JP2015099099A - GLONASS receiver - Google Patents

Abstract

PROBLEM TO BE SOLVED: To prevent occurrences of positioning abnormalities and positioning errors by acquiring leap second correction information without having a leap second adjustment error in real time, and estimating hardware delay difference with high accuracy without increasing a circuit scale in a GLONASS receiver for performing mixture positioning of GLONASS and GPS.SOLUTION: A GLONASS receiver R acquires satellite transmission time information of GLONASS of a GLONASS time system by demodulating a GLONASS satellite signal received from a GLONASS satellite, acquires satellite transmission time information of GPS of a GPS time system by demodulating a GPS satellite signal received from a GPS satellite, and calculates a leap second correction value by comparing the satellite transmission time information of GLONASS of the GLONASS time system with the satellite transmission time information of GPS of the GPS time system after compensating a time difference of standard time.

Description

  The present invention relates to a GLONASS receiver that performs positioning using satellite signals of GLONASS (Global Navigation Satellite System) and GPS (Global Positioning System).

  The GLONASS receiver receives the GLONASS and GPS satellite signals and measures the position of the own device. GLONASS is a Russian satellite positioning system, GPS is a US positioning system, and these are separate satellite positioning systems.

Japanese Patent Laid-Open No. 11-109016

U. Rossbach, “Positioning and Navigation Using the Russian Satrite System GLONASS”, ub. unibw-muenchen. de, Juni 2000.

  By the way, GLONASS and GPS have the following differences. First, the GLONASS time system is adjusted for leap seconds, but the GPS time system is not adjusted for leap seconds. Therefore, when performing GLONASS and GPS mixed positioning, it is necessary to align both time systems, and in order to align both time systems, it is necessary to consider the presence or absence of leap second adjustment for both time systems. If the presence / absence of the leap second adjustment is not taken into account, there is a problem that a positioning abnormality is caused at the leap second adjustment timing which may be performed about once a year. Secondly, GLONASS and GPS satellite signals have different transmission frequencies. Therefore, when performing mixed positioning of GLONASS and GPS, there is a problem that a positioning error is caused unless a hard delay difference caused by a difference in pass bandwidth due to a bandpass filter in the own aircraft for both satellite signals is taken into consideration. .

  In Patent Document 1, a hard delay difference is considered, but since leap second adjustment of the GLONASS time system is not considered at all, a positioning abnormality occurs at leap second adjustment timing when performing mixed positioning of GLONASS and GPS. Also, the hard delay difference is generated by a dedicated generation circuit in the own aircraft with a pilot signal having the same frequency as the transmission frequency of both satellite signals, and the time difference between the generation time and the reception time is received in the dedicated pilot signal reception channel unit It is measured by measuring with. Therefore, the circuit scale of the GLONASS receiver increases by the amount of the pilot signal generation circuit and the pilot signal reception channel section.

  Non-Patent Document 1 considers leap second adjustment and hard delay difference of the GLONASS time system. Here, information for correcting the influence of the leap second adjustment of the GLONASS time system (hereinafter referred to as leap second correction information) is acquired from the GPS navigation data, and using the acquired leap second correction information, the GLONASS time is acquired. By converting the system to the GPS time system, GLONASS and GPS mixed positioning is performed. The hard delay difference is the time constant difference between the GPS and GLONASS band-pass filters, and the time constant difference causes a difference in signal arrival between the GPS and GLONASS, resulting in a difference between the GPS and GLONASS. Relative reception time difference. Therefore, the hard delay difference is estimated by solving a positioning equation in which a relative reception time difference between GPS and GLONASS is added as an unknown quantity. Here, the time constant of the bandpass filter can be calculated theoretically at the time of design, but due to changes in analog circuit characteristics due to temperature changes and individual component differences in the actual environment, theoretical calculation values should be used to reduce positioning errors. It is necessary to estimate without using it. By estimating the hard delay difference, the influence on the positioning result due to the hard delay difference is eliminated.

