A real-time, high-concurrency, high-precision spatiotemporal service method that takes user behavior into account

Abstract

The present invention proposes a real-time, high-concurrency, high-precision spatiotemporal service method that takes into account user behavior, and belongs to the field of navigation and positioning. Based on the dual-frequency pseudorange and carrier phase data of the domestic Beidou and low-orbit satellite ground-based reference stations, combined with the Beidou dual-frequency pseudorange and carrier phase observation data of the onboard low-orbit satellite, the present invention estimates the precise orbits, clock errors, clock speeds and clock drifts of Beidou and low-orbit satellites, as well as the spatial atmospheric delays of the regional ionosphere and troposphere, and establishes relevant models; classifies the resolved enhancement information, and establishes a high-precision spatiotemporal position enhancement information pool that integrates wide-area and regional areas; at the same time, by identifying user behavior or needs, the accuracy requirements of user applications and the types of enhancement information required are judged, and load balancing is used to achieve high-precision positioning services that match massive, multi-class users with high concurrency.

Description

Technical Field

[0001] The present invention relates to the field of navigation and positioning, and specifically refers to a real-time, high-concurrency, and high-precision spatiotemporal service method that takes user behavior into consideration and can meet the real-time, high-concurrency, and high-precision positioning needs of a large number of users. Background Art

[0002] The Global Navigation Satellite System (GNSS) can provide fast and high-precision positioning, navigation and timing (PNT) services to users around the world, playing an important role in multiple military and civilian industries. Nowadays, a large number of mass users have put forward higher requirements for the accuracy and timeliness of location services. How to provide high-concurrency and high-precision positioning services to a large number of users in a wide area and region in real time is an urgent problem to be solved.

[0003] In recent years, many studies have confirmed that low-orbit satellites can help improve the performance of GNSS systems in terms of accuracy, reliability, availability and anti-interference, and the upcoming era of large-scale low-orbit satellite constellations has brought opportunities to solve the shortcomings of traditional satellite positioning. However, there are still shortcomings in terms of precise orbit determination efficiency and providing services to a large number of users, mainly reflected in:

[0004] (1) The precise orbit determination process for GNSS satellites and low-orbit satellites is usually carried out step by step by constellation, which is complex and inefficient.

[0005] (2) The GNSS combined with low-orbit satellite positioning method is rarely combined with the new generation of high-precision, high-dynamic real-time PPP-RTK technology, and a service system has not been formed;

[0006] (3) When facing users with various positioning needs, it is important to focus on intelligently identifying user accuracy requirements and meeting massive and highly concurrent service requests. Summary of the invention

[0007] In order to solve the problems in the prior art, the present invention provides high-concurrency and high-precision positioning needs for massive users in wide areas and regions in real time, and proposes a real-time, high-concurrency and high-precision spatiotemporal service method that takes user behavior into account. The method is based on GNSS / low-orbit satellite observation data, and uses enhanced information such as precise orbits, clock errors, ionosphere, troposphere, and satellite phase deviations for real-time estimation. It also combines user behavior and an adaptive logical load balancing method to achieve high-precision adaptive positioning services that match user types.

[0008] The present invention is achieved through the following technical solutions:

[0009] A real-time, high-concurrency, high-precision spatiotemporal service method taking into account user behavior comprises the following steps:

[0010] (1) Based on the pseudorange and carrier phase data of the base stations of BeiDou and LEO satellite ground-based backbone nodes, combined with the pseudorange and carrier phase observation data of the onboard LEO satellite, a one-step method is used to achieve precise orbit and clock error estimation of BeiDou and LEO satellites;

[0011] (2) Using the precise orbit and clock error data of BeiDou and LEO satellites, combined with the pseudorange and carrier phase data of the reference stations of the regional BeiDou and LEO satellite ground-based backbone nodes, the clock rate and clock drift of the BeiDou and LEO satellite clock error, as well as the spatial atmospheric delay of the regional ionosphere and troposphere are estimated;

[0012] (3) Based on the precise orbit and clock error information of Beidou and low-orbit satellites, as well as clock speed, clock drift, and atmospheric delay of the regional ionosphere and troposphere, a high-precision spatiotemporal position enhancement information pool integrating wide-area and regional fusion is established according to longitude and latitude information;

[0013] (4) Based on the real-time location, speed and needs reported by users, user behavior mining methods are used to identify user application scenarios and accuracy requirements in real time. Load balancing is used in combination with multiple types of differential enhancement information to achieve high-precision positioning services that match massive and multiple types of users with high concurrency.

