CN105738926A - Method for calibrating phase inter-frequency bias between GLONASS system receiving machines - Google Patents

Abstract

The invention provides a method for calibrating a phase inter-frequency bias between GLONASS system receiving machines. A calibration for a carrier wave phase inter-frequency bias between receiving machines is performed by utilizing short-baseline GLONASS double-difference observation quantity. The method comprises the following steps: firstly inputting observation data, obtaining a baseline vector of a fixed solution by utilizing a GPS+BDS double-difference observation quantity, screening the data then, determining a reference satellite, determining the number of parameters, establishing a parameter estimation model, sampling a carrier wave phase inter-frequency bias between receiving machines finally, calculating a double-difference ambiguity floating point solution through batch processing by utilizing least square adjustment, searching a fixed ambiguity through ratio inspection, and evaluating a calibration value of the carrier wave phase inter-frequency bias between the receiving machines. Reliability of the GLONASS whole-circle ambiguity in baseline solution can be improved, and rapid and real-time relative positioning is facilitated.

Description

Phase place inter-frequency deviation scaling method between a kind of GLONASS system receiver

Technical field

The present invention relates to GLONASS relative localization field, the concrete a kind of method carrier phase inter-frequency deviation between any two receivers demarcated for the GLONASS of utilization system double difference.

Background technology

GLONASS is that after GPS, second constellation be complete and GNSS system in orbit, but use extensively not as good as GPS, and the precision of quick, real-time accurate location and relative localization is not high, one of the main reasons is exactly the impact of the carrier phase inter-frequency deviation (inter-frequencybias, IFB) that GLONASS system produces at receiver end.

With adopt CDMA GPS system the difference is that, GLONASS system adopts frequency division multiple access technology, the carrier frequency of every satellite launch is different, and the satellite-signal of different frequency can enter the internal corresponding frequency channel of receiver, thus can produce different IFB.So the gps data processing method of routine can not be used when processing the carrier phase value of GLONASS, during in particular with conventional process double difference data, owing to the IFB of different satellites is different, GLONASS double difference carrier phase equation can not eliminate IFB, have impact on the fixing of integer ambiguity, thus reducing the precision of location.

Based on some characteristics of IFB, the frequency of such as IFB and satellite is linear；L1 and L2 frequency range has the IFB (unit/rice) of formed objects, and highly stable, and size is in the scope of interval [-0.1,0.1] (unit/rice), and major part research method all have employed and estimates IFB and fuzziness parameter simultaneously at present.But due to this two parts parameter strong correlation, it is necessary to long observation data are only possible to resolving and obtain, so not being provided that comparatively accurate IFB priori corrected value, it is difficult to quick, real-time positioning.

In actual applications, once the IFB obtained between any two GLONASS receiver, just can be the same with GPS relative localization, double-differential carrier phase observed quantity is utilized fuzziness accurately to be solved and fixes, and then obtain accurate carrier phase observation data, it is also possible to realize the integrated positioning of GLONASS and other GNSS system better.

Summary of the invention

The present invention is directed to the problems referred to above, devise the scaling method of phase place inter-frequency deviation between a kind of GLONASS system receiver.The method estimates the IFB calibration value between the receiver obtained, for quick, real-time relative positioning, it is possible to improves the reliability of GLONASS integer ambiguity in Baselines and is fixed into power, and then improves positioning precision and the availability of GLONASS；

The technical scheme is that

Phase place inter-frequency deviation scaling method between a kind of GLONASS system receiver, it is characterised in that comprise the steps:

Step 1: preferentially choose short baseline, two survey stations that namely selected distance is short as far as possible.Read observation data file and navigation file carries out data inputting.Observation data are carried out Detection of Cycle-slip scanning, the epoch of all generation cycle slips of labelling.

Step 2: utilize the basic lineal vector that the short-and-medium baseline of GPS+BDS double difference process of solution 1 is formed, it is thus achieved that the basic lineal vector of fixed solution.