  By the way, leap second correction information in GPS navigation data is transmitted at intervals of 12.5 minutes. Therefore, the latest leap second correction information needs to wait for a maximum of 12.5 minutes even in an open sky environment with good reception conditions, and cannot be acquired in real time. If the latest leap second correction information cannot be obtained, the leap second correction information received in the past is used. Since there is no guarantee that the past leap second correction information matches the latest leap second correction information, the latest leap second adjustment may include an error of ± 1 second or more.

The leap second correction error of ± 1 second or more becomes an offset distance error between the GPS and GLONASS pseudoranges if the distance is converted by multiplying the speed of light, and if the distance error cannot be removed, the positioning result is affected. Become. By using a positioning equation in which a leap second correction error of ± 1 second or more is used as a relative reception time difference between GLONASS and GPS without converting the distance, and the relative reception time difference is added as an unknown amount, ± 1 Leap second correction errors of more than a second are absorbed into the unknown with a hard delay difference. The hard delay difference and the leap second correction error of ± 1 second or more are estimated as one unknown amount. In other words, this one unknown amount has a linear combination of a hard delay difference on the order of several tens of ns and a leap second correction error on the order of seconds, and is estimated by a positioning equation. However, since the ratio between the hard delay difference of the order of several tens of ns and the leap second correction error of the second order is about 10 ⁸ times, this one unknown amount changes about 10 ⁸ times at the leap second adjustment timing. When the unknown amount to be estimated changes nonlinearly as much as 10 ⁸ times, a hard delay difference that is an order that is too small for this change cannot be accurately estimated. If the unknown quantity cannot be estimated accurately, a large positioning error occurs.

  Therefore, in order to solve the above-mentioned problem, the present invention obtains leap second correction information without a leap second adjustment error in real time in a GLONASS receiver that performs mixed positioning of GLONASS and GPS without increasing the circuit scale. The purpose is to accurately estimate the hard delay difference and not to cause positioning abnormality and positioning error.

  In order to achieve the above object, the leap second correction value is calculated by comparing the GLONASS satellite transmission time information of the GLONASS time system and the GPS satellite transmission time information of the GPS time system after compensating for the time difference of the standard time. It was decided to.

  Specifically, the present invention demodulates a GLONASS satellite signal received from a GLONASS satellite, acquires a GLONASS satellite signal transmission time information of a GLONASS time system, and a GPS satellite signal received from a GPS satellite. A GPS satellite signal demodulator that acquires GPS satellite transmission time information of the GPS time system, GLONASS satellite transmission time information of the GLONASS time system acquired by the GLONASS satellite signal demodulator, and the GPS satellite A leap second correction value calculation unit that calculates a leap second correction value by comparing the GPS satellite transmission time information of the GPS time system acquired by the signal demodulation unit after compensating for the time difference of the standard time. This is a GLONASS receiver.

  According to this configuration, since the latest leap second correction information can be acquired in the GLONASS receiver, a positioning abnormality can be prevented.

  In addition, the present invention provides a pseudo distance and geometric distance between the GLONASS satellite and the own aircraft, a pseudo distance and geometric distance between the GPS satellite and the own aircraft, and the leap second correction value calculating unit. The hard delay difference is calculated by solving the positioning equation using the second correction value as an observation amount and the hard delay difference that is the time constant difference between the GLONASS satellite signal and the GPS satellite signal as a time constant of the bandpass filter in the own aircraft. The GLONASS receiver further includes a local position measuring unit with a hard delay difference estimation that estimates and measures the position of the local apparatus.

  According to this configuration, in the GLONASS receiver, the hard delay difference can be accurately estimated without increasing the circuit scale, so that no positioning error can be caused.

  Further, in the present invention, the own position measurement unit with hard delay difference estimation calculates the estimation accuracy of the hard delay difference based on a dispersion calculation formula of the unknown amount of the positioning equation, and the unknown amount of the positioning equation is calculated. When the estimation accuracy of the hard delay difference calculated based on the variance calculation formula is less than a predetermined threshold, the position of the own device is measured using the hard delay difference as an unknown amount in the positioning equation, and the unknown amount of the positioning equation When the estimation accuracy of the hard delay difference calculated based on the variance calculation formula is equal to or greater than the predetermined threshold, the position of the own device is measured using the backup delay or the default hard delay difference in the positioning equation. This is a GLONASS receiver.