[0014] Furthermore, the specific method of step (1) is:

[0015] (101) Establish the GNSS dual-frequency observation equations for ground stations and low-orbit satellites:

[0016]

[0017]

[0018]

[0019]

[0020] Where P and L represent pseudorange and carrier phase observation values, respectively; g and leo represent ground stations and low-orbit satellites, respectively; s and j represent GNSS satellites and frequency numbers, respectively. and is the geometric distance between the navigation satellite mass center, the ground station phase center and the low-orbit satellite mass center, c is the speed of light in vacuum, δt g ,δt leo and δt s Respectively represent the station, low-orbit satellite receiver clock error and navigation satellite clock error, λ j represents the signal wavelength of the jth frequency, b g,j , b leo,j and are the pseudorange hardware delays of the station, LEO satellite, and satellite, respectively. g,j, B leo,j and Represent the phase delays of the station, low-orbit satellite, and satellite, respectively. represents the ionospheric delay corresponding to the jth frequency, represents the tropospheric slant delay at the ground station, and represents the integer ambiguity in weeks; and Represent the sum of multipath effect and observation noise on pseudorange and carrier phase respectively;

[0021] (102) An ionosphere-free combined model is constructed to eliminate the influence of ionospheric delay and the combined observation equation is established:

[0022]

[0023]

[0024]

[0025]

[0026] In the formula, the subscript IF represents the deionospheric combination;

[0027] (103) The coordinates of the ground stations are fixed as known values. The a priori Saastamoinen model is used to correct the dry and wet components of the tropospheric delay of the ground stations. The wet component is estimated as a piecewise constant parameter. Then, the combined observation equation is linearized to obtain:

[0028]

[0029]

[0030]

[0031]

[0032] in, and To correct the pseudorange observation value after the known information and model value, and To correct the phase observation value after the known information and model value, the remaining related parameters are calculated as follows:

[0033]

[0034]

[0035]

[0036] α=1 / (μ2-1)

[0037] β=μ2 / (μ2-1)

[0038]

[0039]

[0040] O leo,0 =(x leo,0 ,y leo,0 ,z leo,0 ,v leo,x ,v leo,y, ,v leo,z ,p leo,1 ,p leo,2 ,…,p leo,n ) T

[0041]

[0042]

[0043] In the formula, and are the unit vectors from the ground station and from the low-orbit satellite to the navigation satellite respectively; and They are the state transfer matrices of navigation satellites and low-orbit satellites respectively; Represents the initial state of the navigation satellite, including the position of the navigation satellite at the initial epoch speed And the light pressure parameters to be estimated O leo,0 Represents the state transfer matrix of the low-orbit satellite, including the initial position of the low-orbit satellite (x leo,0 ,y leo,0 ,z leo,0 )、initial velocity (v leo,x ,v leo,y, ,v leo,z ) and force model parameters (p leo,1 ,p leo,2 ,…,p leo,n ); and are the station, low-orbit satellite receiver clock errors and navigation satellite clock errors to be estimated; is the tropospheric parameters to be estimated at the ground station, including the tropospheric wet component Z g , north-south tropospheric gradient parameter G ns and the east-west tropospheric gradient parameter G ew ; Projection function representing the tropospheric wet component; represents the projection function of the tropospheric gradient part; a represents the azimuth;

[0044] For the joint orbit determination in step (1), the parameters to be estimated are expressed as:

[0045]

[0046] (104) Using observation data from wide-area and regional tracking stations and low-orbit satellites, the observation equations in step (103) are solved to accurately determine the orbits and clock errors of the GNSS and low-orbit satellites.

