JP4977053B2 - Ionospheric electron density distribution estimation system and positioning system - Google Patents

Description

  The present invention relates to an ionospheric electron density distribution estimation system and a positioning system, and more particularly to a technique for estimating the total number of electrons based on a plurality of frequency carrier phases and correcting a pseudorange for positioning.

  The pseudorange difference of multiple frequency signals broadcast from satellites such as GPS (Global Positioning System), Galileo, or Quasi Zenith Satellites System (QZSS) is the number of electrons on the propagation path of radio waves. It is known to be related to When estimating the total number of electrons on the propagation path of radio waves (propagation delay amount due to passage through the ionosphere), a method of using a pseudo-range (code distance) difference depending on transmission / reception times of multiple frequency signals, and carriers of multiple frequencies A method using a difference in phase pseudorange (phase distance) is known.

The method of using the pseudo-range difference depending on the transmission / reception time has a large error, while there is a problem that a bias is generated due to an indeterminate value of the phase inherent in the carrier phase using the carrier phase pseudo-range. Further, the carrier phase may cause a data jump called cycle slip. When a simple Kalman filter is used, there is a problem that a transient phenomenon in which the estimated value greatly oscillates after the occurrence of cycle slip occurs, and data until convergence is not available.
A. Komjathy: 'Global Ionospheric Total Electron Content Mapping Using the Global Positioning System', UNB, Technical Report No.188, Sep.1997. Pratap Misra, PerEnge, “GLOBAL POSITIONIG SYSTEM Signals, Measurements, and Performance.” Ganga-Jamuna Press, 2001 New edition of GPS -Precision positioning system using artificial satellite- ", Geodetic Society of Japan, September, 1989. Kiyoshi Nishiyama: “Kalman filter solved on a personal computer”, Maruzen Corporation, 1999. Mohinder S. Grewal, et.al. “Kalman Filtering Theory and Practice Using MATLAB Second Edition”, John Wiley & Sons. Inc, 2001. “Navstar GPS Space Segment / Navigation User Interfaces', ICD-GPS-200, Rev. C, 12. Apr. 2000

  By the way, the error in the pseudo distance due to the carrier phase of the signal broadcast from the satellite is very small. However, there is uncertainty due to the phase cycle, and it is not possible to simply use the pseudorange due to the carrier phase. On the other hand, the pseudo distance generated from the transmission / reception time has no uncertainty due to the phase cycle, but the measurement error is very large.

  Therefore, a method for removing uncertainty due to the phase cycle using a pseudorange has been proposed (see Non-Patent Document 1). The pseudo-range based on the carrier phase has a property that the uncertain value changes when the satellite tracking is interrupted, and is called “cycle slip”. Due to this cycle slip, the uncertainty due to the phase cycle cannot be removed simply by the method disclosed in Non-Patent Document 1.

  The present invention has been made to solve the above-mentioned problems, and its problem is to eliminate the uncertainty included in the multi-carrier phase pseudorange difference and to suppress the transient phenomenon period that occurs in the filter due to cycle slip. Thus, it is an object of the present invention to provide an ionospheric electron density distribution estimation system that accurately estimates the total number of electrons in the ionosphere of the propagation path and uses it for estimating the ionospheric electron density distribution. It is another object of the present invention to provide a positioning system that can use the estimated degree of electron for positioning and remove an ionospheric propagation delay error at a pseudo distance.

In order to solve the above problem, an ionospheric electron density distribution estimation system according to a first aspect of the present invention is a cycle using a multi-phase distance difference, a covariance matrix of a modified Kalman filter based on a velocity component of this distance difference, and a satellite sample time difference. A data processing period that specifies a data collection period that is continued by determining the presence or absence of slip and updates a covariance matrix of the modified Kalman filter when the cycle slip occurs, and a data collection period that is specified by the filter processing section The ionosphere total electron number estimation unit that converts the multiple carrier phase pseudorange difference after removing the uncertain value related to the carrier phase into the ionosphere electron number, and the ionosphere electron number obtained by the ionosphere total electron number estimation unit And an electron density estimation processing unit for estimating the electron density distribution of the ionosphere.

