A method for estimating GNSS receiver observation noise

Abstract

The invention discloses a method for estimating the observation noise of a GNSS receiver, including: S1: obtaining a receiver reference coordinate value and an observation; S2: initializing parameters; S3: obtaining a current observation noise estimate; S4: obtaining an observation correction value; S5: obtaining a receiver position estimate; S6: obtaining a receiver position coordinate error; S7: judging a jump step according to the current receiver position coordinate error; S8: judging whether to accept an inferior solution acceptance condition; S9: assigning an optimal estimate to the observation noise; S10: judging whether the number of iterations exceeds a limit; S11: judging whether to terminate the search for the optimal estimate of the observation noise. The invention does not rely on observation noise modeling, but randomly searches and screens the observation noise in a multidimensional space, avoids the noise modeling process, obtains a global observation noise estimate, and estimates the observation noise of the inaccurately modeled part under the condition that the partial error source is accurately modeled.

Description

Technical Field

[0001] The present invention relates to the field of receiver observation quantity noise, and in particular to a method for estimating the observation quantity noise of a GNSS receiver. Background Art

[0002] The signals broadcast by GNSS satellites are affected by many factors along the propagation path, including satellite clock error, receiver clock error, ionospheric delay, tropospheric delay, ephemeris error, multipath, receiver noise, etc., which causes the observations obtained by the receiver to contain noise, thus interfering with related functions.

[0003] Noise estimation is an important part of the GNSS system and a prerequisite for the normal operation of the GNSS system. Traditional noise estimation mainly relies on the error source models to calculate the error components under the current measurement environment, and then synthesizes the errors to obtain the total observation noise estimate. This model has high requirements for modeling accuracy, and the accuracy of noise estimation is extremely dependent on the accuracy of modeling.

[0004] The iterative search method can search for data with specific requirements in high-dimensional space and can effectively approach the global optimal solution, which facilitates the estimation of receiver observation noise. Summary of the invention

[0005] In order to solve the above technical problems, the present invention proposes a method for estimating the noise of GNSS receiver observations.

[0006] A method for estimating GNSS receiver observation noise comprises the following steps:

[0007] S1: Obtain the receiver reference coordinate value and observation information;

[0008] S2: Initialize the optimal estimate of the observation noise, the inferior solution acceptance condition, the position coordinate error threshold, the termination search condition, and the number of iterations;

[0009] S3: Get the current observation noise estimate;

[0010] S4: Obtain the observed value correction value;

[0011] The observed quantity is a pseudorange or a carrier phase;

[0012] S5: Obtaining an estimated value of the receiver position;

[0013] S6: Obtain the receiver position coordinate error;

[0014] S7: Determine the jump step according to the current receiver position coordinate error;

[0015] S8: Determine whether to accept the inferior solution acceptance condition;

[0016] S9: assigning the current observation noise estimation value to the observation noise optimal estimation value;

[0017] S10: Determine whether the number of iterations exceeds the limit;

[0018] S11: Determine whether to terminate the search for the optimal estimate of the observation noise.

[0019] Furthermore, in a method for estimating GNSS receiver observation noise, step S2 comprises the following sub-steps:

[0020] S21: Initialize the optimal estimate of the observation noise: Assume that the number of available satellite observations is n, and set the optimal estimate of the observation noise to be estimated All are initialized to 0, the formula is:

[0021] ;

[0022] in, represents the optimal estimate of the noise of the observation to be estimated, represents the components corresponding to the optimal estimate of the noise of the 1st to nth observations, T represents the matrix transpose, and n represents the number of available satellite observations;

[0023] S22: Initialize the inferior solution acceptance condition P, which is a probability parameter in the range of 0.80~0.99;

[0024] S23: Initialize position coordinate error threshold , is the expected solution accuracy, within the range of 0.001~0.1 meters;

[0025] S24: Initialize the termination search condition, and terminate when one of condition 1 or condition 2 is met:

[0026] Condition 1: With the update of the inferior solution acceptance condition P, the current inferior solution acceptance condition P is sufficiently small, ranging from 0.0001 to 0.01;

[0027] Condition 2: The number of consecutive failures to update the optimal estimate of the observed quantity exceeds a predetermined value, ranging from 50 to 500;

[0028] S25: Initialization iteration limit: Iteration limit Initialized to a positive integer in the range of 3000~10000.