[0047] Furthermore, the specific method of step (2) is:

[0048] (201) Using the satellite clock error sequence δτ estimated in step (1) s (t) Construct a quadratic polynomial model:

[0049] δτ s (t) = δτ s (t0)+a1(t-t0)+a2(t-t0) 2

[0050] Where, t is the epoch time corresponding to the clock error;

[0051] Calculate the clock speed a1 and clock drift a2 based on the quadratic polynomial model;

[0052] (202) The ground station coordinates, satellite orbits, and satellite clock errors are fixed as known values ​​or estimated values ​​in step (1). The dry and wet components of the tropospheric delay of the ground station are corrected using the a priori Saastamoinen model, and the wet component is estimated as a piecewise constant parameter. In addition, the regional ionospheric delay, the receiver-satellite phase deviation, the receiver clock error, and the integer ambiguity parameters are estimated. The undifferenced and non-combined observation equation constructed is:

[0053]

[0054]

[0055] The relevant parameters are calculated as follows:

[0056]

[0057] definition Then we have:

[0058] α=1 / (μ2-1)

[0059] β=μ2 / (μ2-1)

[0060] γ j=α·(μ j +1)·c / λ j

[0061] κ j =α·(μ j +μ2)·c / λ j

[0062] B base,j =B 1,j +γ j b 1,2 -κ j b 1,1

[0063] (203) The observation equations in (202) are solved by using the dual-frequency pseudorange and carrier phase data of the regional BeiDou and low-orbit satellite ground-based backbone node reference stations, thereby accurately determining the phase deviation, tropospheric wet delay, and ionospheric slant delay of the GNSS and low-orbit satellites.

[0064] Furthermore, the specific method of step (3) is:

[0065] (301) Using the tropospheric wet delay value calculated in step (2), a model related to longitude and latitude is established:

[0066]

[0067] In the formula, α i (i=0…4) represents the coefficient of the zenith tropospheric wet delay polynomial; and Δθ r =θ r -θ0 represents the latitude and longitude of the reference station r The central longitude and latitude of the survey area The difference; h r is the geodetic height of reference station r;

[0068] (302) Using the ionospheric slant delay value calculated in step (2), an ionospheric slant delay model is established:

[0069]

[0070] In the formula, and Δθ IPP =θ IPP -θ0 represents the latitude and longitude of the ionospheric thin layer puncture point from the satellite to the receiver and the latitude and longitude of the center of the measurement area The difference between i (i=0…5) represents the ionospheric slant delay polynomial coefficient;

[0071] (303) Based on the satellite precise orbit, clock error, clock speed, clock drift, satellite phase deviation, ionospheric slant delay model parameters, and tropospheric wet delay model parameters, a high-precision spatiotemporal position enhancement information pool is established that integrates wide-area and regional enhancement information, and the information in the enhancement information pool is pushed from the main service center to multiple CDN (content distribution network) service nodes distributed throughout the country;

[0072] Among them, precise orbit, clock error, clock rate and clock drift are used as the first enhancement information to provide traditional PPP services; precise orbit, clock error, clock rate, clock drift and satellite phase deviation are used as the second enhancement information to provide traditional PPP-AR services; precise orbit, clock error, clock rate, clock drift, satellite phase deviation, ionospheric slant delay model parameters and tropospheric wet delay model parameters are used as the third enhancement information to provide PPP-RTK services.

[0073] Furthermore, the specific method of step (4) is:

[0074] (401) The user sends an enhanced information usage request and real-time location to a normal DNS server; the normal DNS server transmits the information to the CDN DNS server for load balancing and allocates a CDN service node for the user to connect to; the user establishes a connection with the CDN service node based on the allocated information for sending and receiving information;

[0075] (402) The user reports his speed time series and enhanced information type requirements to the connected CDN service node; the service node calculates the standard deviation of the user's speed in each three-dimensional direction: Where (Sx, Sy, Sz) are the standard deviations of the velocities in the three directions, n is the number of epochs in the velocity sequence, (v xi ,v yi ,v zi ) is the velocity in each direction corresponding to epoch i, is the mean of the velocity series in each direction;

[0076] (403) Using the standard deviation of the speed in the three directions, calculate the parameters used to determine the user application scenario If the user does not explicitly report the type of enhanced information required, the user application scenario analysis is performed to provide the corresponding enhanced information: Std>1m / s, general positioning requirements, provide the first enhanced information; 0.1m / s<Std≤1m / s, medium and high precision positioning requirements, provide the second enhanced information; Std≤0.1m / s, high precision positioning requirements, provide the third enhanced information;

[0077] (404) After receiving the enhanced information, the user implements PPP, PPP-AR or PPP-RTK positioning application according to the corresponding algorithm.