Step 3: the original observed data in step 1 is screened, preferentially choose continuous segmental arc, GLONASS satellite number be not less than 4 data.Determine the reference satellite of all epoch, calculate number of parameters to be estimated, wherein, the epoch of the generation cycle slip of labelling in step 1 is arranged new fuzziness parameter, and does not occur keep fuzziness parameter constant the epoch of cycle slip.

Step 4: set up parameter estimation model, including function model and stochastic model, specifically comprises the following steps that

Step 4.1: the satellite of each epoch is formed double-differential carrier phase observational equation:

Wherein, n is frequency range 1 or 2, and A is reference receiver title, and B is rover station receiver title, and i, j are satellite number, and j is reference satellite,For the wavelength of satellite i that frequency range is n,For the wavelength of satellite j that frequency range is n,For the receiver A and the receiver B single difference phase observation value to satellite i,For the receiver A and the receiver B single difference phase observation value to satellite j,For receiver A and receiver B, the double difference of satellite i and satellite j is defended distance,For the receiver A and the receiver B double difference fuzziness to satellite i and satellite j,For the receiver A and the receiver B poor fuzziness of list to satellite j,For the receiver A and the receiver B double difference inter-frequency deviation to satellite i and satellite j,For the receiver A and the receiver B double difference observation noise to satellite i and satellite j.Wherein,Can be expressed as:

${\mathrm{\γ}}_{AB}^{ij}=(k^i-k^j){\mathrm{\Δ\γ}}_{AB}---\left(0.2\right)$

Wherein, kⁱAnd k^jRespectively corresponding with satellite j for satellite i frequency channel number, Δ γ_ABFor the inter-frequency deviation between receiver A and receiver B, it is value to be calibrated.

Step 4.2: utilize broadcast ephemeris to calculate satellite position, according to the double difference Pseudo-range Observations in base station coordinates and basic lineal vector calculating formula (1.1)

Step 4.3: calculate the poor fuzziness of list of reference satellite each epochComputing formula is:

Wherein,For single poor Pseudo-range Observations.

Step 4.4: utilize the least square adjustment, the normal equation of single epoch to be:

Wherein,For parameter to be estimated.The normal equation of all epoch is overlapped, forms final normal equation, order:

Wherein, n is the number of satellite.Total normal equation is abbreviated as:

$N_{BB}\hat{x}-W=0---\left(1.6\right)$

Wherein,For double difference fuzziness.

Step 4.5: use elevation angle function method to determine the variance-covariance matrix of observation.

For observing, at survey station A, the satellite i that elevation angle is Ele, the variance of its observation is expressed as:

Wherein σ²For the prior variance of observation, carrier phase observation data is generally taken as 0.002m.Considering the mathematical correlation of double difference observation, now the power battle array of phase place double difference observational equation is:

Wherein, subscript s represents observation type, D_sFor the variance-covariance matrix of observation, N is the double difference observational equation number sum of all epoch, it is assumed that the variance of unit weight factorIt is 1.

$\underset{AB}{\overset{ij}{\Σ}}{\mathrm{\σ}}_s^2={\left({\mathrm{\σ}}_A^i\right)}^2+{\left({\mathrm{\σ}}_B^i\right)}^2+{\left({\mathrm{\σ}}_A^j\right)}^2+{\left({\mathrm{\σ}}_B^j\right)}^2---\left(1.9\right)$

$\underset{AB}{\overset{i}{\Σ}}{\mathrm{\σ}}_s^2={\left({\mathrm{\σ}}_A^i\right)}^2+{\left({\mathrm{\σ}}_B^i\right)}^2---\left(1.10\right)$

Formula (1.9) represents the observation variance sum of base station A and rover station B simultaneous observation satellite i, j, and formula (1.10) represents base station and the rover station observation variance sum to reference satellite.

Step 5: [-0.1,0.1] interval (unit/rice) is divided into 200 chosen candidate values with the interval of 1 millimeter, to Δ γ_ABSampling, substitute in the normal equation shown in step 4 Chinese style (1.5) respectively, batch processing resolves each epoch in the period, and obtaining double difference fuzziness float-solution is:

$\hat{x}-N_{BB}^{-1}W---\left(1.11\right)$

Its covariance matrix is:

$D_{\hat{x}\hat{x}}={\widehat{\mathrm{\σ}}}_0^2{N}_{BB}^{-1}---\left(1.12\right)$

Wherein,Variance is weighed for unit.