The second invention is the ionospheric electron density distribution estimation system of the first invention, the filter processing unit, the difference between the value and the observed value estimated by the deformation Kalman filter dynamically estimated to have dispersed in the modified Kalman filter The presence or absence of cycle slip is determined based on the value and the data collection time difference . When cycle slip occurs, the slip amount is input and the modified Kalman filter is operated to specify the continuous data collection period. And

  The positioning system of the third invention is a process of subtracting the ionosphere propagation delay amount corresponding to the ionosphere electron number obtained by the ionosphere total electron number estimation unit of the first or second invention from the pseudo distance obtained by the positioning. A pseudo-range correcting unit that performs the positioning, and a positioning processing unit that performs positioning using the corrected pseudo-range sent from the pseudo-range correcting unit and the satellite position obtained from the satellite navigation calendar. .

  According to a fourth aspect of the present invention, in the positioning system of the third aspect, the pseudo-range correction unit further corrects a delay amount caused by the troposphere.

  According to the present invention, the uncertainty included in the multi-carrier phase pseudorange difference can be removed and the transient phenomenon period caused by cycle slip can be suppressed, so that the total number of propagation path ionosphere electrons can be accurately estimated and ionosphere electrons can be accurately estimated. The density distribution can be estimated.

  In addition, the estimated electron degree can be used for positioning, and an ionospheric propagation delay error at a pseudo distance can be removed.

  Further, by using the pseudo distance based on the carrier phase, the measurement accuracy is improved 10 times or more than the pseudo distance generated from the transmission / reception time. As a result, it is possible to improve the measurement accuracy of the total number of electrons in the radio wave passage path by about 10 times by calculating the multiple carry phase pseudorange difference. Thereby, ionospheric electron density observation accuracy can be improved, and HF band communication using reflection by the ionosphere can be performed more effectively.

  In addition, the propagation delay amount of radio waves generated by passing through the area where electrons exist is a cause of positioning error using navigation satellite radio waves. Since the quantity can be estimated accurately, the positioning accuracy can be improved by applying the correction of the estimated value to the observation pseudorange used for positioning.

  Embodiments of an ionospheric electron density distribution estimation system and positioning system according to the present invention will be described below in detail with reference to the drawings.

  FIG. 1 is a diagram showing a configuration of a positioning system including an ionosphere electron density distribution estimation system according to Embodiment 1 of the present invention. This positioning system includes a satellite signal observation system 1, a satellite signal processing system 2, an internet data processing system 3, and a data server system 4. In addition to these, although not shown, a satellite and an external satellite data observation system, for example, the Geographical Survey Institute (GEONET) (GPS Earth Observation Network System) is connected to the outside.

  The satellite signal observation system 1 includes an antenna 11 and a receiver 12. The antenna 11 receives radio waves of a plurality of frequencies broadcast from satellites such as GPS, Galileo, and Quasi-Zenith Satellite System (QZSS), converts them into electrical signals, and sends them to the receiver 12. The receiver 12 performs predetermined processing on the signal sent from the antenna 11 and sends it to the satellite signal processing system 2.

  The satellite signal processing system 2 includes a signal processing device 20 for processing satellite navigation calendar (ephemeris almanac) data, transmission / reception times, and carrier phase pseudoranges included in a signal transmitted from the receiver 12. Has been. As shown in FIG. 2, the signal processing device 20 includes a filter processing unit 21, an ionosphere total electron number estimation unit 22, an electron density estimation processing unit 22 a, a pseudo distance correction unit 23, and a positioning processing unit 24.

  The filter processing unit 21, the ionosphere total electron number estimation unit 22, and the electron density estimation processing unit 22a constitute an ionosphere electron density distribution estimation system.

  The filter processing unit 21 specifies a continuous data collection period by determining the presence or absence of cycle slip using a modified Kalman filter (details will be described later). The processing result in the filter processing unit 21 is sent to the ionosphere total electron number estimation unit 22.

  The ionosphere total electron number estimation unit 22 removes an uncertain value related to the carrier phase by the method disclosed in Non-Patent Document 1 for a period of time during which data is continuously collected by the filter processing unit 21, and multi-carrier phase simulation. Convert the distance difference into the number of ionospheric electrons. The processing result in the ionosphere total electron number estimation unit 22 is sent to the electron density estimation processing unit 22 a and the pseudo distance correction unit 23.

