JP2014041025A - Multi-path estimation device, multi-path detection device, gnss receiver, multi-path estimation method, information terminal appliance and multi-path estimation program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a multi-path estimation device, etc. with small hardware resources and software resources.SOLUTION: A register 330 generates a replica code E advancing to a prompt replica code P by 0.05 chip, a replica code L delayed by 0.05 chip, a replica code VE advancing by 0.1 chip and a replica code VL delayed by 0.1 chip. A correlation processing section 30 generates five correlation values out of correlation processing of the replica codes VE,E,P,L,VL and a C/A code. An estimation section 400 estimates about a multi-path signal from a coefficient value of a linear term of a polynomial of the fourth degree approximated by each phase offset value (-0.1,-0.05,0,0.05,0.1) and each correlation value.

Description

  The present invention relates to a multipath estimation device for estimating a multipath signal generated when a GNSS (Global Navigation Satellite System) signal is received from a satellite, and a multipath detection device for detecting a multipath signal. The present invention relates to a multipath estimation method, a multipath estimation program, a GNSS receiver and an information terminal device to which they are applied.

  GNSS signals transmitted from multiple satellites used in the global positioning system are phased with pseudo-noise codes (PRN codes) such as C (Coarse) / A (Acquisition) codes for positioning. It is modulated. Since each satellite is assigned a PRN code unique to the satellite, the satellite can be identified from the PRN code of the demodulated GNSS signal, and the pseudorange to the identified satellite can be known. Therefore, it is possible to determine the position of the GNSS receiver that has received the GNSS signal with reference to the specified satellite. Such a GNSS receiver is built in, for example, a car navigation device or a mobile phone, and constitutes a positioning system that measures the position of these devices.

  In order to identify the satellite and know the pseudorange, the GNSS receiver generates, for example, a replica code of the C / A code of the GNSS signal to be received. Then, a correlation operation between the C / A code demodulated from the GNSS signal and the replica code is performed, a code phase offset value at the head of the replica code having the maximum correlation value is found, and a peak (tracking point) of the correlation value is obtained. To track.

  However, as described in Patent Document 1 (Japanese Patent Laid-Open No. 2009-159261), the GNSS signal received by the GNSS receiver is not only transmitted directly from the satellite, but also a direct wave is transmitted from the building. Indirect waves such as reflected waves and diffracted waves generated by reflection are also included. Since these indirect waves that have passed through paths other than the direct path (multipath) are multipath signals, the multipath signal also includes the same C / A code as the GNSS signal. Therefore, when a multipath signal is superimposed on a GNSS signal received directly through the direct path, the autocorrelation function of the C / A code and the replica code is distorted, and a phase error occurs in the code phase offset value at the tracking point. Arise. This phase error causes an error in the pseudorange, and the positioning accuracy is lowered.

  Since the phase of the C / A code of the directly received GNSS signal also changes, it is difficult to detect the presence or absence of a multipath signal. However, if the presence or absence of a multipath signal can be determined, positioning accuracy can be obtained by using a GNSS signal that does not generate a multipath signal or by reducing the weight in the calculation of a GNSS signal that generates a multipath signal in the positioning calculation. Can be improved.

  An object of the present invention is to add a multipath signal detection function to a device such as an inexpensive GNSS receiver for consumers, which requires a multipath estimation device and multipath that require less hardware resources and software resources. It is to provide a detection apparatus, a multipath estimation method, and a multipath estimation program, and to provide a GNSS receiver and an information equipment terminal to which they are applied.

  A multipath estimation apparatus according to the present invention is a multipath estimation apparatus that estimates a multipath signal of a GNSS signal modulated using a pseudo noise code, and has a prompt having a phase corresponding to the phase of the pseudo noise code. A replica code, a first early replica code that is advanced by a first phase relative to a prompt replica code, a first late replica code that is delayed by a first phase, a second early replica code that is advanced by a second phase, and a second that is delayed by a second phase A replica code generating unit that generates a two-rate replica code, a correlation processing unit that correlates each replica code and pseudo-noise code generated by the replica code generating unit to calculate each correlation value, and a phase of each replica code 4th order or higher order polynomials approximated using each correlation value corresponding to the quantity and phase of each replica code Those comprising an estimation unit for estimating about signal strength and / or phase of the multipath signal by using the coefficient values of the definitive predetermined section.

  A multipath detection apparatus according to the present invention includes the above-described multipath estimation apparatus and a detection unit that detects the presence / absence of a multipath signal using coefficient values.