Step 6: the double difference fuzziness float-solution resolved according to step 5And covariance matrixUtilizing LAMBDA method search fuzziness, whether ambiguity search is successful to adopt ratio inspection to judge, and preserves ratio value.Finally, the Δ γ that the maximum in 200 ratio values is corresponding_ABIt is between final receiver the calibration value of inter-frequency deviation.

Phase place inter-frequency deviation scaling method between above-mentioned a kind of GLONASS system receiver, utilizes the known basic lineal vector that GPS+BDS double difference resolves in described step 2, need not take basic lineal vector parameter in parameter estimation into account.

Phase place inter-frequency deviation scaling method between above-mentioned a kind of GLONASS system receiver, keeps fuzziness parameter constant, and the epoch that cycle slip occurs is reset fuzziness parameter continuous segmental arc in described step 3.

Phase place inter-frequency deviation scaling method between above-mentioned a kind of GLONASS system receiver, in described step 5, will process the normal equation superposition of each epoch in data period, the method adopting batch processing, all of double difference fuzziness float-solution in the unified resolving period.

Phase place inter-frequency deviation scaling method between above-mentioned a kind of GLONASS system receiver, fixes integer ambiguity in described step 6 and searches for, with the double-differential carrier phase inter-frequency deviation between GLONASS receiver, the method combined.

Plant phase place inter-frequency deviation between GLONASS system receiver to demarcate, including such as lower module:

Data inputting module: be used for reading observation file and navigation file；

Baselines module: for calculating the basic lineal vector obtaining fixed solution；

Data screening module: preferentially choose continuous segmental arc, GLONASS satellite number be not less than 4 data, and choose the reference satellite that elevation angle is higher, it is determined that number of parameters；

Set up model module: be used for setting up function model and stochastic model, solving model coefficient；

Search fuzziness module: solve, for batch processing, the double difference fuzziness floating-point system of solutions that inter-frequency deviation between different receivers is corresponding, and search for fixing double difference fuzziness；

Assessment module: for determining the calibration value of inter-frequency deviation between receiver.

Therefore, the present invention can improve the reliability of GLONASS integer ambiguity in Baselines, is conducive to quick, real-time relative localization.

Accompanying drawing explanation

Fig. 1 is the schematic flow sheet of the present invention.

Detailed description of the invention

Phase place inter-frequency deviation scaling method between a kind of GLONASS system receiver, comprises the steps:

Step 1: preferentially choose short baseline, two survey stations that namely selected distance is short as far as possible.Read observation data file and navigation file carries out data inputting.Observation data are carried out Detection of Cycle-slip scanning, the epoch of all generation cycle slips of labelling.

Step 2: utilize the basic lineal vector that the short-and-medium baseline of GPS+BDS double difference process of solution 1 is formed, it is thus achieved that the basic lineal vector of fixed solution.

Step 3: the original observed data in step 1 is screened, preferentially choose continuous segmental arc, GLONASS satellite number be not less than 4 data.Determine the reference satellite of all epoch, calculate number of parameters to be estimated, wherein, the epoch of the generation cycle slip of labelling in step 1 is arranged new fuzziness parameter, and does not occur keep fuzziness parameter constant the epoch of cycle slip.

Step 4: set up parameter estimation model, including function model and stochastic model, specifically comprises the following steps that

Step 4.1: the satellite of each epoch is formed double-differential carrier phase observational equation:

Wherein, n is frequency range 1 or 2, and A is reference receiver title, and B is rover station receiver title, and i, j are satellite number, and j is reference satellite,For the wavelength of satellite i that frequency range is n,For the wavelength of satellite j that frequency range is n,For the receiver A and the receiver B single difference phase observation value to satellite i,For the receiver A and the receiver B single difference phase observation value to satellite j,For receiver A and receiver B, the double difference of satellite i and satellite j is defended distance,For the receiver A and the receiver B double difference fuzziness to satellite i and satellite j,For the receiver A and the receiver B poor fuzziness of list to satellite j,For the receiver A and the receiver B double difference inter-frequency deviation to satellite i and satellite j,For the receiver A and the receiver B double difference observation noise to satellite i and satellite j.Wherein,Can be expressed as:

${\mathrm{\γ}}_{AB}^{ij}=(k^i-k^j){\mathrm{\Δ\γ}}_{AB}---\left(0.4\right)$

Wherein, kⁱAnd k^jRespectively corresponding with satellite j for satellite i frequency channel number, Δ γ_ABFor the inter-frequency deviation between receiver A and receiver B, it is value to be calibrated.