[0029] Furthermore, in a method for estimating GNSS receiver observation noise, step S3 comprises the following sub-steps:

[0030] S31: Generate a set of random observation noise disturbance values ​​according to the current random noise disturbance intensity;

[0031] S32: Superimpose the random observation noise disturbance value on the optimal estimation value of the observation noise to be estimated to obtain the current observation noise estimation value, the formula is:

[0032] ;

[0033] in, represents the optimal estimate of the noise of the observation to be estimated, represents the noise disturbance value of the random observation quantity, Represents the noise estimate of the current observation.

[0034] Furthermore, in a method for estimating GNSS receiver observation noise, step S4 comprises the following sub-steps:

[0035] S41: Correct the observation quantity using the current observation quantity noise estimation value, the formula is:

[0036] ;

[0037] in, represents the correction value component of the i-th observation, represents the raw measurement component of the ith observation, Represents the optimal estimate component of the i-th observation noise;

[0038] S42: Obtain the observed value correction value, the formula is:

[0039] ;

[0040] in, represents the observed value correction, Represents the correction value component of the 1st to nth observations.

[0041] Furthermore, in a method for estimating GNSS receiver observation noise, step S5 comprises the following sub-steps:

[0042] S51: Calculate the receiver position estimate value through the observation correction value and the satellite position, and use the single point positioning method to iteratively estimate the receiver position. First, obtain the unit vector determined in the kth iteration. The formula is:

[0043] ;

[0044] in, represents the coefficient matrix consisting of the unit vectors determined by the receiver-to-satellite position in the kth iteration, , , They represent the three components of the unit vector determined from the receiver position to the j-th satellite position in three-dimensional coordinates, and the subscript N represents the total number of satellites;

[0045] S52: Obtain the observation residual matrix in the kth iteration, the formula is:

[0046] ;

[0047] in, represents the residual matrix of observations in the kth iteration, represents the geometric distance between the nth satellite and the receiver, the superscript s represents the satellite position, and the subscript r represents the receiver position;

[0048] S53: Calculate the incremental vector of the receiver position and local clock error during the k iterations. The formula is:

[0049] ;

[0050] in, represents the incremental vector of the receiver position and local clock error during k iterations, represent Transpose;

[0051] S54: After each iteration, The value of ,when The iteration is terminated when the 2-norm of the vector is less than 0.1. The formula is:

[0052] ;

[0053] in, represents the estimated value in the kth iteration, Represents the initial value in the kth iteration.

[0054] Furthermore, in a method for estimating GNSS receiver observation noise, the S6 comprises the following sub-steps:

[0055] S61: Calculate the current receiver position coordinate error through the receiver position estimate value and the receiver position reference value, and the formula is:

[0056] ;

[0057] in, Represents the current receiver position coordinate error, represents the receiver position estimate, , , Represent the coordinate points of the three-dimensional coordinates of the receiver position estimate, Represents the receiver position reference value, , , They respectively represent the coordinate points of the three-dimensional coordinates of the receiver position reference value.

[0058] Furthermore, a method for estimating GNSS receiver observation noise, said S7 comprises the following sub-steps:

[0059] Determine the current receiver position coordinate error and position coordinate error threshold Size relationship:

[0060] like , then continue to step S8;

[0061] like , then update the position coordinate error threshold for , jump to step S9.

[0062] Furthermore, a method for estimating GNSS receiver observation noise, said S8 comprises the following sub-steps:

[0063] When the random number generated by the random number generator is less than the current inferior solution acceptance condition P, the inferior solution is accepted, otherwise it is not accepted;

[0064] If accepted, go to step S9, if not accepted, go to step S10;

[0065] The S10 includes the following sub-steps:

[0066] During each iteration, the iteration counter is incremented by 1. When the value in the iteration counter exceeds Time exceeds the limit, otherwise it does not exceed the limit;

[0067] If not, go to step S1 and continue iterating; if yes, jump to step S11;

[0068] The S11 comprises the following sub-steps:

[0069] If not, update the inferior solution acceptance condition and reset the number of iterations, and jump to step S1;

[0070] If yes, the optimal estimate of the current observation noise is used as the final observation noise estimation result.