  The GNSS receiver according to the present invention also receives an antenna that receives a GNSS signal modulated using a pseudo-noise code, down-converts the GNSS signal, performs AD conversion, and outputs digital received data including the pseudo-noise code. A high-frequency processing unit, a positioning unit that performs a positioning calculation from a pseudo distance obtained based on a pseudo-noise code, and the above-described multi-path detection device that detects the presence or absence of a multi-path signal from digital reception data output from the high-frequency processing unit; The positioning unit is configured to optimize the positioning calculation according to the signal strength and / or phase of the multipath signal detected by the multipath detection device.

  An information equipment terminal according to the present invention includes the above-described GNSS receiver and a display device that performs display based on positioning information obtained by positioning calculation performed by the GNSS receiver.

  The multipath estimation apparatus, multipath detection apparatus, GNSS receiver, and information equipment terminal of the present invention include a replica code generation unit, a correlation processing unit, and an estimation unit as hardware resources. Among these, the replica code generation unit and the correlation processing unit are hardware resources for demodulating the pseudo noise code from the GNSS signal, for example, and thus can be shared with the GNSS receiver. Moreover, since the estimation unit has a simple configuration in which estimation is performed using a coefficient value of a predetermined term of a polynomial, the estimation unit can be configured with few hardware resources. Thereby, in a GNSS receiver or an information equipment terminal, a multipath estimation apparatus can be provided with a small amount of hardware resources. The multipath detection device, the GNSS receiver, and the information equipment terminal can be easily downsized.

  The multipath estimation method according to the present invention is a multipath detection method for estimating a multipath signal of a GNSS signal modulated using a pseudo noise code, and has a prompt having a phase corresponding to the phase of the pseudo noise code. A replica code, a first early replica code that is advanced by a first phase relative to a prompt replica code, a first late replica code that is delayed by a first phase, a second early replica code that is advanced by a second phase, and a second that is delayed by a second phase A replica code generating step for generating a two-rate replica code, a correlation processing step for calculating each correlation value by correlating each replica code and pseudo-noise code generated by the replica code generating unit, and a phase of each replica code 4th order approximated with each correlation value corresponding to the quantity and phase of each replica code In which and a estimation step for estimating about signal strength and / or phase of the multipath signal by using a coefficient value of a predetermined term in the polynomial above.

  In the multipath estimation method of the present invention, the replica code generation step and the correlation processing step for obtaining each replica code and each correlation value are necessary steps for demodulating the pseudo noise code from the GNSS signal. Alternatively, if each replica code or each correlation value is obtained from the information equipment terminal, the arithmetic processing is not increased as compared with the conventional case. Further, in the estimation step, only a simple process of estimating with the coefficient value of the predetermined term of the polynomial is performed, so that the calculation process is reduced. Thereby, a navigation apparatus having a GNSS receiver or the like can provide a multipath estimation method with a small amount of software resources.

  ADVANTAGE OF THE INVENTION According to this invention, the hardware resource and software resource which are required for the estimation of a multipath signal can be reduced, and the function which estimates a multipath signal can be added to a consumer-oriented GNSS receiver or information equipment terminal at low cost.

The block diagram for demonstrating the GNSS receiver with which the multipath detection apparatus concerning 1st Embodiment of this invention is applied. 1 is a block diagram illustrating an example of a configuration of a multipath detection device according to a first embodiment. The graph for demonstrating the result of the correlation process of a C / A code and a replica code. The block diagram which shows an example of a structure of the multipath estimation apparatus of a processor. 6 is a flowchart illustrating an example of a multipath signal detection procedure according to the first embodiment. The graph which shows the simulation result of the delay of a multipath signal, and a pseudorange error. Graph showing the relationship between the delay and the normalized coefficient values a ₁ / a ₀ of the multipath signal. The block diagram which shows an example of a structure of the multipath detection apparatus which concerns on 2nd Embodiment. The block diagram which shows an example of a structure of the multipath estimation apparatus of a processor. 9 is a flowchart showing an example of a multipath signal detection procedure according to the second embodiment. The graph which shows the relationship between the delay of a multipath signal, and a phase difference. Graph showing the relationship between the delay and the normalized coefficient values a ₁ / a ₀ of the multipath signal.

<First Embodiment>
The multipath detection apparatus according to the first embodiment will be described below with reference to FIGS. The multipath detection apparatus is used in the GNSS receiver 10 shown in FIG. 1, for example.

(1) Outline of GNSS Receiver A GNSS receiver 10 in FIG. 1 includes an antenna 1, an RF (Radio Frequency) unit 2, and a baseband unit 3.