Step 4.2: utilize broadcast ephemeris to calculate satellite position, according to the double difference Pseudo-range Observations in base station coordinates and basic lineal vector calculating formula (1.1)

Step 4.3: calculate the poor fuzziness of list of reference satellite each epochComputing formula is:

Wherein,For single poor Pseudo-range Observations.

Step 4.4: utilize the least square adjustment, the normal equation of single epoch to be:

Wherein,For parameter to be estimated.The normal equation of all epoch is overlapped, forms final normal equation, order:

Wherein, n is the number of satellite.Total normal equation is abbreviated as:

$N_{BB}\hat{x}-W=0---\left(1.6\right)$

Wherein,For double difference fuzziness.

Step 4.5: use elevation angle function method to determine the variance-covariance matrix of observation.

For observing, at survey station A, the satellite i that elevation angle is Ele, the variance of its observation is expressed as:

Wherein σ²For the prior variance of observation, carrier phase observation data is generally taken as 0.002m.Considering the mathematical correlation of double difference observation, now the power battle array of phase place double difference observational equation is:

Wherein, subscript s represents observation type, D_sFor the variance-covariance matrix of observation, N is the double difference observational equation number sum of all epoch, it is assumed that the variance of unit weight factorIt is 1.

$\underset{AB}{\overset{ij}{\Σ}}{\mathrm{\σ}}_s^2={\left({\mathrm{\σ}}_A^i\right)}^2+{\left({\mathrm{\σ}}_B^i\right)}^2+{\left({\mathrm{\σ}}_A^j\right)}^2+{\left({\mathrm{\σ}}_B^j\right)}^2---\left(1.9\right)$

$\underset{AB}{\overset{i}{\Σ}}{\mathrm{\σ}}_s^2={\left({\mathrm{\σ}}_A^i\right)}^2+{\left({\mathrm{\σ}}_B^i\right)}^2---\left(1.10\right)$

Formula (1.9) represents the observation variance sum of base station A and rover station B simultaneous observation satellite i, j, and formula (1.10) represents base station and the rover station observation variance sum to reference satellite.

Step 5: [-0.1,0.1] interval (unit/rice) is divided into 200 chosen candidate values with the interval of 1 millimeter, to Δ γ_ABSampling, substitute in the normal equation shown in step 4 Chinese style (1.5) respectively, batch processing resolves each epoch in the period, and obtaining double difference fuzziness float-solution is:

$\hat{x}=N_{BB}^{-1}W---\left(1.11\right)$

Its covariance matrix is:

$D_{\hat{x}\hat{x}}={\widehat{\mathrm{\σ}}}_0^2{N}_{BB}^{-1}---\left(1.12\right)$

Wherein,Variance is weighed for unit.

Step 6: the double difference fuzziness float-solution resolved according to step 5And covariance matrixUtilizing LAMBDA method search fuzziness, whether ambiguity search is successful to adopt ratio inspection to judge, and preserves ratio value.Finally, the Δ γ that the maximum in 200 ratio values is corresponding_ABIt is between final receiver the calibration value of inter-frequency deviation.

Step 2 utilizes the known basic lineal vector that GPS+BDS double difference resolves, parameter estimation need not be taken into account basic lineal vector parameter.Continuous segmental arc is kept fuzziness parameter constant by step 3, and the epoch that cycle slip occurs is reset fuzziness parameter.In step 5, the normal equation superposition of each epoch in data period, the method adopting batch processing, all of double difference fuzziness float-solution in the unified resolving period will be processed.Integer ambiguity is fixed by step 6 and searches for, with the double-differential carrier phase inter-frequency deviation between GLONASS receiver, the method combined.