[0071] The beneficial effects of the present invention are: a method for estimating the observation noise of a GNSS receiver, which does not rely on accurate observation noise modeling, thereby avoiding the problem of inaccurate observation noise estimation results when the noise model is inaccurate or incomplete; through a certain search strategy, the observation noise is randomly searched and screened in a multi-dimensional space, avoiding the noise modeling process, and obtaining a global observation noise estimate. BRIEF DESCRIPTION OF THE DRAWINGS

[0072] Figure 1 It is a flow chart of the method of the present invention. DETAILED DESCRIPTION

[0073] The present invention is further described below, but the protection scope of the present invention is not limited to the following description.

[0074] As attached Figure 1 As shown, a method for estimating the noise of a GNSS receiver observation quantity comprises the following steps:

[0075] S1: Obtain the receiver reference coordinate value and observation information;

[0076] S2: Initialize the optimal estimate of the observation noise, the inferior solution acceptance condition, the position coordinate error threshold, the termination search condition, and the number of iterations;

[0077] S3: Get the current observation noise estimate;

[0078] S4: Obtain the observed value correction value;

[0079] The observed quantity is a pseudorange or a carrier phase;

[0080] S5: Obtaining an estimated value of the receiver position;

[0081] S6: Obtain the receiver position coordinate error;

[0082] S7: Determine the jump step according to the current receiver position coordinate error;

[0083] S8: Determine whether to accept the inferior solution acceptance condition;

[0084] S9: assigning the current observation noise estimation value to the observation noise optimal estimation value;

[0085] S10: Determine whether the number of iterations exceeds the limit;

[0086] S11: Determine whether to terminate the search for the optimal estimate of the observation noise.

[0087] Furthermore, in a method for estimating GNSS receiver observation noise, step S2 comprises the following sub-steps:

[0088] S21: Initialize the optimal estimate of the observation noise: Assume that the number of available satellite observations is n, and set the optimal estimate of the observation noise to be estimated All are initialized to 0, the formula is:

[0089] ;

[0090] in, represents the optimal estimate of the noise of the observation to be estimated, represents the components corresponding to the optimal estimate of the noise of the 1st to nth observations, T represents the matrix transpose, and n represents the number of available satellite observations;

[0091] S22: Initialize the inferior solution acceptance condition P, which is a probability parameter in the range of 0.80~0.99;

[0092] S23: Initialize position coordinate error threshold , is the expected solution accuracy, within the range of 0.001~0.1 meters;

[0093] S24: Initialize the termination search condition, and terminate when one of condition 1 or condition 2 is met:

[0094] Condition 1: With the update of the inferior solution acceptance condition P, the current inferior solution acceptance condition P is sufficiently small, ranging from 0.0001 to 0.01;

[0095] Condition 2: The number of consecutive failures to update the optimal estimate of the observed quantity exceeds a predetermined value, ranging from 50 to 500;

[0096] S25: Initialization iteration limit: Iteration limit Initialized to a positive integer in the range of 3000~10000.

[0097] Furthermore, in a method for estimating GNSS receiver observation noise, step S3 comprises the following sub-steps:

[0098] S31: Generate a set of random observation noise disturbance values ​​according to the current random noise disturbance intensity;

[0099] S32: Superimpose the random observation noise disturbance value on the optimal estimation value of the observation noise to be estimated to obtain the current observation noise estimation value, the formula is:

[0100] ;

[0101] in, represents the optimal estimate of the noise of the observation to be estimated, represents the noise disturbance value of the random observation quantity, Represents the noise estimate of the current observation.

[0102] Furthermore, in a method for estimating GNSS receiver observation noise, step S4 comprises the following sub-steps:

[0103] S41: Correct the observation quantity using the current observation quantity noise estimation value, the formula is:

[0104] ;

[0105] in, represents the correction value component of the i-th observation, represents the raw measurement component of the ith observation, Represents the optimal estimate component of the i-th observation noise;

[0106] S42: Obtain the observed value correction value, the formula is:

[0107] ;

[0108] in, represents the observed value correction, Represents the correction value component of the 1st to nth observations.