[0078] Compared with the prior art, the present invention has the following beneficial effects:

[0079] (1) The present invention is based on GNSS / low-orbit satellite dual-frequency observation data, and integrates enhanced information such as precise orbit, clock error, ionosphere, troposphere, and satellite phase deviation for real-time estimation.

[0080] (2) It can use the CDN's DNS server to load balance the user's enhanced information requests, establish a connection between the user and the CDN service node, and efficiently provide the user with real-time and high-precision positioning enhancement information.

[0081] (3) The CDN service node can determine the user's demand for positioning accuracy based on the speed information submitted by the user and the user's motion status information in the absence of clear requirements, and provide the user with the corresponding type of positioning enhancement information.

[0082] (4) It can provide users with enhanced information on three types of positioning services: wide-area and regional real-time PPP, PPP-AR, and PPP-RTK. BRIEF DESCRIPTION OF THE DRAWINGS

[0083] Figure 1 This is a business connection diagram between the service center and the user terminal in an embodiment of the present invention. DETAILED DESCRIPTION

[0084] In order to better illustrate the purpose and advantages of the present invention, the technical solution of the present invention is further described below.

[0085] A real-time, high-concurrency, high-precision spatiotemporal service method that takes user behavior into consideration. This method is based on the pseudorange and carrier phase data of the base stations of the domestic Beidou and low-orbit satellite ground-based backbone nodes, combined with the pseudorange and carrier phase observation data of the onboard low-orbit satellite, to estimate the precise orbit and clock error, clock speed and clock drift of Beidou and low-orbit satellites, as well as the spatial atmospheric delays such as the regional ionosphere and troposphere, and establish relevant models; classify the resolved enhancement information, and establish a high-precision spatiotemporal position enhancement information pool that integrates wide-area and regional areas; at the same time, by identifying user behavior or needs, determine the accuracy requirements of user applications and the type of enhancement information required, and use load balancing to achieve high-precision positioning services that match massive, multi-category users with high concurrency. Specifically, it includes the following steps:

[0086] (1) Based on the pseudorange and carrier phase data of the base stations of the domestic Beidou and low-orbit satellite ground-based backbone nodes, combined with the pseudorange and carrier phase observation data of the onboard low-orbit satellite, the "one-step method" is used to achieve the joint precise orbit and clock error estimation of Beidou and low-orbit satellites;

[0087] (2) Using the precise orbit and clock error data of BeiDou and LEO satellites, combined with the pseudorange and carrier phase data of the reference stations of the regional BeiDou and LEO satellite ground-based backbone nodes, the clock rate and clock drift of the BeiDou and LEO satellite clock error, as well as the atmospheric delays in the regional ionosphere and troposphere, are estimated;

[0088] (3) Based on the precise orbit and clock error information of Beidou and low-orbit satellites, as well as clock speed, clock drift, regional ionosphere and troposphere atmospheric delays, the enhanced information is divided into different regions and layers to establish a high-precision spatiotemporal position enhanced information pool that integrates wide-area and regional information.

[0089] (4) Based on the real-time location, speed and needs reported by users, user behavior mining methods are used to identify user application scenarios and accuracy requirements in real time. Load balancing is used in combination with multiple types of differential enhancement information to achieve high-precision positioning services that match massive and multiple types of users with high concurrency.

