calculate the vector about The first derivative of Step 18: Based on and Calculate the iterative value of the ground radiation source position vector And calculate about The first derivative of Step 19: Right Perform Newton iterative update to get the updated value If the update amount Then go to step 20; otherwise, And go to step 12; Step 20: Order as well as like Then stop the calculation; otherwise update the iteration index k:=k+1 and go to step 8.
2. The multi-satellite TDOA / FDOA joint positioning method for ground radiation sources based on the earth ellipsoid constraint according to claim 1 is characterized in that: In step 1, the distance difference observation is calculated in the following manner: in represents the distance difference between the mth satellite and the first satellite, represents the TDOA observation of the ground radiation source signal reaching the mth satellite and the first satellite, c is the signal propagation speed, u is the ground radiation source position vector, s m is the position vector of the mth satellite, 2≤m≤M, s1 is the position vector of the first satellite, Δr m1 Represents the observation error of the distance difference between the mth satellite and the first satellite; The distance difference change rate observation is calculated as follows In the formula represents the observed rate of change of the distance difference between the mth satellite and the first satellite, represents the FDOA observation of the ground radiation source signal reaching the mth satellite and the first satellite, f0 represents the signal transmission frequency, is the velocity vector of the mth satellite, is the velocity vector of the first satellite, It represents the observation error of the rate of change of the distance difference between the mth satellite and the first satellite.
3. The multi-satellite TDOA / FDOA joint positioning method for ground radiation sources based on earth ellipsoid constraints according to claim 1 is characterized in that: In step 2, an M×M scalar product matrix is constructed as follows: and the M×M scalar product matrix In the formula represents the distance between the m1th satellite and the m2th satellite, and 1≤m1; m2≤M; represents the inner product of the position vector difference and the velocity vector difference between the m1th satellite and the m2th satellite; 4. The multi-satellite TDOA / FDOA joint positioning method for ground radiation sources based on earth ellipsoid constraints according to claim 1, characterized in that: In step 3, an M×4-order measurement matrix is constructed as follows: and the M×4-order observation matrix Where s1 is the position vector of the first satellite, s m is the position vector of the mth satellite, 2≤m≤M, is the velocity vector of the first satellite, is the velocity vector of the mth satellite.
5. The multi-satellite TDOA / FDOA joint positioning method for ground radiation sources based on earth ellipsoid constraints according to claim 1, characterized in that: In step 4, a pseudo-inverse matrix of order M×5 is constructed as follows: and the M×5 pseudo-inverse matrix Where 1 M represents an M×1-order all-one column vector; 0 M Represents an M×1-order all-zero column vector.
6. The multi-satellite TDOA / FDOA joint positioning method for ground radiation sources based on earth ellipsoid constraints according to claim 1, characterized in that: In step 5, a 2M×6 order pseudo linear measurement matrix is constructed as follows: In the formula Represents the 5th column vector in the 5×5-order identity matrix I5; 0 M represents an M×1-order all-zero column vector; Representation Matrix The first column vector in ; Represented by the matrix The matrix consisting of the 2nd to 4th columns in ; Representation Matrix The fifth column vector in ; Representation Matrix The sixth column vector in .
7. The multi-satellite TDOA / FDOA joint positioning method for ground radiation sources based on earth ellipsoid constraints according to claim 1, characterized in that: In step 6, four first-order perturbation matrices are calculated as follows: as well as Where I M-1 and I M denote the (M-1)×(M-1)-order and M×M-order identity matrices respectively; The diag function is used to create a diagonal matrix; 0 M-1 represents a (M-1)×1-order all-zero column vector; O (M-1)×(M-1) and O (3M+1)×(M-1) They represent (M-1)×(M-1)-order and (3M+1)×(M-1)-order all-0 matrices respectively; and They represent the distance difference observation vector and the distance difference change rate observation vector respectively; Π represents the permutation matrix, which satisfies The vec function is used to convert the matrix into a column vector in the order of the matrix columns; 1 M represents an M×1-order all-one column vector; I4 represents a 4×4 identity matrix, and 04 represents a 4×1 all-zero column vector.
8. The multi-satellite TDOA / FDOA joint positioning method for ground radiation sources based on earth ellipsoid constraints according to claim 6, characterized in that: In step 7, the initial value of the iteration is calculated as follows: in express The first 4 elements of the column vector, express The last element of ; Then we get the iterative initial value of the ground radiation source position vector I3 represents a 3×3 identity matrix; 03 represents a 3×1 all-zero column vector.
9. The multi-satellite TDOA / FDOA joint positioning method for ground radiation sources based on earth ellipsoid constraints according to claim 8, characterized in that: In step 8, the 2M×2(M-1) order observation error coefficient matrix is calculated as follows: In the formula and O M×5M Represents M×M 2 M×5M all-zero matrices; Represents the 5th column vector in the 5×5 identity matrix I5.
10. The multi-satellite TDOA / FDOA joint positioning method for ground radiation sources based on earth ellipsoid constraints according to claim 6, characterized in that: In step 9, the matrix Perform singular value decomposition to obtain In the formula represents a 2M×2M order left singular matrix; represents a 2(M-1)×2(M-1) right singular matrix; represents a 2(M-1)×2(M-1) order singular value diagonal matrix; O 2×2(M-1) Represents a 2×2(M-1)-order all-zero matrix; Then use Calculate the 2(M-1)×2(M-1) order pseudo-linear observation error covariance matrix Where E represents the 2(M-1)×2(M-1) order observation error covariance matrix; Preferably, in step 10, the 2M×2M order optimal weighting matrix is calculated in the following manner: Preferably, in step 11, the 3×3 matrix is calculated in the following manner: 3 3×1 rank vectors as well as 6 scalars as well as Preferably, in step 12, the calculation is performed in the following manner: as well as In the formula s1 is the position vector of the first satellite, is the velocity vector of the first satellite; e=0.081819790992113 represents the first eccentricity of the earth, and the diag function is used to create a diagonal matrix; Then calculate as follows as well as In the formula Preferably, in step 13, the calculation is performed in the following manner: as well as Then calculate as follows as well as Preferably, in step 14, two distance β and distance change rate between the ground radiation source and the first satellite are constructed in the following manner: The bivariate polynomial equation for : The coefficients of the above two-variable polynomial equation are calculated as follows: Where the symbols <·>1 and <·>2 represent the first and second elements in the vector respectively; Then calculate them as follows and about The first derivative of and Preferably, in step 15, a univariate seventh-order polynomial equation about the distance β between the ground radiation source and the first satellite is constructed in the following manner: In the formula Obtained through sequential linear convolution operations: Where * represents the sequential linear convolution operation; Preferably, in step 16, the rate of change of the distance between the ground radiation source and the first satellite is calculated in the following manner: Iteration value of Preferably, in step 17, the vector is calculated in the following manner: about The first derivative of In the formula Preferably, in step 18, the iterative value of the ground radiation source position vector is calculated in the following manner: Then calculate as follows On the Estimates of Lagrange Multipliers The first derivative of Preferably, in step 19, the Lagrange multiplier estimation value is calculated in the following manner: Perform Newton iterative update: Where r e =6378.137km represents the equatorial radius of the earth.
Citation Information
Patent Citations
External radiation source TDOA / FDOA error correction based positioning method
CN109633581A
Moving radiation source TDOA and FDOA positioning method based on weighted multidimensional scale and Lagrange multiplier technology
CN111551895A
Globally-convergent geo-location algorithm
US20100315290A1
Cited By
Material source analysis visualization method and device based on local singularity
CN121478875A