[0109] Furthermore, in a method for estimating GNSS receiver observation noise, step S5 comprises the following sub-steps:

[0110] S51: Calculate the receiver position estimate value through the observation correction value and the satellite position, and use the single point positioning method to iteratively estimate the receiver position. First, obtain the unit vector determined in the kth iteration. The formula is:

[0111] ;

[0112] in, represents the coefficient matrix consisting of the unit vectors determined by the receiver-to-satellite position in the kth iteration, , , They represent the three components of the unit vector determined from the receiver position to the j-th satellite position in three-dimensional coordinates, and the subscript N represents the total number of satellites;

[0113] S52: Obtain the observation residual matrix in the kth iteration, the formula is:

[0114] ;

[0115] in, represents the residual matrix of observations in the kth iteration, represents the geometric distance between the nth satellite and the receiver, the superscript s represents the satellite position, and the subscript r represents the receiver position;

[0116] S53: Calculate the incremental vector of the receiver position and local clock error during the k iterations. The formula is:

[0117] ;

[0118] in, represents the incremental vector of the receiver position and local clock error during k iterations, represent Transpose;

[0119] S54: After each iteration, The value of ,when The iteration is terminated when the 2-norm of the vector is less than 0.1. The formula is:

[0120] ;

[0121] in, represents the estimated value in the kth iteration, Represents the initial value in the kth iteration.

[0122] Furthermore, in a method for estimating GNSS receiver observation noise, the S6 comprises the following sub-steps:

[0123] S61: Calculate the current receiver position coordinate error through the receiver position estimate value and the receiver position reference value, and the formula is:

[0124] ;

[0125] in, Represents the current receiver position coordinate error, represents the receiver position estimate, , , Represent the coordinate points of the three-dimensional coordinates of the receiver position estimate, Represents the receiver position reference value, , , They respectively represent the coordinate points of the three-dimensional coordinates of the receiver position reference value.

[0126] Furthermore, a method for estimating GNSS receiver observation noise, said S7 comprises the following sub-steps:

[0127] Determine the current receiver position coordinate error and position coordinate error threshold Size relationship:

[0128] like , then continue to step S8;

[0129] like , then update the position coordinate error threshold for , jump to step S9.

[0130] Furthermore, a method for estimating GNSS receiver observation noise, said S8 comprises the following sub-steps:

[0131] When the random number generated by the random number generator is less than the current inferior solution acceptance condition P, the inferior solution is accepted, otherwise it is not accepted;

[0132] If accepted, go to step S9, if not accepted, go to step S10;

[0133] The S10 includes the following sub-steps:

[0134] During each iteration, the iteration counter is incremented by 1. When the value in the iteration counter exceeds Time exceeds the limit, otherwise it does not exceed the limit;

[0135] If not, go to step S1 and continue iterating; if yes, jump to step S11;

[0136] The S11 comprises the following sub-steps:

[0137] If not, update the inferior solution acceptance condition and reset the number of iterations, and jump to step S1;

[0138] If yes, the optimal estimate of the current observation noise is used as the final observation noise estimation result.

[0139] This scheme adopts a GNSS receiver observation noise estimation method that does not rely on accurate observation noise modeling, avoiding the problem of inaccurate observation noise estimation results when the noise model is inaccurate or incomplete; through a certain search strategy, the observation noise is randomly searched and screened in multi-dimensional space, avoiding the noise modeling process, and obtaining the global observation noise estimate.

[0140] The above shows and describes the basic principles and main features of the present invention and the advantages of the present invention. It should be understood by those skilled in the art that the present invention is not limited to the above embodiments. The above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention to be protected. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A method for estimating GNSS receiver observation noise, characterized in that: The following steps are involved: S1: Obtain the receiver reference coordinate value and observation information; S2: Initialize the optimal estimate of the observation noise, the inferior solution acceptance condition, the position coordinate error threshold, the termination search condition, and the number of iterations; S3: Get the current observation noise estimate; S4: Obtain the observed value correction value; The observed quantity is a pseudorange or a carrier phase; S5: Obtaining an estimated value of the receiver position; S6: Obtain the receiver position coordinate error; S7: Determine the jump step according to the current receiver position coordinate error; S8: Determine whether to accept the inferior solution acceptance condition; S9: assigning the current observation noise estimation value to the observation noise optimal estimation value; S10: Determine whether the number of iterations exceeds the limit; S11: Determine whether to terminate the search for the optimal estimate of the observation noise; The S7 comprises the following sub-steps: Determine the current receiver position coordinate error and position coordinate error threshold Size relationship: like , then continue to step S8; like , then update the position coordinate error threshold for , jump to step S9; The S8 comprises the following sub-steps: When the random number generated by the random number generator is less than the current inferior solution acceptance condition P, the inferior solution is accepted, otherwise it is not accepted; If accepted, go to step S9, if not accepted, go to step S10; The S10 includes the following sub-steps: During each iteration, the iteration counter is incremented by 1. When the value in the iteration counter exceeds Time exceeds the limit, otherwise it does not exceed the limit; If not, go to step S1 and continue iterating; if yes, jump to step S11; The S11 comprises the following sub-steps: If not, update the inferior solution acceptance condition and reset the number of iterations, and jump to step S1; If yes, the optimal estimate of the current observation noise is used as the final observation noise estimation result.