  When the estimation accuracy of the hard delay difference is expected to be high, the hard delay difference is accurately estimated as an unknown amount in the positioning equation. When the estimation accuracy of the hard delay difference is expected to be low, the hard delay difference in the positioning equation is set to a known number that is estimated to have a certain degree of estimation accuracy. Therefore, no positioning error can be caused.

  As described above, according to the present invention, in a GLONASS receiver that performs mixed positioning of GLONASS and GPS, leap second correction information without a leap second adjustment error is obtained in real time, and the hardware delay difference is accurately obtained without increasing the circuit scale. It is possible to estimate well and to prevent positioning errors and positioning errors.

It is a figure which shows the structure of the GLONASS receiver of this invention. It is a figure which shows the relationship of each time system. It is a flowchart which shows the leap second correction value calculation process of this invention. It is a flowchart which shows the own position measurement process with a hard delay difference estimation of this invention.

  Embodiments of the present invention will be described with reference to the accompanying drawings. The embodiments described below are examples of the present invention, and the present invention is not limited to the following embodiments. In the present specification and drawings, the same reference numerals denote the same components.

  The configuration of the GLONASS receiver of the present invention is shown in FIG. The GLONASS receiver R includes an antenna A, down converters 11 and 12, band pass filters 13 and 14, GLONASS satellite signal reception channels 15-1 to 15-N, GPS satellite signal reception channels 16-1 to 16-N, and GLONASS satellites. It comprises signal demodulating units 17-1 to 17-N, GPS satellite signal demodulating units 18-1 to 18-N, leap second correction value calculating unit 19, and own position measuring unit 20 with hardware delay difference estimation. The antenna A receives GLONASS and GPS satellite signals.

  The down converter 11 down-converts the GLONASS satellite signal so that it can be easily processed by a digital circuit. The down-converter 12 down-converts the GPS satellite signal so that it can be easily processed by a digital circuit.

  The bandpass filter 13 passes a signal within a band of about 9 MHz around a carrier frequency of 1602 MHz + k × 562.5 kHz (k is an integer from −7 to 6) when extracting the L1 signal of the GLONASS satellite signal. Let The bandpass filter 14 passes a signal having a band within a range of about 2 MHz around the carrier frequency 1575.42 MHz when extracting the L1 signal of the GPS satellite signal. A hard delay difference between the GLONASS satellite signal and the GPS satellite signal in the own device R may be generated based on a difference in pass bandwidth between the bandpass filter 13 and the bandpass filter 14.

  The GLONASS satellite signal reception channels 15-1 to 15-N receive the GLONASS N satellite signals through the N channels. The GPS satellite signal reception channels 16-1 to 16-N receive N satellite signals of GPS through N channels, respectively. “Receiving” here refers to an operation of capturing satellite signals, tracking them, and extracting bit information of navigation data.

  The GLONASS satellite signal demodulation units 17-1 to 17-N demodulate the N GLONASS satellite signals for each of the N channels. The GPS satellite signal demodulation units 18-1 to 18-N decode GPS N satellite signals for each of the N channels. Here, “demodulate” refers to extracting information necessary for positioning, such as navigation data information, satellite transmission time information, and various correction parameters, from the bit information of the navigation data extracted by tracking.

  The navigation data information is ephemeris information and almanac information for calculating the satellite position, and the satellite transmission time information is an elapsed time from a predetermined reference time of the satellite system. The time system of the satellite transmission time information is a GPS time system in the case of GPS, and a GLONASS time system in the case of GLONASS. The various correction parameters are satellite clock correction information, ionospheric correction information, and leap second correction information.

  The content of the correction parameter differs between GPS and GLONASS, and the leap second correction information can be extracted every 12.5 minutes with GPS, but there is no leap second correction information in the GLONASS correction parameter. In an environment where reception is poor, such as traveling in urban areas where radio waves are easily blocked or signal strength is degraded, satellite signals may be received, but may not be demodulated. For this reason, the GPS satellite signal demodulation units 18-1 to 18-N cannot always acquire leap second correction information every 12.5 minutes. In the present invention, the leap second correction information 19 extracted from the GPS correction parameter is not used, but the newly provided leap second correction value calculation unit 19 is used.

