Disclosure of Invention
In view of the above, the invention provides a cycle slip detection and repair method which is real-time, is based on a low-cost receiver and is suitable for a single-frequency terminal in a complex environment.
The technical scheme of the invention is realized as follows: a cycle slip detection and repair method suitable for a single frequency terminal in a complex environment comprises the following steps:
s1: establishing a single-frequency non-difference carrier phase observation equation;
S2: performing inter-epoch difference on a single-frequency non-difference carrier phase observation equation;
s3: solving unknown parameters in a single-frequency non-difference carrier phase observation equation;
S4: carrying out three-time difference among satellites, measuring stations and epochs on a single-frequency non-difference carrier phase observation equation, and establishing a three-difference observation equation;
s5: substituting the unknown parameters calculated in the third step into a three-difference observation equation, and detecting and repairing cycle slip according to the difference result between the carrier phase epochs of the current epoch and the previous epoch.
On the basis of the above technical solution, preferably, in step S1, the single-frequency non-differential carrier phase observation equation is established by making the single-frequency non-differential carrier phase observation equation be: Wherein/> Representing carrier phase observations; λ represents a wavelength; p denotes the geometric distance of the satellite to the receiver; δt r and δt s represent receiver clock difference and satellite clock difference, respectively; n represents integer ambiguity; i represents ionospheric delay error; t represents tropospheric delay error; epsilon L represents measurement noise; c is the speed of light.
Preferably, in step S2, the inter-epoch difference of the single-frequency non-differential carrier phase observation equation is expressed as: Wherein Δ represents an inter-epoch single difference calculation factor; /(I) And/>Carrier phase observations representing a current epoch t j and a previous epoch t j-1, respectively; Δi and Δt are ionosphere error difference and troposphere delay error difference, which constitute an atmospheric delay error parameter; delta t r and delta t s are clock error parameters; the difference Δp=p (t j)-P(tj-1) between the satellite-to-receiver distances of the current epoch t j and the previous epoch t j-1, the satellite-to-receiver distance P (t j)=e(tj)*[rs(tj)-ru(tj) at the current epoch t j, the satellite-to-receiver distance P (t j-1)=e(tj-1)*[rs(tj-1)-ru(tj-1) at the previous epoch t j-1, where e (t j) is the directional cosine of epoch t j, e (t j-1) is the directional cosine of epoch t j-1, r s is the satellite geometry position, r u is the receiver geometry position, let r s(tj)=ru(tj-1)+Δru,Δru be the position change parameter, and the result is: /(I)Δd represents the difference in adjacent epoch satellite positions and Δg represents the difference in adjacent epoch receiver positions; ignoring the atmospheric delay error parameters Δi and Δt, satellite clock difference/>The calculation formula of (2) is as follows: /(I)Where Δδt r is the difference in receiver clock difference, Δε L is the difference in measured noise, and the unknown parameters are Δr u and Δδt r.
Preferably, in step S3, the solving of the unknown parameters in the single-frequency non-differential carrier-phase observation equation is to decompose the position change parameter Δr u intoLet the number of satellites observed by a satellite navigation system be m, solve the unknown parameters by the following formula: y=hx+ε L, where/>Satellite clock differences for m satellites; /(I)Is an unknown parameter matrix; /(I)Is a coefficient matrix,/>The direction cosine of each satellite in the x-axis direction of epoch t j; /(I)The direction cosine of each satellite in the y-axis direction of element t j; /(I)The direction cosine of each satellite in the z-axis direction of epoch t j; solving a position parameter matrix x, x= (H TpH)-1HT py, where p is a weight matrix,/>, by a least squares methodThe elements on the diagonal of the weight matrix are calculated/>, using the following methodWhere i=1, 2,3,..m,/>Is a constant variance value; C/N 0i is the signal to noise ratio; el i is the satellite altitude.
Preferably, in step S4, the three-time difference between satellites, between stations and between epochs is performed on the single-frequency non-differential carrier phase observation equation, and the three-time difference observation equation is established by establishing the following equation: Wherein/> A double difference calculation factor representing the difference between satellites or between measuring stations, and when the difference between the satellites is carried out, a satellite is selected as a reference satellite; /(I)And/>A double difference of the geometric distances of satellites of adjacent epochs to the receiver; and/> Double differences of measurement noise for adjacent epochs; neglecting the influence of observation noise, linearizing the above equation to obtain the following linearization three-difference observation equation:
preferably, the selecting a satellite as the reference satellite is selecting a satellite with the largest satellite altitude angle as the reference satellite in the satellite navigation system observation.