[0090] The specific method of step (1) is as follows:

[0091] (101) Establish the GNSS dual-frequency observation equations for ground stations and low-orbit satellites:

[0092]

[0093]

[0094]

[0095]

[0096] Where P and L represent pseudorange and carrier phase observation values, respectively; g and leo represent ground stations and low-orbit satellites, respectively; s and j represent GNSS satellites and frequency numbers, respectively. and is the geometric distance between the navigation satellite mass center, the ground station phase center and the low-orbit satellite mass center, c is the speed of light in vacuum, δt g ,δt leo and δt s Respectively represent the station, low-orbit satellite receiver clock error and navigation satellite clock error, λ j represents the signal wavelength of the jth frequency, b g,j 、b leo,j and are the pseudorange hardware delays of the station, LEO satellite, and satellite, respectively. g,j , B leo,j and Represent the phase delays of the station, low-orbit satellite, and satellite, respectively. represents the ionospheric delay corresponding to the jth frequency, represents the tropospheric slant delay at the ground station, and represents the integer ambiguity in weeks; and Represent the sum of multipath effect and observation noise on pseudorange and carrier phase respectively;

[0097] (102) An ionosphere-free combined model is constructed to eliminate the influence of ionospheric delay and the combined observation equation is established:

[0098]

[0099]

[0100]

[0101]

[0102] In the formula, the subscript IF represents the deionospheric combination;

[0103] (103) The coordinates of the ground stations are fixed as known values. The a priori Saastamoinen model is used to correct the dry and wet components of the tropospheric delay of the ground stations. The wet component is estimated as a piecewise constant parameter. Then, the combined observation equation is linearized to obtain:

[0104]

[0105]

[0106]

[0107]

[0108] in, and To correct the pseudorange observation value after the known information and model value, and To correct the phase observation value after the known information and model value, the remaining related parameters are calculated as follows:

[0109]

[0110]

[0111]

[0112] α=1 / (μ2-1)

[0113] β=μ2 / (μ2-1)

[0114]

[0115]

[0116] O leo,0 =(x leo,0 ,y leo,0 ,z leo,0 ,v leo,x ,v leo,y ,,v leo,z ,p leo,1 ,p leo,2 ,…,p leo,n ) T

[0117]

[0118]

[0119] In the formula, and are the unit vectors from the ground station and from the low-orbit satellite to the navigation satellite respectively; and They are the state transfer matrices of navigation satellites and low-orbit satellites respectively; Represents the initial state of the navigation satellite, including the position of the navigation satellite at the initial epoch speed And the light pressure parameters to be estimated O leo,0 Represents the state transfer matrix of the low-orbit satellite, including the initial position of the low-orbit satellite (x leo,0 ,y leo,0 ,z leo,0 )、initial velocity (v leo,x ,v leo,y, ,v leo,z ) and force model parameters (p leo,1 ,p leo,2 ,…,p leo,n ); and are the station, low-orbit satellite receiver clock errors and navigation satellite clock errors to be estimated; is the tropospheric parameters to be estimated at the ground station, including the tropospheric wet component Z g , north-south tropospheric gradient parameter G ns and the east-west tropospheric gradient parameter G ew ; Projection function representing the tropospheric wet component; represents the projection function of the tropospheric gradient part; a represents the azimuth;

[0120] For the joint orbit determination in step (1), the parameters to be estimated are expressed as:

[0121]

[0122] (104) Using observation data from wide-area and regional tracking stations and low-orbit satellites, the observation equations in step (103) are solved to accurately determine the orbits and clock errors of the GNSS and low-orbit satellites.

[0123] The specific method of step (2) is as follows:

[0124] (201) Using the satellite clock error sequence δτ estimated in step (1) s (t) Construct a quadratic polynomial model:

[0125] δτ s (t) = δτ s (t0)+a1(t-t0)+a2(t-t0) 2

[0126] Where, t is the epoch time corresponding to the clock error;

[0127] Calculate the clock speed a1 and clock drift a2 based on the quadratic polynomial model;

[0128] (202) The ground station coordinates, satellite orbits, and satellite clock errors are fixed as known values ​​or estimated values ​​in step (1). The dry and wet components of the tropospheric delay of the ground station are corrected using the a priori Saastamoinen model, and the wet component is estimated as a piecewise constant parameter. In addition, the regional ionospheric delay, the receiver-satellite phase deviation, the receiver clock error, and the integer ambiguity parameters are estimated. The undifferenced and non-combined observation equation constructed is:

[0129]

[0130]

[0131] The relevant parameters are calculated as follows:

[0132]

[0133] definition Then we have:

[0134] α=1 / (μ2-1)

[0135] β=μ2 / (μ2-1)

[0136] γ j =α·(μ j +1)·c / λ j

[0137] κ j =α·(μj +μ2)·c / λ j

[0138] B base,j =B 1,j +γ j b 1,2 -κ j b 1,1

[0139] (203) The observation equations in (202) are solved by using the dual-frequency pseudorange and carrier phase data of the regional BeiDou and low-orbit satellite ground-based backbone node reference stations, thereby accurately determining the phase deviation, tropospheric wet delay, and ionospheric slant delay of the GNSS and low-orbit satellites.