  In the GNSS receiver 10, the antenna 1 receives a high frequency signal (GNSS signal) transmitted from the satellite 60. The high-frequency signal transmitted from the satellite 60 and demodulated by the GNSS receiver 10 has a transmission frequency of L1 band (center frequency 1575.42 MHz) and is modulated with a C / A code (first code). However, the GNSS receiver 10 has the same configuration in the case of a high-frequency signal having a frequency such as the L2 band (center frequency 1227.6 MHz) or a high-frequency signal modulated by another PRN code such as a P code. It is possible. In FIG. 1, for convenience of explanation, only one satellite is shown.

  The RF unit 2 takes in the high-frequency signal RF output from the antenna 1, down-converts the high-frequency signal RF to an intermediate frequency, and converts it into a 1-bit (binary) digital signal. In this embodiment, as an example, the case where the RF unit 2 converts into a 1-bit digital signal is shown, but the same configuration can be used when converting into a 2-bit digital signal. The RF unit 2 outputs the converted intermediate frequency signal (GNSS signal) IF to the baseband unit 3 at the subsequent stage. Further, the RF unit 2 generates a sampling clock for the baseband unit 3 (hereinafter referred to as “clock CLK”) and outputs the sampling clock to the baseband unit 3.

  The baseband unit 3 captures and tracks the satellite from the correlation process of the C / A code of the intermediate frequency signal IF output from the RF unit 2. Then, based on the navigation message, time information, and the like that are extracted by decoding the intermediate frequency signal IF of the tracking satellite, the baseband unit 3 performs a pseudorange calculation operation, a positioning operation, and the like.

  In FIG. 1, in order to track a large number of satellites, the baseband unit 3 includes signal processing units 31 corresponding to the channels 1, 2,..., N for receiving the intermediate frequency signal IF, and 1 One processor 35 is provided. For example, the value of N is 16, and each of the 16-channel signal processing units 31 can simultaneously receive the intermediate frequency signal IF output from the RF unit 2 and perform correlation processing. Each channel is similarly configured except that it uses a different PRN code. This correlation process will be described later in detail.

  The processor 35 is communicably connected to each signal processing unit 31 in the 16 channels. The processor 35 is further incorporated in the information equipment terminal 70. The information equipment terminal 70 uses a positioning result of a GNSS receiver such as a car navigation device. If the terminal receives the signal from the processor 35 and uses the positioning result, for example, it is a device such as a wristwatch or a mobile phone that does not directly provide position information to the user using the positioning result inside the device. Alternatively, a processing device that does not specify the use of position information, such as a PDA (Personal Digital Assistant) that simply processes positioning results, may be used. The information equipment terminal 70 of this embodiment includes an information equipment 71 for using positioning information, such as a CPU, a memory, a display device, and an input device.

  In FIG. 1, the processor 35 includes a ROM 351 and a RAM 352. The ROM 351 can be configured separately from the processor 35 and externally attached to the processor 35.

(2) Configuration of Baseband Unit FIG. 2 is a block diagram for explaining the configuration of each signal processing unit 31 and processor 35 configuring the baseband unit 3. In the following description, each signal processing unit is simply referred to as the signal processing unit 31 in the description common to each signal processing unit 31. The signal processing unit 31 can be realized by a digital circuit such as an ASIC (Application Specific Integrated Circuit).

(2-1) Configuration of Correlation Processing Unit The signal processing unit 31 obtains the in-phase component and the quadrature component from the intermediate frequency signal IF, and performs correlation processing for obtaining a correlation result of each component described later. For this purpose, in the signal processing unit 31, as shown in FIG. 2, the correlation processing unit 30 includes a pair of multipliers 301 and 302, a SIN map 303, a COS map 304, a carrier wave NCO (Numerically Controlled Oscillator) 305, Multipliers 310 to 319 and ten correlators 323 to 336 are included. Each correlator is, for example, an integration and dump filter.

  Further, the signal processing unit 31 includes a register 330, a code generator 331, and a code NCO (Numerically Controlled Oscillator) 332.

  The correlation processing unit 30 in FIG. 2 shows a case where a pair of multipliers 301 and 302, a SIN map 303, a COS map 304, a carrier wave NCO 305, multipliers 310 to 319, and correlators 323 to 336 are provided. The pair of multipliers 301 and 302, the SIN map 303, the COS map 304, and the carrier wave NCO 305 may be configured separately from the correlation processing unit 30.

  The multiplier 301 multiplies the intermediate frequency signal IF and the sine signal output from the SIN map 303 in order to extract the in-phase signal I having the in-phase component from the intermediate frequency signal IF. The sine signal is an in-phase component signal of a carrier wave (hereinafter referred to as “replica carrier wave”) generated by the carrier wave NCO 305, and the SIN map 303 has a known function for extracting the in-phase component sine signal from the replica carrier wave. The sine signal extracted by this function is output to the multiplier 301.