  The leap second correction value calculation unit 19 will be described later with reference to FIGS. 2 and 3. The own position measurement unit 20 with hard delay difference estimation will be described later with reference to FIG.

  The leap second correction value calculation unit 19 acquires GLONASS satellite transmission time information of the GLONASS time system output from the GLONASS satellite signal demodulation units 17-1 to 17-N, and acquires GPS satellite signal demodulation units 18-1 to 18-. GPS satellite transmission time information of the GPS time system output from N is acquired, the acquired GLONASS satellite transmission time information of the GLONASS time system is converted to UTC (Coordinated Universal Time), and the GLONASS of the converted GLONASS time system is converted. The leap second correction value is calculated by comparing the satellite transmission time information with the GPS satellite transmission time information of the GPS time system.

The relationship of each time system is shown in FIG. UTC is a universal time based on an international agreement. The leap second adjustment is performed for the GLONASS time and UTC. On the other hand, leap second adjustment is not performed for GPS time. Recently, +1 second has been inserted into the GLONASS time and UTC for the leap second adjustment on the end of June 2012, but +1 second has not been inserted into the GPS time. If 3 hours are subtracted from the GLONASS time, it can be converted to UTC. Furthermore, if the difference between UTC and GPS time is calculated, the leap second correction value Δt _LS can be calculated.

Therefore, the leap second correction value calculation unit 19 calculates UTC by subtracting 3 hours from the GLONASS satellite transmission time of the GLONASS time system, and subtracts UTC from the GPS satellite transmission time of the GPS time system, thereby correcting the leap second. The value Δt _LS can be calculated. Here, the leap second correction value calculation unit 19 can acquire the GLONASS satellite transmission time of the GLONASS time system and the GPS satellite transmission time of the GPS time system in real time by receiving the GLONASS and GPS satellite signals. it can. Therefore, the leap second correction value calculation unit 19 can calculate the leap second correction value Δt _LS in real time.

A flowchart showing the leap second correction value calculation processing of the present invention is shown in FIG. The leap second correction value calculation unit 19 acquires GLONASS satellite transmission time information output from the GLONASS satellite signal demodulation units 17-1 to 17-N as GLONASS time information, and uses GPS satellite signals as GPS time information. The GPS satellite transmission time information output from the demodulating units 18-1 to 18-N is acquired, and the difference amount after compensation of the three-hour time difference between the GLONASS time system and UTC is calculated for the GLONASS and GPS satellite transmission times. The leap second correction value Δt _LS is calculated by rounding off the first decimal place of the difference amount.

  This will be specifically described below. It is determined whether the shared positioning of GPS and GLONASS is possible (step S1). If it is determined that the shared positioning is possible (YES in step S1), the process proceeds to step S2. If it is determined that shared positioning is not possible (NO in step S1), this process is terminated. Hereinafter, steps S2 to S5 will be described.

GLONASS satellite transmission time information output from GLONASS satellite signal demodulation units 17-1 to 17-N and GPS satellite transmission time information output from GPS satellite signal demodulation units 18-1 to 18-N are acquired. (Step S2). As shown in FIG. 2, the GLONASS satellite transmission time is converted from the GLONASS time system to UTC by subtracting 3 hours from the GLONASS satellite transmission time (step S3). As shown in FIG. 2, the difference between the converted GLONASS satellite transmission time and the GPS satellite transmission time is calculated (step S4). The leap second correction value Δt _LS is calculated by rounding off the first decimal place of the difference amount (step S5).

GLONASS and GPS satellite transmission times contain errors on the order of much less than one second in the GLONASS and GPS time systems. Therefore, for the satellite transmission time of GLONASS and GPS, the difference amount after compensation of the time difference (3 hours) between the GLONASS time system and UTC is calculated, and by rounding off the first decimal place of the difference amount, the leap second error is calculated. A non-leap second correction value Δt _LS can be calculated.