[0140] The specific method of step (3) is:

[0141] (301) Using the tropospheric wet delay value calculated in step (2), a model related to longitude and latitude is established:

[0142]

[0143] In the formula, α i (i=0…4) represents the coefficient of the zenith tropospheric wet delay polynomial; and Δθ r =θ r -θ0 represents the latitude and longitude of the reference station r The central longitude and latitude of the survey area The difference; h r is the geodetic height of reference station r;

[0144] (302) Using the ionospheric slant delay value calculated in step (2), an ionospheric slant delay model is established:

[0145]

[0146] In the formula, and Δθ IPP =θ IPP -θ0 represents the latitude and longitude of the ionospheric thin layer puncture point from the satellite to the receiver and the latitude and longitude of the center of the measurement area The difference between i (i=0…5) represents the ionospheric slant delay polynomial coefficient;

[0147] (303) Based on the satellite precise orbit, clock error, clock speed, clock drift, satellite phase deviation, ionospheric slant delay model parameters, and tropospheric wet delay model parameters, a high-precision spatiotemporal position enhancement information pool is established that integrates wide-area and regional enhancement information, and the information in the enhancement information pool is pushed from the main service center to multiple CDN (content distribution network) service nodes distributed throughout the country;

[0148] Among them, precise orbit, clock error, clock rate and clock drift are used as the first enhancement information to provide traditional PPP services; precise orbit, clock error, clock rate, clock drift and satellite phase deviation are used as the second enhancement information to provide traditional PPP-AR services; precise orbit, clock error, clock rate, clock drift, satellite phase deviation, ionospheric slant delay model parameters and tropospheric wet delay model parameters are used as the third enhancement information to provide PPP-RTK services.

[0149] The business connection method between the service center and the user end in step (4) is as follows Figure 1 As shown, the specific steps of step (4) are:

[0150] (401) The user sends an enhanced information usage request and real-time location to a normal DNS server; the normal DNS server transmits the information to the CDN DNS server for load balancing and allocates a CDN service node for the user to connect to; the user establishes a connection with the CDN service node based on the allocated information for sending and receiving information;

[0151] (402) The user reports his speed time series and enhanced information type requirements to the connected CDN service node; the service node calculates the standard deviation of the user's speed in each three-dimensional direction: Where (Sx, Sy, Sz) are the standard deviations of the velocities in the three directions, n is the number of epochs in the velocity sequence, (v xi ,v yi ,v zi ) is the velocity in each direction corresponding to epoch i, is the mean of the velocity series in each direction;

[0152] (403) Using the standard deviation of the speed in the three directions, calculate the parameters used to determine the user application scenario If the user does not explicitly report the type of enhanced information required, the user application scenario analysis is performed to provide the corresponding enhanced information: Std>1m / s, general positioning requirements, provide the first enhanced information; 0.1m / s<Std≤1m / s, medium and high precision positioning requirements, provide the second enhanced information; Std≤0.1m / s, high precision positioning requirements, provide the third enhanced information;

[0153] (404) After receiving the enhanced information, the user implements PPP, PPP-AR or PPP-RTK positioning application according to the corresponding algorithm.