2. A method for estimating GNSS receiver observation noise according to claim 1, characterized in that: The S2 comprises the following sub-steps: S21: Initialize the optimal estimate of the observation noise: Assume that the number of available satellite observations is n, and set the optimal estimate of the observation noise to be estimated All are initialized to 0, the formula is: ； in, represents the optimal estimate of the noise of the observation to be estimated, represents the components corresponding to the optimal estimate of the noise of the 1st to nth observations, T represents the matrix transpose, and n represents the number of available satellite observations; S22: Initialize the inferior solution acceptance condition P, which is a probability parameter in the range of 0.80~0.99; S23: Initialize position coordinate error threshold , is the expected solution accuracy, within the range of 0.001~0.1 meters; S24: Initialize the termination search condition, and terminate when one of condition 1 or condition 2 is met: Condition 1: With the update of the inferior solution acceptance condition P, the current inferior solution acceptance condition P is sufficiently small, ranging from 0.0001 to 0.01; Condition 2: The number of consecutive failures to update the optimal estimate of the observed quantity exceeds a predetermined value, ranging from 50 to 500; S25: Initialization iteration limit: Iteration limit Initialized to a positive integer in the range of 3000~10000.

3. The method for estimating GNSS receiver observation noise according to claim 1, characterized in that: The S3 comprises the following sub-steps: S31: Generate a set of random observation noise disturbance values ​​according to the current random noise disturbance intensity; S32: Superimpose the random observation noise disturbance value on the optimal estimation value of the observation noise to be estimated to obtain the current observation noise estimation value, and the formula is: ； in, represents the optimal estimate of the noise of the observation to be estimated, represents the noise disturbance value of the random observation quantity, Represents the noise estimate of the current observation.

4. The method for estimating GNSS receiver observation noise according to claim 1, characterized in that: The S4 comprises the following sub-steps: S41: Correct the observation quantity using the current observation quantity noise estimation value, the formula is: ； in, represents the correction value component of the i-th observation, represents the raw measurement component of the ith observation, Represents the optimal estimate component of the i-th observation noise; S42: Obtain the observed value correction value, the formula is: ； in, represents the observed value correction, Represents the correction value component of the 1st to nth observations.

5. The method for estimating GNSS receiver observation noise according to claim 1, characterized in that: The S5 comprises the following sub-steps: S51: Calculate the receiver position estimate value through the observation correction value and the satellite position, and use the single point positioning method to iteratively estimate the receiver position. First, obtain the unit vector determined in the kth iteration. The formula is: ； in, represents the coefficient matrix consisting of the unit vectors determined by the receiver-to-satellite position in the kth iteration, , , They represent the three components of the unit vector determined from the receiver position to the j-th satellite position in three-dimensional coordinates, and the subscript N represents the total number of satellites; S52: Obtain the observation residual matrix in the kth iteration, the formula is: ； in, represents the residual matrix of observations in the kth iteration, represents the geometric distance between the nth satellite and the receiver, the superscript s represents the satellite position, and the subscript r represents the receiver position; S53: Calculate the incremental vector of the receiver position and local clock error during the k iterations. The formula is: ； in, represents the incremental vector of the receiver position and local clock error during k iterations, represent Transpose; S54: After each iteration, The value of ,when The iteration is terminated when the 2-norm of the vector is less than 0.

1. The formula is: ； in, represents the estimated value in the kth iteration, Represents the initial value in the kth iteration.

6. The method for estimating GNSS receiver observation noise according to claim 1, characterized in that: The S6 comprises the following sub-steps: S61: Calculate the current receiver position coordinate error through the receiver position estimate value and the receiver position reference value, and the formula is: ； in, Represents the current receiver position coordinate error, represents the receiver position estimate, , , Represent the coordinate points of the three-dimensional coordinates of the receiver position estimate, Represents the receiver position reference value, , , They respectively represent the coordinate points of the three-dimensional coordinates of the receiver position reference value.

Images