  The electron density estimation processing unit 22a uses the ionosphere electron number estimated by the ionosphere total electron number estimation unit 22 to estimate the electron density distribution of the ionosphere.

  The pseudorange correction unit 23 obtains the ionospheric propagation delay amount corresponding to the multicarrier phase pseudorange difference after removing the indeterminate value, that is, the ionosphere electron number sent from the ionosphere total electron number estimation unit 22 by positioning. A process of subtracting from the obtained pseudo distance is performed (details will be described later). The pseudo distance corrected by the pseudo distance correcting unit 23 is sent to the positioning processing unit 24.

  The positioning processing unit 24 performs positioning using the corrected pseudo distance sent from the pseudo distance correcting unit 23 and the satellite position obtained from the satellite navigation calendar.

  The Internet data processing system 3 is the ionosphere-related information published, the GPS observation data (GEONET data) published by the Geospatial Information Authority of Japan, and the IGS (International GPS Service for Geodynamics) that is publishing international observation results Collect data etc. via the Internet. Data collected by the Internet data processing system 3 is transmitted to the data server system 4 via the LAN and stored in the data collection server 41. The data collection server 41 also stores data received from satellites and processing results in the satellite signal processing system 2.

  The Internet data processing system 3 can be configured to include a firewall using a switching hub or router in consideration of security. This switching hub or router is not essential for the Internet data processing system 3, but is an option.

  Next, the operation of the positioning system including the ionospheric electron density distribution estimation system according to Embodiment 1 of the present invention configured as described above will be described.

(1) Basics The pseudo distance (code distance, pseudo range) based on the transmission / reception times of the two-frequency signals L1 and L2 broadcast from the satellite and the pseudo distance (phase distance) based on the carrier phase are expressed by the following equation (1.1) And (1.2).

The relationship between the result of subtracting the observed values of the two frequencies and the total electron content (TEC) is expressed by the following equations (1.3) and (1.4).

here,
ρ: code distance λφ: phase distance C: speed of light λ: wavelength I: ionospheric propagation delay T: tropospheric propagation delay ε: observation error Φ: phase r: true distance f: frequency δtu: receiver time error δts: satellite Time error Namb: Integer uncertain value Δbias: Bias between frequencies in satellite / receiver (2) Cycle slip determination by modified Kalman filter The cycle slip is determined by the filter processing unit 21. As disclosed in Non-Patent Document 4 and Non-Patent Document 5, a basic system of a Kalman filter that performs cycle slip determination and the like includes a state equation and an observation equation.

The state vector shown in the following formula (2.1) is a multi-phase distance difference (ΔI _k ) and its time change.

The equation of state is shown in equation (2.2) and the observation equation is shown in equation (2.3). Where x_ is a predicted value, Φ is a state transition matrix, ΔF _k is an observed value, Δt is a sampling interval, R _k is an observation error covariance matrix, H is an observation matrix, P _k is a true value covariance matrix, K _k represents Kalman gain, and ν _k represents noise.

  Note that I in the equation (2.7) represents a unit matrix, the subscript t in the equations (2.4) and (2.8) represents a transposed matrix, and in the equation (2.8) E [•] represents an average value.

  Next, cycle slip determination processing by the modified Kalman filter executed by the filter processing unit 21 will be described with reference to the flowchart shown in FIG.

  First, initialization is performed (step S11). That is, a state vector is defined, and the Hb and Φb matrices are set as data (initial values) immediately before the H and Φ matrices. Next, the next predicted value is obtained using the state equation (2.2) (step S12). Next, it is checked whether or not there is any missing data (step S13). Specifically, whether the time difference D between the previous data and the current data is greater than 2Δt, that is, whether the collection time difference D between the previous data and the current data is greater than twice the sample time interval Δt. Be examined.

  If it is determined in step S13 that data is missing, the previous data (past data before updating) is used, that is, the Hb and Φb stored in the previous time as H and Φ matrices. Each matrix is set, and a flag flag indicating that a cycle slip has occurred is set to “1” (step S14). On the other hand, if it is determined in step S13 that there is no missing data, the process in step S14 is skipped.