Between a kind of GLONASS system receiver, phase place inter-frequency deviation is demarcated, including such as lower module:

Data inputting module: be used for reading observation file and navigation file；

Baselines module: for calculating the basic lineal vector obtaining fixed solution；

Data screening module: preferentially choose continuous segmental arc, GLONASS satellite number be not less than 4 data, and choose the reference satellite that elevation angle is higher, it is determined that number of parameters；

Set up model module: be used for setting up function model and stochastic model, solving model coefficient；

Search fuzziness module: solve, for batch processing, the double difference fuzziness floating-point system of solutions that inter-frequency deviation between different receivers is corresponding, and search for fixing double difference fuzziness；

Assessment module: for determining the calibration value of inter-frequency deviation between receiver.

Specific embodiment described herein is only to present invention spirit explanation for example.Described specific embodiment can be made various amendment or supplements or adopt similar mode to substitute by those skilled in the art, but without departing from the spirit of the present invention or surmount the scope that appended claims is defined.

Claims (6)

1. phase place inter-frequency deviation scaling method between a GLONASS system receiver, it is characterised in that comprise the steps:

Step 1: preferentially choose short baseline, two survey stations that namely selected distance is short as far as possible；Read observation data file and navigation file carries out data inputting；Observation data are carried out Detection of Cycle-slip scanning, the epoch of all generation cycle slips of labelling；

Step 2: utilize the basic lineal vector that the short-and-medium baseline of GPS+BDS double difference process of solution 1 is formed, it is thus achieved that the basic lineal vector of fixed solution；

Step 3: the original observed data in step 1 is screened, preferentially choose continuous segmental arc, GLONASS satellite number be not less than 4 data；Determine the reference satellite of all epoch, calculate number of parameters to be estimated, wherein, the epoch of the generation cycle slip of labelling in step 1 is arranged new fuzziness parameter, and does not occur keep fuzziness parameter constant the epoch of cycle slip；

Step 4: set up parameter estimation model, including function model and stochastic model, specifically comprises the following steps that

Step 4.1: the satellite of each epoch is formed double-differential carrier phase observational equation:

Wherein, n is frequency range 1 or 2, and A is reference receiver title, and B is rover station receiver title, and i, j are satellite number, and j is reference satellite,For the wavelength of satellite i that frequency range is n,For the wavelength of satellite j that frequency range is n,For the receiver A and the receiver B single difference phase observation value to satellite i,For the receiver A and the receiver B single difference phase observation value to satellite j,For receiver A and receiver B, the double difference of satellite i and satellite j is defended distance,For the receiver A and the receiver B double difference fuzziness to satellite i and satellite j,For the receiver A and the receiver B poor fuzziness of list to satellite j,For the receiver A and the receiver B double difference inter-frequency deviation to satellite i and satellite j,For the receiver A and the receiver B double difference observation noise to satellite i and satellite j；Wherein,Can be expressed as:

${\mathrm{\γ}}_{AB}^{ij}=(k^i-k^j){\mathrm{\Δ\γ}}_{AB}---\left(0.2\right)$

Wherein, kⁱAnd k^jRespectively corresponding with satellite j for satellite i frequency channel number, Δ γ_ABFor the inter-frequency deviation between receiver A and receiver B, it is value to be calibrated；

Step 4.2: utilize broadcast ephemeris to calculate satellite position, according to the double difference Pseudo-range Observations in base station coordinates and basic lineal vector calculating formula (1.1)

Step 4.3: calculate the poor fuzziness of list of reference satellite each epochComputing formula is:

Wherein,For single poor Pseudo-range Observations；

Step 4.4: utilize the least square adjustment, the normal equation of single epoch to be:

Wherein,For parameter to be estimated；The normal equation of all epoch is overlapped, forms final normal equation, order:

Wherein, n is the number of satellite；Total normal equation is abbreviated as:

$N_{BB}\hat{x}-W=0---\left(1.6\right)$

Wherein,For double difference fuzziness；

Step 4.5: use elevation angle function method to determine the variance-covariance matrix of observation；