Claims

1. A real-time, high-concurrency, high-precision spatiotemporal service method that takes into account user behavior, characterized in that: The following steps are involved: (1) Based on the pseudo-range and carrier phase data of the base stations of BeiDou and LEO satellite ground-based backbone nodes, combined with the pseudo-range and carrier phase observation data of the onboard LEO satellite, a one-step method is used to achieve precise orbit and clock error estimation of BeiDou and LEO satellites. The specific method is as follows: (101) Establish the GNSS dual-frequency observation equations for ground stations and low-orbit satellites: Where P and L represent pseudorange and carrier phase observation values, respectively; g and leo represent ground stations and low-orbit satellites, respectively; s and j represent GNSS satellites and frequency numbers, respectively. and is the geometric distance between the navigation satellite mass center, the ground station phase center and the low-orbit satellite mass center, c is the speed of light in vacuum, δt g ,δt leo and δt s Respectively represent the station, low-orbit satellite receiver clock error and navigation satellite clock error, λ j represents the signal wavelength of the jth frequency, b g,j , b leo,j and are the pseudorange hardware delays of the station, LEO satellite, and satellite, respectively. g,j , B leo,j and Represent the phase delays of the station, low-orbit satellite, and satellite, respectively. represents the ionospheric delay corresponding to the jth frequency, represents the tropospheric slant delay at the ground station, and represents the integer ambiguity in weeks; and Represent the sum of multipath effect and observation noise on pseudorange and carrier phase respectively; (102) An ionosphere-free combined model is constructed to eliminate the influence of ionospheric delay and the combined observation equation is established: In the formula, the subscript IF represents the deionospheric combination; (103) The coordinates of the ground stations are fixed as known values. The a priori Saastamoinen model is used to correct the dry and wet components of the tropospheric delay of the ground stations. The wet component is estimated as a piecewise constant parameter. Then, the combined observation equation is linearized to obtain: in, and To correct the pseudorange observation value after the known information and model value, and To correct the phase observation value after the known information and model value, the remaining related parameters are calculated as follows: a=1 / (μ2-1) β=m2 / (μ2-1) O leo,0 =(x leo,0 ,y leo,0 ,z leo,0 ,v leo,x ,v leo,y ,,v leo,z ,p leo,1 ,p leo,2 ,…,p leo,n ) T In the formula, and are the unit vectors from the ground station and from the low-orbit satellite to the navigation satellite respectively; and They are the state transfer matrices of navigation satellites and low-orbit satellites respectively; Represents the initial state of the navigation satellite, including the position of the navigation satellite at the initial epoch speed And the light pressure parameters to be estimated O leo,0 Represents the state transfer matrix of the low-orbit satellite, including the initial position of the low-orbit satellite (x leo,0 ,y leo,0 ,z leo,0 )、initial velocity (v leo,x ,v leo,y ,,v leo,z ) and force model parameters (p leo,1 ,p leo,2 ,…,p leo,n ); and are the station, low-orbit satellite receiver clock errors and navigation satellite clock errors to be estimated; is the tropospheric parameters to be estimated at the ground station, including the tropospheric wet component Z g , north-south tropospheric gradient parameter G ns and the east-west tropospheric gradient parameter G ew ; Projection function representing the tropospheric wet component; represents the projection function of the tropospheric gradient part; a represents the azimuth; For the joint orbit determination in step (1), the parameters to be estimated are expressed as: (104) using observation data from wide-area and regional tracking stations and low-orbit satellites, solving the observation equations in step (103) to accurately determine the orbits and clock errors of the GNSS and low-orbit satellites; (2) Using the precise orbit and clock error data of BeiDou and LEO satellites, combined with the pseudorange and carrier phase data of the reference stations of the regional BeiDou and LEO satellite ground-based backbone nodes, the clock rate and clock drift of the BeiDou and LEO satellite clock error, as well as the spatial atmospheric delay of the regional ionosphere and troposphere are estimated; the specific method is as follows: (201) Using the satellite clock error sequence dt estimated in step (1) s (t) Construct a quadratic polynomial model: dt s (t)=dt s (t0)+a1(t-t0)+a2(t-t0) 2 Where, t is the epoch time corresponding to the clock error; Calculate the clock speed a1 and clock drift a2 based on the quadratic polynomial model; (202) The ground station coordinates, satellite orbits, and satellite clock errors are fixed as known values ​​or estimated values ​​in step (1). The dry and wet components of the tropospheric delay of the ground station are corrected using the a priori Saastamoinen model, and the wet component is estimated as a piecewise