Next, the Kalman filter covariance matrix is updated (step S15). In step S15, a difference dE between the observed value ΔF _k and the estimated value is obtained. Next, it is checked whether or not the difference dE obtained in step S15 is larger than six times the error variance value or the flag flag is “1” (step S16). Here, the variance value σ is “√ (P (1,1))”.

  In step S16, if it is determined that the difference dE is greater than 6 times the variance value or the flag flag is “1”, it is recognized that there has been a cycle slip or that there has been data loss. The flag flag is cleared to “0”, and the difference dE is added to the predicted value x_ to obtain a new predicted value (step S17). Transient phenomena can be minimized by the process of adding the difference dE in step S17.

  On the other hand, if it is determined in step S16 that the difference dE is not larger than 6 times the variance value and the flag flag is not “1”, the process of step S17 is skipped.

  Next, the next estimated value of the Kalman filter is calculated (step S18). In step S18, a collection time difference D between the current data and the next data is calculated. Next, it is checked whether or not there is data missing (step S19). That is, it is checked whether the collection time difference D is greater than twice the sample time interval Δt.

  If it is determined in step S19 that there is data loss, the H and Φ matrices at that time are stored as Hb and Φb matrices representing the previous H and Φ matrices, respectively (step S20). . In other words, if it is determined in step S19 that there is no missing data, the process of step S20 is skipped.

  Next, the H and Φ matrices are updated (step S21). Next, it is checked whether the sample number is smaller than the total number of samples (step S22). If it is determined in step S22 that the sample number is smaller than the total number of samples, it is recognized that the processing for all the samples has not been completed, and the process returns to step S12 and the above-described processing is repeated.

  On the other hand, if it is determined in step S22 that the sample number is not smaller than the total number of samples, it is recognized that the processing for all the samples has been completed, and the cycle slip determination processing ends.

  By the processing described above, when the cycle slip occurs, the H and Φ matrices are temporarily saved and updated, and then the saved H and Φ matrices are restored. Can be reduced.

  Further, since the collection time difference D between the previous data and the current data is assigned to the Φ matrix for data update every time, when the collection time difference D becomes very large, for example, the satellite becomes invisible and becomes visible again. In some cases, the update data x_ becomes a very large value, and can be recognized by cycle slip determination.

(3) Ionosphere Total Electron Count Estimation The ionosphere total electron count is estimated by the ionosphere total electron count estimation unit 22. (1.4) in the expression _{(λ L1 ΔN L1, amb -λ} L2 ΔN L2, amb), the delta _bias, is fixed without change unless occur cycle slip. Using this, in Non-Patent Document 1, this fixed value is deleted by the following equation.

Here, M indicates the number of samples that can be continuously collected without cycle slip. TEC _k indicates the total number of electrons of the k-th data of the satellite s.

Therefore, from (3.1), the TEC is obtained as follows.

Here, the bias term is divided into a bias generated by the receiver 12 and a bias generated by the satellite. The bias by the receiver 12 is a bias common to all the satellites at one observation location, and the bias generated in the satellite is a different bias for each satellite. The bias generated in the satellite is broadcast as a group delay Tgd shown in Non-Patent Document 5 included in the satellite navigation calendar.

(T _L1 −t _L2 ) in the formula (3.3) is a time difference between L1 and L2 measured on the ground. The actual satellite bias may be changing over time. By applying Equation (3.3) to Equation (3.2), the satellite bias component error can be further reduced.

  The error component related to the inter-frequency bias described in the equation (3.4) is estimated and deleted by another method.

(4) Pseudo distance correction for positioning The pseudo distance correction used for positioning is executed by the pseudo distance correction unit 23. The ionospheric propagation delay amount included in the carrier phase L1 can be obtained from the contents disclosed in Non-Patent Document 2 and the equation (3.4) as follows.

When correction is performed by subtracting the ionospheric propagation delay amount from the pseudo distances of the equations (1.1) and (1.2) using the equation (4.1), the following is obtained.

  The subscript k indicates the data number of the satellite s, and u indicates the number for identifying the receiver 12.