For observing, at survey station A, the satellite i that elevation angle is Ele, the variance of its observation is expressed as:

Wherein σ²For the prior variance of observation, carrier phase observation data is generally taken as 0.002m；Considering the mathematical correlation of double difference observation, now the power battle array of phase place double difference observational equation is:

Wherein, subscript s represents observation type, D_sFor the variance-covariance matrix of observation, N is the double difference observational equation number sum of all epoch, it is assumed that the variance of unit weight factorIt is 1；

$\underset{AB}{\overset{ij}{\Σ}}{\mathrm{\σ}}_s^2={\left({\mathrm{\σ}}_A^i\right)}^2+{\left({\mathrm{\σ}}_B^i\right)}^2+{\left({\mathrm{\σ}}_A^j\right)}^2+{\left({\mathrm{\σ}}_B^j\right)}^2---\left(1.9\right)$

$\underset{AB}{\overset{i}{\Σ}}{\mathrm{\σ}}_s^2={\left({\mathrm{\σ}}_A^i\right)}^2+{\left({\mathrm{\σ}}_B^i\right)}^2---\left(1.10\right)$

Formula (1.9) represents the observation variance sum of base station A and rover station B simultaneous observation satellite i, j, and formula (1.10) represents base station and the rover station observation variance sum to reference satellite；

Step 5: [-0.1,0.1] interval (unit/rice) is divided into 200 chosen candidate values with the interval of 1 millimeter, to Δ γ_ABSampling, substitute in the normal equation shown in step 4 Chinese style (1.5) respectively, batch processing resolves each epoch in the period, and obtaining double difference fuzziness float-solution is:

$\hat{x}=N_{BB}^{-1}W---\left(1.11\right)$

Its covariance matrix is:

$D_{\hat{x}\hat{x}}={\widehat{\mathrm{\σ}}}_0^2{N}_{BB}^{-1}---\left(1.12\right)$

Wherein,Variance is weighed for unit；

Step 6: the double difference fuzziness float-solution resolved according to step 5And covariance matrixUtilizing LAMBDA method search fuzziness, whether ambiguity search is successful to adopt ratio inspection to judge, and preserves ratio value；Finally, the Δ γ that the maximum in 200 ratio values is corresponding_ABIt is between final receiver the calibration value of inter-frequency deviation.

2. phase place inter-frequency deviation scaling method between a kind of GLONASS system receiver according to claim 1, it is characterized in that, described step 2 utilizes the known basic lineal vector that GPS+BDS double difference resolves, parameter estimation need not be taken into account basic lineal vector parameter.

3. phase place inter-frequency deviation scaling method between a kind of GLONASS system receiver according to claim 1, it is characterised in that in described step 3, continuous segmental arc is kept fuzziness parameter constant, and the epoch that cycle slip occurs is reset fuzziness parameter.

4. phase place inter-frequency deviation scaling method between a kind of GLONASS system receiver according to claim 1, it is characterized in that: in described step 5, the normal equation superposition of each epoch in data period will be processed, the method adopting batch processing, all of double difference fuzziness float-solution in the unified resolving period.

5. phase place inter-frequency deviation scaling method between a kind of GLONASS system receiver according to claim 1, it is characterized in that, integer ambiguity is fixed by described step 6 and searches for, with the double-differential carrier phase inter-frequency deviation between GLONASS receiver, the method combined.

6. between a GLONASS system receiver, phase place inter-frequency deviation is demarcated, including such as lower module:

Data inputting module: be used for reading observation file and navigation file；

Baselines module: for calculating the basic lineal vector obtaining fixed solution；

Data screening module: preferentially choose continuous segmental arc, GLONASS satellite number be not less than 4 data, and choose the reference satellite that elevation angle is higher, it is determined that number of parameters；

Set up model module: be used for setting up function model and stochastic model, solving model coefficient；

Search fuzziness module: solve, for batch processing, the double difference fuzziness floating-point system of solutions that inter-frequency deviation between different receivers is corresponding, and search for fixing double difference fuzziness；

Assessment module: for determining the calibration value of inter-frequency deviation between receiver.