Preferably, in step S5, the unknown parameters calculated in step S3 are substituted into the three-difference observation equation, and cycle slip detection and repair are performed according to the difference between the carrier phase epochs of the current epoch and the previous epoch, which is to use the position change parameters calculated in step S3Substituting the sum of the left term and the right term of the formula of the linearized three-difference observation equation obtained in the step S4 into the linearized three-difference observation equation, respectively solving the sum of the left term and the right term of the formula of the linearized three-difference observation equation, and considering that the satellite observed by the current epoch has cycle slip if the difference epsilon between the left term and the right term of the formula of the linearized three-difference observation equation is larger than a set threshold value, further judging the size of the cycle slip, and repairing the cycle slip.
Preferably, step S5 further includes a reference star cycle slip detection step, that is, each linearized three-difference observation equation obtained in step S4 includes information of two satellites, by calculating a difference value of a sum of a term on a left side of a formula and a term on a right side of the formula of each linearized three-difference observation equation, whether each difference value has a deviation of the same degree with respect to a set threshold value is observed, if so, it indicates that a cycle slip occurs in the reference star, the reference star needs to be removed, one satellite is reselected as the reference star, the step S4 is repeated to construct a linearized three-difference observation equation to perform a judgment of the cycle slip of the reference star again, and after determining that the cycle slip does not occur in the reference star, a judgment of the cycle slip is performed.
Preferably, in step S5, it is considered that the satellite observed in the current epoch has a cycle slip, and the size of the cycle slip is further determined, and when the difference value between the sum of the term on the left side of the formula and the term on the right side of the formula of the linearized three-difference observation equation is greater than the set threshold, the difference value between the sum of the term on the left side of the formula and the term on the right side of the formula of the linearized three-difference observation equation and the wavelength of the corresponding satellite are rounded to obtain the size of the cycle slip: lambda 0 is the wavelength corresponding to the satellite frequency; ROUND is a rounding operation; Δl is the size of the cycle slip.
Preferably, the observation time interval of adjacent epochs is not more than 10 seconds.
The cycle slip detection and repair method suitable for the single-frequency terminal in the complex environment has the following beneficial effects compared with the prior art:
(1) According to the application, only single-frequency cycle slip detection is adopted, only the observation data of the current epoch and the previous epoch are needed, a single-phase carrier phase equation is formed by constructing a differential equation between epochs, the coordinate correction parameters are calculated through a least square method, then the coordinate correction parameters are substituted into a three-difference observation equation, and the satellite with cycle slip and the corresponding frequency thereof can be detected by observing the difference value of the left side and the right side of the three-difference observation equation; the method is simple in algorithm, can detect in real time, and is suitable for real-time navigation and positioning of the low-cost receiver in a complex environment.
Detailed Description
The following description of the embodiments of the present invention will clearly and fully describe the technical aspects of the embodiments of the present invention, and it is apparent that the described embodiments are only some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the present invention without making any inventive effort, are intended to fall within the scope of the present invention.
As shown in fig. 1, the invention provides a cycle slip detection and repair method suitable for a single-frequency terminal in a complex environment, which comprises the following steps:
S1: and establishing a single-frequency non-difference carrier phase observation equation.
Specifically, let the single-frequency non-differential carrier phase observation equation be: Equation 1; wherein/> Representing carrier phase observations; λ represents a satellite wavelength; p represents the geometric distance from the satellite to the receiver, let the geometric position of the ith satellite be (x i,yi,zi), and the geometric position of the receiver be (x, y, z), then/>For dynamic relative positioning, the receiver position of the current epoch may use the result calculated by single point or differential pseudorange positioning, and the receiver position of the last epoch may use the floating solution or fixed solution result obtained from the last epoch. δt r and δt s represent receiver clock difference and satellite clock difference, respectively; n represents integer ambiguity; i represents ionospheric delay error; t represents tropospheric delay error; epsilon L represents measurement noise; c is the speed of light.