As for the pseudo distance used for positioning, the observation error can be reduced and the positioning accuracy can be improved by using the equation (4.3) based on the carrier phase. The integer uncertain value N _amb term is also left in the equation (4.3). The method for deleting the integer uncertainty value in this case is not disclosed in Non-Patent Document 1, but the integer uncertainty value is deleted by a method similar to the method disclosed in Non-Patent Document 1.

Here, the bias term is mainly a bias generated by the receiver 12, but a bias correction error caused by the satellite due to the aging of the satellite still remains. The bias due to the receiver 12 becomes a bias common to all the satellites at one observation place, and cannot be distinguished from the receiver clock error. The bias generated by the satellite cannot be distinguished from the satellite time error. The satellite time error can be corrected by the satellite navigation calendar broadcast from the satellite. This correction term is a term described in Non-Patent Document 6 as Δtsv. For this reason, the error amount varies from satellite to satellite due to the bias error after the bias correction between satellite frequencies. In the simple positioning calculation methods described in Non-Patent Document 2 and Non-Patent Document 3, an error commonly included in the satellite is regarded as a time error component of the receiver 12. For this reason, the bias error and the time error included in the receiver 12 do not affect the position error of the positioning calculation.

Note that the delay amount T caused by the troposphere of the satellite is corrected by using the method by Sasstamoinen disclosed in Non-Patent Document 2. Further, the satellite time error Δt ^s is corrected using a time correction coefficient existing in the satellite navigation calendar broadcast from the satellite. The satellite bias can be estimated and corrected by other methods to improve positioning accuracy.

  The present invention can be applied to a positioning system including an ionosphere electron density distribution estimation system that requires accurate positioning using a satellite.

It is a figure which shows the structure of the positioning system containing the ionosphere electron density distribution estimation system which concerns on Example 1 of this invention. It is a block diagram which shows the detailed structure of the signal processing apparatus which comprises the satellite signal processing system of the positioning system containing the ionosphere electron density distribution estimation system which concerns on Example 1 of this invention. It is a flowchart which shows operation | movement of the filter process part of the signal processing apparatus which comprises the satellite signal processing system of the positioning system containing the ionosphere electron density distribution estimation system which concerns on Example 1 of this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Satellite signal observation system 2 Satellite signal processing system 3 Internet data processing system 4 Data server system 11 Antenna 12 Receiver 20 Signal processing apparatus 21 Filter processing part 22 Ionospheric total electron number estimation part 22a Electron density estimation processing part 23 Pseudo distance correction part 24 Positioning processing unit 41 Data collection server

Claims (4)

By determining the presence or absence of cycle slip using the multi-phase distance difference and the covariance matrix of the modified Kalman filter based on the velocity component of this distance difference and the sampling time difference of the satellite, the continuous data collection period is specified and the cycle slip is determined. A filter processing unit that updates the covariance matrix of the modified Kalman filter when it occurs ,
For the data collection period specified by the filter processing unit, an ionosphere total electron number estimation unit that converts the multiple carrier phase pseudorange difference after removing the uncertain value related to the carrier phase into the ionosphere electron number,
An electron density estimation processing unit that estimates the electron density distribution of the ionosphere using the ionosphere electron number obtained by the ionosphere total electron number estimation unit;
An ionospheric electron density distribution estimation system comprising: The filter processing unit
Based on the difference between the value estimated by the modified Kalman filter and the observed value, the variance value dynamically estimated by the modified Kalman filter, and the data collection time difference, the presence or absence of a cycle slip was determined, and the cycle slip occurred. 2. The ionospheric electron density distribution estimation system according to claim 1, wherein the data collection period continued by operating the modified Kalman filter is specified with the slip amount as an input. A pseudorange correction unit that performs a process of subtracting the ionosphere propagation delay amount corresponding to the ionosphere electron number obtained by the ionosphere total electron number estimation unit according to claim 1 or 2, from the pseudorange obtained by positioning;
A positioning processing unit that performs positioning using the corrected pseudorange sent from the pseudorange correction unit and the satellite position obtained from the satellite navigation calendar,
A positioning system characterized by comprising.   The positioning system according to claim 3, wherein the pseudo distance correcting unit further corrects a delay amount caused by a troposphere.

Images