  The carrier NCO 305 receives from the processor 35 a carrier phase control amount φ1 for controlling the phase or frequency of the replica carrier in order to match the phase or frequency between the intermediate frequency signal IF and the replica carrier. When the carrier wave NCO 305 receives the carrier phase control amount φ 1, the carrier wave NCO 305 generates a replica carrier wave determined based on the carrier phase control amount φ 1 and outputs the replica carrier wave to the SIN map 303 and the COS map 304. Note that the clock CLK described above is used as the sampling clock of the carrier wave NCO 305.

  The COS map 304 has a known function for extracting an orthogonal component cosine signal from a replica carrier wave, and outputs the cosine signal extracted by this function to the multiplier 302.

  The multiplier 302 multiplies the intermediate frequency signal IF and the cosine signal output from the COS map 304 in order to extract the orthogonal signal Q of the orthogonal component from the intermediate frequency signal IF.

  Each multiplier 310 to 314 multiplies the in-phase signal I output from the multiplier 301 and each C / A code (hereinafter referred to as “replica code”) output from the register 319. The replica code means a replica of the C / A code included in the intermediate frequency signal IF. Each multiplier 315 to 319 multiplies the orthogonal signal Q output from the multiplier 302 and each replica code output from the register 330.

  In this embodiment, the case where the register 330 is a 5-bit register will be described as an example. Therefore, the register 330 synchronizes one replica code output from the code generator 331 in synchronization with the clock CLK (accordingly, ) By latching and outputting, as a whole, a replica code having five types of phases is output. In other words, as shown in FIG. 2, the register 330 outputs the replica codes VE, E, P, L, and VL in the order of the phase. In this case, the replica codes VE and E indicate an early phase replica code, the prompt replica code P indicates a regular replica code, and the replica codes L and VL indicate a late phase replica code. In the description of this embodiment, as an example, the phase difference (interval) between the five types of replica codes is set to about 0.05 chips (50 nsec).

  A pair of multipliers (multiplier on the in-phase signal I side and multiplier on the quadrature signal Q side) is connected to each output port provided in the register 330. Thereby, each replica code VE, E, P, L, VL is output to a corresponding pair of multipliers. For example, the replica code VE is output to the pair of multipliers 310 and 315.

  Here, the code generator 331 generates one replica code at the timing when the output signal of the code NCO 332 is input, and outputs it to the register 330. The code NCO 332 receives, from the processor 35, a code phase control amount φ2 for controlling the phase of the replica code so as to make the phase between the C / A code included in the intermediate frequency signal IF and the replica code coincide. The code phase control amount φ2 is, for example, a value indicating an increase in phase per clock. When the code NCO 332 receives the code phase control amount φ2, the code NCO 332 outputs a signal to the code generator 331 at a timing determined based on the code phase control amount φ2.

  Each correlator 320 to 324 receives a signal output from each of the multipliers 310 to 314. Each correlator 320 to 324 integrates the output of the corresponding multiplier by one bit and outputs the result to the processor 35. Further, each correlator 325 to 329 receives a signal output from each of the multipliers 315 to 319. Each correlator 325 to 329 integrates the output of the corresponding multiplier by one bit and outputs the result to the processor 35. Thereby, a correlation result between the C / A code and the replica code is obtained for each of the in-phase signal I and the quadrature signal Q.

  Note that the type of replica code (that is, the number of phases of the replica code output from the register) may be configured to be five or more as described above.

(2-2) Configuration of Processor Next, the processor 35 will be described. The processor 35 executes processing for capturing and tracking the intermediate frequency signal IF (GNSS signal), setting of a code phase control amount, acquisition of a navigation message, positioning calculation, multipath detection, and the like. Therefore, in this embodiment, as illustrated in FIG. 2, the processor 35 includes a control amount setting unit 353, a navigation message acquisition unit 354, a positioning calculation unit 355, an estimation unit 400, and a multipath presence / absence detection unit 403. These functions are realized, for example, when the processor 35 reads a program recorded in the ROM 351 and operates according to the program.

  The tracking process is performed for the purpose of continuously tracking the correlation peak (tracking point) while following the intermediate frequency signal IF. The tracking process includes a C / A code tracking process and a carrier frequency tracking process.

  The tracking process of the C / A code is performed by a code tracking error (received C / A code and prompt replica) from the correlation values output from the correlators 320 to 324 and the correlators 325 to 326 by a known method. The code phase control amount φ2 is generated by the control amount setting unit 353 so that the phase difference of the code P is minimized (zero).