S2: and carrying out inter-epoch difference on the single-frequency non-difference carrier phase observation equation.
Preferably, in step S2, the inter-epoch difference of the single-frequency non-differential carrier phase observation equation is expressed as: Equation 2; wherein Δ represents an inter-epoch single difference calculation factor; /(I) And/>Carrier phase observations representing a current epoch t j and a previous epoch t j-1, respectively; Δi and Δt are ionosphere error difference and troposphere delay error difference, which constitute an atmospheric delay error parameter; delta t r and delta t s are clock error parameters; the difference Δp=p (t j)-P(tj-1) between the satellite-to-receiver distances of the current epoch t j and the previous epoch t j-1, the satellite-to-receiver distance P (t j)=e(tj)*[rs(tj)-ru(tj) at the current epoch t j, the satellite-to-receiver distance P (t j-1)=e(tj-1)*[rs(tj-1)-ru(tj-1) at the previous epoch t j-1, where e (t j) is the directional cosine of epoch t j, e (t j-1) is the directional cosine of epoch t j-1, r s is the satellite geometry position, r u is the receiver geometry position, let r s(tj)=ru(tj-1)+Δru,Δru be the position change parameter, and the result is: /(I)Equation 3; Δd represents the difference in adjacent epoch satellite positions and Δg represents the difference in adjacent epoch receiver positions; ignoring the atmospheric delay error parameters Δi and Δt, satellite clock difference/>The calculation formula of (2) is as follows: /(I)Equation 4; where Δδt r is the difference in receiver clock difference, Δε L is the difference in measured noise, and the unknown parameters are Δr u and Δδt r.
The direction cosine e (t j) here can be decomposed into [ e x(tj),ey(tj),ez(tj) ], and the coordinates r s(tj of the epoch t j satellite can be decomposed intoSimilar decomposition can be performed by the same principles e (t j-1) and r s(tj-1), and the content of the above formula 3 can be rewritten as follows:
in equation 4, the left side of the equation Representing what can be calculated by known information, the right side of the formula (t j)*Δru]+cΔδtr+ΔεL represents the part of the parameter to be estimated, in the actual calculation process, the influence of the measured noise parameter delta epsilon L is often ignored, and only the first half position change parameter delta r u and the delta t r of the clock error parameter are estimated.
S3: and solving unknown parameters in the single-frequency non-difference carrier phase observation equation.
Specifically, the position change parameter Deltar u is decomposed intoLet the number of satellites observed by a satellite navigation system be m, solve the unknown parameters by the following formula: y=hx+ε L, where/>Satellite clock differences for m satellites; /(I)Is an unknown parameter matrix; /(I)As a matrix of coefficients,The direction cosine of each satellite in the x-axis direction of epoch t j; /(I)The direction cosine of each satellite in the y-axis direction of element t j; /(I)The direction cosine of each satellite in the z-axis direction of epoch t j; solving a position parameter matrix x, x= (H TpH)-1HT py, where p is a weight matrix,/>, by a least squares methodThe elements on the diagonal of the weight matrix p are calculated/>, using the following methodWhere i=1, 2,3,..m,/>Is a constant variance value; C/N 0i is the signal to noise ratio; el i is the satellite altitude.
To obtain the coefficient matrix H, the direction cosine e (t j) needs to be linearized to obtain the non-1 elements in the matrix, respectively. Here, the direction cosine e (t j) is taken as an example for explanation:
[ x i(tj),yi(tj),zi(tj) ] represents the geometric position of the satellite at time t j; [ x (t j),y(tj),z(tj) ] represents the geometrical position of the receiver at time t j. The direction cosine e (t j-1) can be linearized in the same way.
The weight matrix p is determined based on the carrier noise variance, the signal to noise ratio C/N 0i and the satellite altitude angle el i Is a constant variance value, which can be empirically set; the signal to noise ratio C/N 0i can be obtained by observing a file, and the signal to noise ratio of a general satellite is 40dbHz-50dbHz; the satellite altitude angle el i is obtained through calculation, in actual calculation, threshold value judgment is needed for the altitude angle, the cut-off altitude angle is generally set to be 15 degrees, satellites smaller than the cut-off altitude angle are eliminated, and meanwhile follow-up resolving processes are not involved.