  In the above description, the leap second correction value calculation unit 19 subtracts 3 hours from the GLONASS satellite transmission time of the GLONASS time system to convert it to the UTC GLONASS satellite transmission time, and converts the UTC GLONASS satellite transmission time to the GPS. A leap second correction value is calculated by comparing the time-based GPS satellite transmission time.

  As a modification, the leap second correction value calculating unit 19 adds 3 hours to the GPS satellite transmission time of the GPS time system, and adds the GPS satellite transmission time of 3 hours to the GLONASS satellite transmission time of the GLONASS time system. May be used to calculate the leap second correction value.

  More generally, the leap second correction value calculation unit 19 compares the GLONASS satellite transmission time of the GLONASS time system and the GPS satellite transmission time of the GPS time system after compensating for the time difference from the standard time, thereby calculating the leap second correction value. May be calculated.

  The own position measurement unit 20 with hard delay difference estimation is output from the GLONASS satellite information output from the GLONASS satellite signal demodulation units 17-1 to 17-N and the GPS satellite signal demodulation units 18-1 to 18-N. Using GPS satellite information and leap second correction information acquired by the leap second correction value calculation unit 19, the difference is caused by a difference in pass bandwidth between the bandpass filter 13 for the GLONASS satellite signal and the bandpass filter 14 for the GPS satellite signal. By solving the positioning equation in which the hard delay difference is an unknown amount, the hard delay difference is estimated and the position of the aircraft R is measured.

The pseudo distance, which is the observation amount of the positioning equation, is output from the GLONASS satellite transmission times output from the GLONASS satellite signal demodulation units 17-1 to 17-N and from the GPS satellite signal demodulation units 18-1 to 18-N. It is calculated from Equation 1 and Equation 2 from the GPS satellite transmission time.

ρ _GPS is a pseudo distance between the GPS satellite and the own aircraft R. ρ _GLO is a pseudorange between the GLONASS satellite and the aircraft R. c is the speed of light. t _u is the time of the clock of own device R. t _GPS and t _GLO are GPS and GLONASS satellite transmission times, respectively. S _i ≡ [x ^Si y ^Si z ^Si ] ^T is the position of the i-th satellite (GPS satellite or GLONASS satellite). u≡ [x _u y _u z _u ] ^T is the position of the aircraft R.

In the present embodiment, further, ionospheric delay, tropospheric delay, clock error of own aircraft R, clock error of satellite, and white Gaussian noise are considered. The pseudoranges observed by receiving GPS and GLONASS satellite signals are expressed by Equations 3 and 4.

r is the geometric distance between the i-th satellite and the aircraft R. δI is an ionospheric delay between the i-th satellite and the own aircraft R. δT is a tropospheric delay between the i-th satellite and own aircraft R. .DELTA.t _u is a clock error of its own R. δt ^Si is the clock error of the i-th satellite. e is white Gaussian noise between the i-th satellite and the own aircraft R, the average is 0, and the variance is σ ² .

Δt _LS is a leap second correction value calculated by the leap second correction value calculation unit 19. Δt _UTC-GLO is a time difference of 3 hours between the UTC and GLONASS time systems. δb _GLO is a hard delay difference caused by a difference in pass bandwidth between the bandpass filter 13 for the GLONASS satellite signal and the bandpass filter 14 for the GPS satellite signal.

幾何学距離ｒを線形化するために、自機Ｒの概略位置ｕ_０の周りで、１次の項までテーラー展開する（数６）。

ここで、ｇは、方向余弦行列と呼び、ｉ番目の衛星の位置Ｓ_ｉ及び自機Ｒの概略位置ｕ_０の間の幾何学距離の勾配ベクトルの転置行列である（数７）。

The geometric distance r is a nonlinear function based on the positions of the i-th satellite and the own aircraft R (Equation 5).

In order to linearize the geometric distance r, Taylor expansion is performed up to the first order term around the approximate position u ₀ of the aircraft R (Equation 6).

Here, g is called a direction cosine matrix, and is a transposed matrix of a gradient vector of a geometric distance between the position S _i of the i-th satellite and the approximate position u ₀ of the own aircraft R (Equation 7).