  The GNSS signal is a signal obtained by spectrum-spreading a navigation message with a C / A code unique to each satellite. The spectrum despreading process is performed on the intermediate frequency signal IF by the above-described correlation processing in the correlation processing unit 30, and the navigation message acquisition unit 354 performs navigation specific to the satellite from the GNSS signal while the C / A code is being tracked. Get the message.

  The positioning calculation unit 355 calculates the position of the GNSS receiver 10 and the like from the positions and pseudoranges of a plurality of satellites obtained from the observation amount of the satellite obtained from the GNSS signal and the ephemeris data of the navigation message, for example.

  The estimator 400, for each GNSS signal, in a tracked state, the replica code VE, E, P, L, VL phase offset value group obtained from the correlation processor 30 and each replica code VE, E, P, The multipath signal is estimated using a correlation value group composed of correlation values of L, VL and C / A code. Based on the estimation by the estimation unit 400, the multipath presence / absence detection unit 403 detects the presence / absence of a multipath signal. The determination as to whether or not tracking is performed is performed by the processor 35 by the same processing as in the prior art, and the above-described processing by the estimation unit 400 and the multipath presence / absence detection unit 403 is performed according to the result.

(3) Detection of multipath signal The tracking processing is performed by parallel processing of the signal processing unit 31 provided for each channel, as in the prior art. The code phase control amount φ2 is set by the control amount setting unit 355 so that the correlation value between the prompt replica code P and the C / A code is maximized in each signal processing unit 31. Here, it is preset in the register 330, and the code phase offset value is set to -0.1, -0.05, 0, 0.05, 0.1 chip.

  FIG. 3 is a graph showing the result of correlation processing between a replica code and a C / A code when an arbitrary GNSS signal is being tracked. The autocorrelation function shown in FIG. 3 is defined with reference to the code phase offset value of the prompt replica code P that gives the maximum value RP of the correlation value R (t). The correlation value R (t) (t = −0.05) between the replica code E and the C / A code whose chip phase is advanced by 0.05 chip from the prompt replica code P is RE, and is 0 from the prompt replica code P. The correlation value R (t) (t = −0.1) between the replica code VE advanced by 1 chip phase and the C / A code is RVE, which satisfies the relationship of RP> RE> RVE. Further, the correlation value R (t) (t = 0.05) between the replica code L and the C / A code delayed by 0.05 chip phase from the prompt replica code P is RL, and is more than the prompt replica code P. The correlation value R (t) (t = 0.1) between the replica code VL with a delay of 0.1 chip phase and the C / A code is RVL, which satisfies the relationship of RP> RL> RVL. Further, since the autocorrelation function shown in FIG. 3 is for the GNSS signal on which the multipath signal is not superimposed, the relationship of RE = RL and RVE = RVL is satisfied.

However, when the multipath signal is superimposed on the GNSS signal, the above relationship is lost. Therefore, if the general expression of the autocorrelation function is approximated by a quartic function, for example, it is given by the following expression.
R (t) = a ₀ + a ₁ · t + a ₂ · t ² + a ₃ · t ³ + a ₄ · t ⁴ (1)

Since each point of the above (t, R (t)) satisfies this general formula (1), (−0.1, RVE), (−0.05, RE), (0, RP), (0 .05, RL), (0.1, RVL), the coefficient values a _{0 to} a ₄ of the five terms can be obtained. As will be described later, the presence or absence of a multipath signal can be detected using the obtained coefficient.

(3-1) Configuration of Multipath Detection Device As described above, in order to detect the presence or absence of a multipath signal using a coefficient of a predetermined term of a fourth or higher order polynomial, the multipath detection device 50 includes a code NCO 332 and A code generator 331, a register 330, a correlation processing unit 30, an estimation unit 400, a multipath presence / absence detection unit 403, and a control amount setting unit 353 are provided. The estimation unit 400 includes a correlator output acquisition unit 401 and a coefficient calculation unit 2. In the multipath detection device 50, the code NCO 332, the code generator 331, the register 330, the correlation processing unit 30, and the estimation unit 400 perform estimation on the multipath signal of the GNSS signal modulated using the pseudo noise code. The multipath estimation device 51 is configured.

  Correlator output acquisition section 401 acquires correlation values RVE, RE, RP, RL, and RVL output from five correlators 320 to 324.