Before the position parameter matrix x is solved by the least square method and the differential solution between epochs is realized, a judgment is needed to be carried out on the observation time intervals of the front epoch and the rear epoch, and continuous observation is the premise of cycle slip judgment; when the observation time interval is set to be more than 10 seconds, the data is considered to be interrupted, at this time, the epoch is not judged to be cycle-slip any more, the current epoch observation value is stored, and the next epoch is calculated. I.e. the maximum interval between adjacent epochs is not more than 10 seconds.
In order to ensure the precision of the coordinate correction, before the inter-epoch difference solution calculation is realized, large cycle slip detection is needed in advance, one cycle slip detection can be carried out in advance through a TurboEdit method, the satellite with the excessive cycle slip is removed, then the residual satellite observation value is subjected to a least square method to solve the position parameter matrix x, and then the cycle slip repair is carried out on all satellites with cycle slip in the subsequent steps.
S4: and carrying out three-time difference among satellites, measuring stations and epochs on the single-frequency non-difference carrier phase observation equation, and establishing a three-difference observation equation.
The three differences represent three differences among satellites, measuring stations and epochs, and most of parameters such as satellite clock differences, receiver clock differences, atmospheric delay correction parameters, whole-cycle ambiguity parameters and the like can be eliminated through the three differences, and only position parameters and measurement noise remain in the observation equation. Specifically, the following three-difference observation equation is established: Wherein/> A double difference calculation factor representing the difference between satellites or between measuring stations, and when the difference between the satellites is carried out, a satellite is selected as a reference satellite; and/> A double difference of the geometric distances of satellites of adjacent epochs to the receiver; /(I)And/>Double differences of measurement noise for adjacent epochs; neglecting the influence of observation noise, linearizing the above equation to obtain the following linearization three-difference observation equation:
s5: substituting the unknown parameters calculated in the third step into a three-difference observation equation, and detecting and repairing cycle slip according to the difference result between the carrier phase epochs of the current epoch and the previous epoch.
And S4, each linearization three-difference observation equation obtained in the step contains information of two satellites, whether the difference values have the same degree of deviation relative to a set threshold value or not is observed by calculating the difference value of the sum of the left term of the formula and the right term of the formula of each linearization three-difference observation equation, if so, the reference star is represented to have cycle slip, the reference star is needed to be removed, one satellite is selected again as the reference star, the step S4 is repeated to construct the linearization three-difference observation equation to judge the cycle slip of the reference star again, and the cycle slip size is judged after the fact that the reference star has no cycle slip is determined.
The cycle slip detection process is to calculate the position change parameters in step S3Substituting the sum of the left term and the right term of the formula of the linearized three-difference observation equation obtained in the step S4 into the linearized three-difference observation equation, respectively solving the sum of the left term and the right term of the formula of the linearized three-difference observation equation, and considering that the satellite observed by the current epoch has cycle slip if the difference epsilon between the left term and the right term of the formula of the linearized three-difference observation equation is larger than a set threshold value, further judging the size of the cycle slip, and repairing the cycle slip.
After confirming that the satellite observed in the current epoch has cycle slip, further judging the size of the cycle slip, and when the difference value of the sum of the left term of the formula and the right term of the formula of the linearized three-difference observation equation is larger than a set threshold value, rounding the difference value of the sum of the left term of the formula and the right term of the formula of the linearized three-difference observation equation and the wavelength of the corresponding satellite to obtain the size of the cycle slip: lambda 0 is the wavelength corresponding to the satellite frequency; ROUND is a rounding operation; Δl is the size of the cycle slip.
And after the size of the cycle slip is obtained, repairing the cycle slip. This section of content can be correspondingly repaired by using a conventional cycle slip repair model. If the satellite with cycle slip is determined to be in the epoch K, returning to the original carrier phase three-difference observed value, and if the satellite with cycle slip is a non-standard satellite, rounding and inverting the K-2 th three-difference observed value of the satellite with cycle slip to obtain a correct cycle slip repair value.
The foregoing description of the preferred embodiments of the invention is not intended to be limiting, but rather is intended to cover all modifications, equivalents, alternatives, and improvements that fall within the spirit and scope of the invention.