The unknown quantities in Equations (3), (4), and (6) are the position u of own aircraft R, the clock error δt _u of own aircraft R, and the hard delay difference δb _GLO in GLONASS satellite signal and GPS satellite signal within own device R. It is. In addition, the observed amount and the known amount in Equations 3, 4, and 6 are the GPS and GLONASS pseudoranges ρ _GPS and ρ _GLO , the satellite position S _i , the approximate position u ₀ of the own aircraft R, the geometry The distance r, the tropospheric delay δT, the ionospheric delay δI, the satellite clock error δt ^Si , the leap second correction value Δt _LS, and the time difference Δt _UTC-GLO between the GLONASS time system and UTC.

Here, the satellite position S _i and the satellite clock error δt ^Si are navigation data information (ephemeris information) output from the GLONASS satellite signal demodulation units 17-1 to 17-N and the GPS satellite signal demodulation units 18-1 to 18-N. ) And the correction parameter, the tropospheric delay δT is calculated using a well-known model equation, and the ionospheric delay δI uses the correction parameter output from the GPS satellite signal demodulation units 18-1 to 18-N. Is calculated.

ここで、ｙは観測ベクトル、Ｈは計画行列、θは未知ベクトル、εは誤差ベクトルであり、ｙ、Ｈ、θ、εは、それぞれ、数９〜数１２で表される。

When the observed amount and the known amount obtained by receiving the GPS and GLONASS satellite signals are collected on the left side, and the unknown amount to be obtained is collected on the right side, the positioning equation is rewritten as in Expression 8.

Here, y is an observation vector, H is a design matrix, θ is an unknown vector, ε is an error vector, and y, H, θ, and ε are expressed by Equation 9 to Equation 12, respectively.

The first satellite is a GPS satellite. The i-th satellite is a GLONASS satellite. Here, when the number of captured GPS satellites is m and the number of captured GLONASS satellites is n, the observation vector y and the error vector ε are a matrix of m + n rows and one column. The direction cosine matrix g is a 1 × 3 matrix, and thus the design matrix H is an m + n × 5 matrix. Position u and approximate position _{u 0} of its own R is a vector of three rows and one column, clock error .DELTA.t _u of its own R is first row and first column, a hard delay difference [delta] b _GLO is one row and one column Therefore, the unknown vector θ is a 5 × 1 vector.

ここで、Ｐは、観測量の分散値で構成される重み行列である。また、数１１の自機Ｒの概略位置ｕ_０が不明な時は、自機Ｒの概略位置を地球中心（ゼロ点）にして数１３を解き、数１３を解いて得られた自機Ｒの推定位置ｕを自機Ｒの概略位置ｕ₀に設定して再び数１３を解くという、数１３の繰り返し計算を実行するようにする。ここでは最小二乗法で未知ベクトルを推定する（解く）方法を説明したが、カルマンフィルタで未知ベクトルを推定するようにしても良い。カルマンフィルタは未知量ベクトルの時間変化モデルが適切に設計できれば最小二乗法より未知量の推定精度が良くなるという特徴がある。 In order to estimate the unknown vector θ of the positioning equation of Equation 8, the least square method of Equation 13 is used.

Here, P is a weight matrix composed of variance values of observation amounts. Further, when the approximate position u ₀ of the own aircraft R in Equation 11 is unknown, the own device R obtained by solving Equation 13 with the approximate position of the own device R being the center of the earth (zero point) and solving Equation 13 is obtained. The estimated position u is set to the approximate position u ₀ of the own aircraft R, and the equation 13 is solved again to execute the equation 13. Although the method for estimating (solving) an unknown vector by the least square method has been described here, the unknown vector may be estimated by a Kalman filter. The Kalman filter is characterized in that the estimation accuracy of the unknown quantity is better than that of the least square method if the time variation model of the unknown quantity vector can be appropriately designed.