The coefficient calculation unit 2 includes a correlation value group including five correlation values RVE, RE, RP, RL, and RVL given from the correlator output acquisition unit 401, and a phase offset value − corresponding to each element of these correlation value groups − For example, coefficient values a _{0 to} a ₄ of a fourth-order polynomial are obtained using a phase offset value group consisting of 0.1, −0.05, 0, 0.05, and 0.1 chip. Here, the phase offset value group is a value preset in the register 330, and is stored in the RAM 352, for example. In such a case, the coefficient calculation unit 2 captures and holds the value of the phase offset value group from the RAM 352 simultaneously with the activation of the GNSS receiver 10.

Here, the phase difference between the elements of the phase offset value group arranged in order is set equal to 0.05 chip. Therefore, using the Savitzky-Golay method (hereinafter referred to as SGM), each coefficient value a _{0 to} a ₄ can be obtained by a simple method of product of a constant matrix and a correlation value. In the case of a quartic polynomial, the coefficient matrix determined by the SGM is a portion that is solved as in the following equation (2) and is a quartic square matrix.

The multipath presence / absence detection unit 403 detects the presence / absence of a multipath signal based on the coefficient values a _{0 to} a _{4 of} the polynomial calculated by the coefficient calculation unit 2.

(3-2) Detection of presence / absence of multipath signal Assuming an ideal state where the multipath signal is not superimposed on the GNSS signal and an error due to other factors does not occur in the GNSS signal, the fourth order polynomial is obtained. There are no first-order or third-order terms in the approximated autocorrelation function. That is, the correlation waveforms when multipath signal is not inserted has a symmetrical shape, this is because it indicates that the function is an even function, ₁ and 3 order coefficient value of the primary term a The coefficient value a ₃ of the term becomes zero. The coefficient value a ₁ and the coefficient value a ₃ increase as the signal strength of the superimposed multipath signal increases. Specifically, for example, when the autocorrelation function is approximated by a fourth-order polynomial, detection is preferably performed using the coefficient value a ₁ of the first-order term. This is because, when approximated by a fourth-order polynomial, when a multipath signal is superimposed, it appears remarkably in the coefficient value a ₁ of the first-order term.

Since it has been found that the above relationship can be used to detect the presence or absence of a multipath signal, whether or not the signal strength of the superimposed multipath signal is so high that the GNSS signal cannot be used for positioning calculation is set as a threshold value. Comparison can be made. That is, it is possible to normalize by dividing the coefficient values a ₁ of the first-order terms in the coefficient values a ₀ constant term, discriminating by comparing with a threshold value Th which is set in advance. At this time, if Th> | a ₁ / a ₀ |, it is determined that the multipath signal is not superimposed on the GNSS signal (no multipath signal), and if Th ≦ | a ₁ / a ₀ | It is determined that the signal is superimposed (there is a multipath signal). In other words, the determination of “no multipath signal” means that the signal strength of the multipath signal is sufficiently small so that the error due to the multipath signal can be ignored. On the contrary, the determination of “there is a multipath signal” means that the error due to the multipath signal cannot be ignored because the signal strength of the multipath signal is large.

  The procedure up to the detection of the presence or absence of the multipath signal is summarized as shown in the flowchart of FIG.

  First, after capturing the intermediate frequency signal IF, the control amount setting unit 353 of the processor 35 determines the code phase control amount φ2 based on the result of correlation processing output from the signal processing unit 31 (output values of the correlators 320 to 329). The tracking processing of the intermediate frequency signal IF is performed (step S1).

  In the tracked state, the multipath detecting device 50 generates a replica code by the code generator 331 (step S2). Based on the phase offset value group (−0.1, −0.05, 0, 0.05, 0.1 chip), the register 330 generates a replica code group (replica code) from the replica code generated by the code generator 331. Codes VE, E, P, L, VL) are generated (step S3).

  The correlation value processing unit 30 performs correlation processing between the I-phase C / A code and the replica code group (replica codes VE, E, P, L, and VL), which is a correlation value group (output from the correlators 320 to 324). RVE, RE, RP, RL, RVL) are generated (step S4).

Then, using these phase offset value groups (−0.1, −0.05, 0, 0.05, 0.1 chip) and correlation value groups (RVE, RE, RP, RL, RVL), autocorrelation is performed. The function is approximated by a polynomial. In other words, (−0.1, RVE), (−0.05, RE), (0, RP), (0.05, RL), (0.1, RVL) are used to relate five terms. Numerical values a _{0 to} a ₄ can be obtained. The presence / absence of a multipath signal is detected by the multipath presence / absence detecting unit 403 of the processor 35 using the coefficient value of the approximated polynomial term (step S5).