constant parameter. In addition, the regional ionospheric delay, the receiver-satellite phase deviation, the receiver clock error, and the integer ambiguity parameters are estimated. The undifferenced and non-combined observation equation constructed is: The relevant parameters are calculated as follows: definition Then we have: a=1 / (μ2-1) β=m2 / (μ2-1) c j =a×(μ j +1)×c / λ j k j =a×(μ j +m2)×c / λ j B base,j =B 1,j +g j ×b 1,2 -k j ×b 1,1 (203) using the dual-frequency pseudorange and carrier phase data of the regional BeiDou and low-orbit satellite ground-based backbone node reference stations, solving the observation equations in (202), thereby accurately determining the phase deviation, tropospheric wet delay, and ionospheric slant delay of the GNSS and low-orbit satellites; (3) Based on the precise orbit and clock error information of Beidou and low-orbit satellites, as well as clock speed, clock drift, and atmospheric delay of the regional ionosphere and troposphere, a high-precision spatiotemporal position enhancement information pool integrating wide-area and regional fusion is established according to longitude and latitude information. The specific methods are as follows: (301) Using the tropospheric wet delay value calculated in step (2), a model related to longitude and latitude is established: In the formula, α i (i=0…4) represents the coefficient of the zenith tropospheric wet delay polynomial; and Dq r =q r -q0 represents the latitude and longitude of the reference station r The central longitude and latitude of the survey area The difference; h r is the geodetic height of reference station r; (302) Using the ionospheric slant delay value calculated in step (2), an ionospheric slant delay model is established: In the formula, and Dq IPP =q IPP -q0 represents the latitude and longitude of the ionosphere puncture point from the satellite to the receiver and the latitude and longitude of the center of the measurement area The difference between i (i=0…5) represents the ionospheric slant delay polynomial coefficient; (303) Based on the satellite precise orbit, clock error, clock speed, clock drift, satellite phase deviation, ionospheric slant delay model parameters, and tropospheric wet delay model parameters, a high-precision spatiotemporal position enhancement information pool is established that integrates wide-area and regional enhancement information, and the information in the enhancement information pool is pushed from the main service center to multiple CDN content distribution network service nodes distributed throughout the country; Among them, precise orbit, clock error, clock speed and clock drift are used as the first enhancement information to provide traditional PPP services; precise orbit, clock error, clock speed, clock drift and satellite phase deviation are used as the second enhancement information to provide traditional PPP-AR services; precise orbit, clock error, clock speed, clock drift, satellite phase deviation, ionospheric slant delay model parameters and tropospheric wet delay model parameters are used as the third enhancement information to provide PPP-RTK services; (4) Based on the real-time location, speed and needs reported by users, user behavior mining methods are used to identify user application scenarios and accuracy requirements in real time. Load balancing is used in combination with multiple types of differential enhancement information to achieve high-precision positioning services that match massive and multi-type users with high concurrency. The specific methods are as follows: (401) The user sends an enhanced information usage request and real-time location to a normal DNS server; the normal DNS server transmits the information to the CDN DNS server for load balancing and allocates a CDN service node for the user to connect to; the user establishes a connection with the CDN service node based on the allocated information for sending and receiving information; (402) The user reports his speed time series and enhanced information type requirements to the connected CDN service node; the service node calculates the standard deviation of the user's speed in each three-dimensional direction: and Where (Sx, Sy, Sz) are the standard deviations of the velocities in the three directions, n is the number of epochs in the velocity sequence, (v xi ,v yi ,v zi ) is the velocity in each direction corresponding to epoch i, is the mean of the velocity series in each direction; (403) Calculate the parameters for judging the user application scenario by using the standard deviations of the velocities in three directions If the user does not explicitly report the requirement for the enhanced information type, perform user application scenario analysis to provide corresponding enhanced information: when Std > 1 m / s, for general positioning requirements, provide the first enhanced information; when 0.1 m / s < Std ≤ 1 m / s, for medium-high-precision positioning requirements, provide the second enhanced information; when Std ≤ 0.1 m / s, for high-precision positioning requirements, provide the third enhanced information; (404) After receiving the enhanced information, the user implements the PPP, PPP-AR or PPP-RTK positioning application according to the corresponding algorithm.

Images