  FIG. 4 is a flowchart showing the own position measurement process with hardware delay difference estimation according to the present invention. Before solving the positioning equation, the own position measurement unit with hard delay difference estimation 20 determines whether the estimation accuracy of the hard delay difference caused by the difference in the pass bandwidth of the GLONASS and GPS satellite signals is good or bad. When the estimation accuracy is expected to be good, the position equation and the hard delay difference of the aircraft R are estimated by solving the positioning equation using the hard delay difference as an unknown quantity. When the estimation accuracy of the hard delay difference is expected to be poor, the estimation is performed. The position of the aircraft R is estimated by solving the positioning equation using the backup value or the default value with better accuracy than the hard delay difference.

DOP (Dilution of Precision) is used as an evaluation index of the estimation accuracy of the hard delay difference. DOP is used as an evaluation index of the estimation accuracy of the clock error .DELTA.t _u positions u and own apparatus R of its own R estimated by solving the positioning equations, in the present invention, evaluation of the estimation accuracy of the hard delay difference Extend to be used as an indicator. Whether the estimation accuracy is good or bad is determined based on the value of the evaluation index.

The weight matrix P of Equation 13 is assumed to be P = I. The variance of the observation vector y in Equation 9 is assumed to be Cov (y) = I. Here, Cov (y) means the variance of the variable y, and I is a unit matrix. The variance of the unknown vector θ in Expression 11 is expressed by Expression 14.

ＧＤＯＰ（Ｇｅｏｍｅｔｒｉｃａｌ ＤＯＰ）、ＰＤＯＰ（Ｐｏｓｉｔｉｏｎ ＤＯＰ）、ＴＤＯＰ（Ｔｉｍｅ ＤＯＰ）及びＧＧＴＤＯＰ（ＧＰＳ ｔｏ ＧＬＯＮＡＳＳ Ｔｉｍｅ ＤＯＰ）は、それぞれ、数式１６〜１９で表される。

The trace of the variance Cov (θ) of the unknown vector θ is expressed as in Expression 15.

GDOP (Geometric DOP), PDOP (Position DOP), TDOP (Time DOP), and GGTDOP (GPS to GLONASS Time DOP) are represented by Equations 16 to 19, respectively.

  When the TDOP of Expression 18 and the GGTDOP of Expression 19 are smaller than a predetermined threshold (for example, 3), the estimation accuracy of the hard delay difference is improved. When Equation 18 TDOP or Equation 19 GGTDOP is larger than a predetermined threshold (for example, 3), the estimation accuracy of the hard delay difference is deteriorated. When it is determined that the estimation accuracy of the hard delay difference is poor, the positioning equation is solved using the hard delay difference estimated with a good accuracy in the past.

  This will be specifically described below. A positioning equation of Formula 8 is established (step S11). It is determined whether or not the number of captured satellites is equal to or greater than a threshold value (here, a number equal to the number of rows of the unknown vector θ) (step S12). It is determined whether the TDOP of Equation 18 and the GGTDOP of Equation 19 are less than a threshold (for example, 3) (step S13). The following processing is performed according to the determination result.

  If the number of acquired satellites is equal to or greater than the threshold (YES in step S12) and TDOP of equation 18 and GGTDOP of equation 19 are less than the threshold (YES in step S13), it is determined that the estimation accuracy of the hard delay difference is good. To do. Therefore, the positioning equation (8) in which the estimated value of the hard delay difference is an unknown quantity is solved by the least square method (step S14). After solving the positioning equation, the estimated hard delay difference is backed up (step S15).

  Even if the number of captured satellites is greater than or equal to the threshold (YES in step S12), if the TDOP of equation 18 or GGTDOP of equation 19 is greater than or equal to the threshold (NO in step S13), the estimation accuracy of the hard delay difference is poor. to decide.

  If it is determined that the estimation accuracy of the hard delay difference is poor, if there is a backed-up hard delay difference (YES in step S16), the positioning equation is solved using the backed-up hard delay difference (step S17), and the backed-up hard delay If there is no difference (NO in step S16), the positioning equation is solved using the default hard delay difference (step S18). Here, the default value is a value measured in advance under good conditions at the time of factory shipment or receiver development, and is a value written in the ROM as an initial value.