  That is, a series of procedures from step S2 to step S5 is a method for detecting the presence or absence of a multipath signal. The tracking process (step S1) is performed by the multipath detection device 50 when the multipath detection device 50 is applied to a device having no tracking function other than the GNSS receiver 10 or the like.

(3-3) Use of Detection Result The detection result of the presence / absence of a multipath signal is provided from the multipath presence / absence detection unit 403 to the positioning calculation unit 355. For example, when the positioning calculation unit 355 receives GNSS signals from six satellites, the GNSS signal determined by the multipath presence / absence detection unit 403 to be superimposed is not used for the positioning calculation. Can be. If the multipath signal is superimposed on one of the six satellites, the positioning calculation unit 355 performs the positioning calculation using the remaining five hygiene.

<Features>
(1)
In the multipath estimation apparatus 51 of the GNSS receiver 10, the replica code generated by the code generator 331 is used to generate a tracking point prompt replica code P by the register 330 (an example of a replica code generation unit). Replica code E (an example of a first early replica code) whose phase has advanced by 0.05 chips (an example of a first phase) and a replica code L (a first phase) that has a phase delayed by 0.05 chips (an example of a first phase) An example of a late replica code), and a phase that is delayed by 0.1 chip (an example of a second phase) and a phase that is delayed by 0.1 chip (an example of a second phase) and a phase that is delayed by 0.1 chip (an example of a second phase) A replica code VL (an example of a second late replica code) is generated. Correlation value groups (RVE, RE, RP, RL, RVL) are generated by the multipliers 310 to 314 and the correlators 320 to 324 of the correlation processing unit 30.

  The above configuration is a configuration originally provided in the GNSS receiver 10 for receiving the GNSS signal, and these configurations are also used as the configuration of the multipath estimation apparatus 51. For this reason, when the multipath estimation apparatus 51 is provided in the GNSS receiver 10, it is not necessary to newly provide a shared part, so that hardware resources of the multipath estimation apparatus 51 are reduced. However, the above-described replica code generation unit and correlation processing unit 30 may be provided separately from the configuration of the GNSS receiver as components of the multipath estimation apparatus.

  Then, the estimation unit 400 of the processor 35 can determine whether or not the signal strength of the multipath signal superimposed on the GNSS signal is small enough to be used in the positioning calculation unit 355. In the estimation unit 400, since the presence or absence of a multipath signal is detected using a coefficient value of a predetermined term of the polynomial by approximating a fourth-order or higher order polynomial, software with few multipath signals that are not suitable for positioning calculation Can be detected with resources.

  As described above, the positioning calculation unit 355 performs the positioning calculation using the GNSS signal on which the multipath signal is not superimposed, thereby calculating the pseudo distance with high accuracy and performing the positioning with high accuracy.

FIG. 5 is a graph showing the relationship between multipath signal delay and pseudorange. FIG. 5 shows an error when a multipath signal having a signal strength half that of a direct wave is generated using a simulator and the pseudorange is calculated by the GNSS receiver 10. FIG. 7 is a graph showing the relationship between the delay of the multipath signal and the normalized value (a ₁ / a ₀ ) of the coefficient of the first-order term. FIG. 7 shows a first-order coefficient / constant term when a multipath signal having a signal strength half that of a direct wave is generated using a simulator and approximated to a fourth-order polynomial in the GNSS receiver 10. The coefficient (a ₁ / a ₀ ) is shown. 5 and 7, when the phase difference between the direct wave and the multipath signal is 0 degree, it is ◯, when it is 90 degrees, it is ×, when it is 180 degrees, it is ☆, and when it is 270 degrees, it is ▲. It is shown. For comparison, when the multipath signal is not superimposed (no MP), it is indicated by ◇.

From FIG. 7, the coefficient value (a ₁ / a) of the first-order term is compared with the case where the multipath delay amount exceeds about 50 m and when the phase difference is 0 degree and 180 degrees as compared with the case where there is no multipath signal. It can be seen that ₀ ) is clearly separated. In addition, referring to FIG. 7, when the multipath signal is superimposed, compared with the case where the multipath signal is not superimposed, the normalized coefficient value (a ₁ / a ₀ ) of the first-order term. It can be seen that the absolute value of has a large value. In particular, when the phase difference is 0 degree and 180 degrees, the tendency is remarkable. On the other hand, referring to FIG. 6, compared to the case where the multipath signal is not superimposed, the pseudorange error does not tend to take a large value even when the multipath signal is superimposed. In other words, if the coefficient value (a ₁ / a ₀ ) of the first-order term is monitored, it can be detected that there is a multipath signal even if the pseudorange error is not large. In this way, it is possible to reduce the pseudo-range error that is difficult to find due to the multipath signal by not using the GNSS signal on which the multipath signal is superimposed for the positioning calculation, leading to an improvement in positioning accuracy.