When it is determined that the estimation accuracy of the hard delay difference is poor, the backup equation or the default value is used as the hard delay difference to solve the positioning equation using the hard delay difference that is an unknown amount in Equation 8 as the known amount. (Step S17 or Step S18). Specifically, in Equation 8, the observation vector y, the design matrix H, and the unknown vector θ are reconfigured according to the following Equations 20 to 22, respectively, and the positioning equation is solved.

  If the number of captured satellites is less than the threshold value (NO in step S12), since the number of observations is smaller than the number of unknowns, the positioning equation of Equation 8 cannot be solved. If there is no backed-up hard delay difference (NO in step S16), the back-up hard delay difference is solved (step S17). The positioning equation composed of Equations 20 to 22 is solved using the hard delay difference of the default value (Step S18).

  The case where there is no backed-up hardware delay difference (NO in step S16) corresponds to the time of factory shipment in which the backup memory is cleared or the time when backup information is erased due to battery exhaustion. For this reason, the situation where there is no backed-up hardware delay difference does not occur frequently.

  When the estimation accuracy of the hard delay difference is expected to be good, the hard delay difference is accurately estimated using the hard delay difference as an unknown amount in the positioning equation. When the estimation accuracy of the hard delay difference is expected to be poor, a known amount is used by using the past hard delay difference with a high estimation accuracy to some extent in the positioning equation. If the estimation accuracy of the hard delay difference is poor, the positioning error also deteriorates, but the positioning error can be reduced by determining whether the estimation accuracy of the hard delay difference is good or bad before solving the positioning equation.

  The GLONASS receiver of the present invention obtains leap second correction information without a leap second error without increasing the circuit scale, accurately estimates a hard delay difference, prevents positioning abnormality, and reduces positioning error. It is possible to apply for the purpose.

R: GLONASS receiver A: antenna 11, 12: down converter 13, 14: band pass filters 15-1 to 15-N: GLONASS satellite signal reception channels 16-1 to 16-N: GPS satellite signal reception channel 17-1 -17-N: GLONASS satellite signal demodulator 18-1 to 18-N: GPS satellite signal demodulator 19: Leap second correction value calculator 20: Own position measurement unit with hard delay difference estimation

Claims (3)

A GLONASS satellite signal demodulator that demodulates a GLONASS satellite signal received from a GLONASS (Global Navigation Satellite System) satellite and acquires GLONASS satellite transmission time information of a GLONASS time system;
A GPS satellite signal demodulator that demodulates a GPS satellite signal received from a GPS (Global Positioning System) satellite and acquires GPS satellite transmission time information of a GPS time system;
Compensation of standard time difference between GLONASS satellite transmission time information of the GLONASS time system acquired by the GLONASS satellite signal demodulation unit and GPS satellite transmission time information of the GPS time system acquired by the GPS satellite signal demodulation unit A leap second correction value calculation unit for calculating a leap second correction value by comparing later,
A GLONASS receiver comprising: The pseudo distance and geometric distance between the GLONASS satellite and the own aircraft, the pseudo distance and geometric distance between the GPS satellite and the own aircraft, and the leap second correction value calculated by the leap second correction value calculating unit are observed amounts. The hard delay difference is estimated by solving the positioning equation with the hard delay difference that is the time constant difference between the GLONASS satellite signal and the GPS satellite signal as a time constant of the bandpass filter in the own apparatus, Its own position measurement unit with hard delay difference estimation to measure the position of
The GLONASS receiver of claim 1, further comprising: The own position measurement unit with the hard delay difference estimation,
The estimation accuracy of the hard delay difference is calculated based on an unknown dispersion calculation formula of the positioning equation,
When the estimation accuracy of the hard delay difference calculated based on the unknown dispersion calculation formula of the positioning equation is less than a predetermined threshold, the position of the own device is measured using the hard delay difference as an unknown amount in the positioning equation. ,
When the estimation accuracy of the hard delay difference calculated based on the unknown dispersion calculation formula of the positioning equation is equal to or higher than the predetermined threshold, the backup equation or the default hard delay difference is used in the positioning equation. The GLONASS receiver according to claim 2, wherein the position of the machine is measured.

Images