  In the first embodiment, when approximating a fourth or higher order polynomial using the phase offset value group and the correlation value group, the phase offset value group and the correlation value group may be used as they are. Detection may be performed after modifying the shape of the autocorrelation function so as not to break the symmetry around the tracking point of the autocorrelation function determined by the phase offset value group and the correlation value group.

  (2) The autocorrelation function indicating the correlation between the C / A code and the replica code is centered on the phase of the tracking point when the GNSS signal in an ideal state where the multipath signal is not superimposed is accurately tracked. The side where the phase is delayed and the side where the phase is advanced are symmetrical. Therefore, it is preferable that the polynomial to be approximated is an even function, and therefore the order of the polynomial is preferably an even number. In the first embodiment, the fourth-order polynomial has been described as an example of the even function. However, it may be a 6th-order or higher 2n-order polynomial (n is a positive integer).

In the case of a fourth-order polynomial, as described above, it has been experimentally confirmed that the coefficient value a ₁ of the first-order term changes significantly depending on the presence or absence of the multipath signal. For this reason, the multipath presence / absence detection unit 403 (an example of a detection unit) detects the presence / absence of a multipath signal using only the coefficient value a ₁ of the first-order term. However, even when the presence or absence of a multipath signal is detected using a fourth-order polynomial, the detection obtained by the coefficient value a _{1 of the first-} order term is corrected with the coefficient value a ₃ of the third-order term. Also good. That is, it may be determined that there is a multipath signal when the absolute value of the calculation result of the function having the coefficient values a1 and a3 as variables exceeds a preset threshold value. Similarly, when a polynomial of 6th order or higher is used for detecting the presence or absence of a multipath signal, only the coefficient value of any odd-order term may be used, depending on the coefficient value of a different odd-order term (odd-order term). The absolute value of the calculation result may be used. Even in such a case, since the coefficient value of the odd-order term is used, the absolute value of the coefficient value or the absolute value of the calculation result by the coefficient value of the different odd-order term (odd-order term) exceeds a predetermined threshold value. It is determined that there is a multipath signal. In order to obtain a coefficient value of a polynomial of 6th order or higher, 7 or more simultaneous equations are required. In such a case, for example, the number of correlators is increased to 7 or more and the phase offset is increased. Correlation values between two or more replica codes and C / A codes having different values may be used.

  (3) As described above, when an even-order polynomial is used and the absolute value of the coefficient value of the odd-order term or the operation value of the coefficient values of the odd-order term is used for detection, a predetermined value is used. It is determined that the multipath signal is superimposed on the GNSS signal when the threshold value is exceeded. However, the absolute value of the coefficient value of the even-order term or the absolute value of the calculation result based on the coefficient value of the different even-order term (even-order term) can also be used for detection. In that case, when the multipath presence / absence detection unit 403 calculates the absolute value of the coefficient value or the absolute value of the calculation result based on the coefficient value of a different even-order term (even-order term) below a predetermined threshold, It is determined that there is a pass signal.

(4) As described above, in order for the multipath presence / absence detection unit 403 to detect the presence / absence of a multipath signal, the coefficient calculation unit 2 must calculate a coefficient value necessary for detecting the presence / absence of a multipath signal. Don't be. In order to calculate coefficient values, it is necessary to have more equations than the number of unknown coefficient values. As described above, in order to solve an equation for obtaining an unknown coefficient value, a large number of operations are required. In the said 1st Embodiment, although the coefficient value was calculated | required using SGM, you may obtain | require a coefficient value by methods other than SGM. However, since the amount of calculation can be reduced and software resources can be reduced, it is preferable to use SGM as in the first embodiment when solving such an equation. For example, when approximating a quartic polynomial, the coefficient calculation unit 2 may perform the calculation of the following equation (3) using the above equation (2).
a ₃ = (2000 × (RVL−RVE) + 4000 × (RE−RL)) ÷ 3
(3)

  In the first embodiment, the phase offset value group (−0.1, −0.05, 0, 0.05, 0.1 chip) of the replica codes VE, E, P, L, and VE arranged in order of phase. The difference in phase offset value (phase difference) between the elements of (1) is equal at 0.05 chips, so SGM can be applied. However, even when the phase differences are not equal, SGM can be applied to reduce the amount of calculation and software resources. For example, when the phase offset value group is (−0.15, −0.05, 0, 0.05, 0.15 chip), the phase difference is 0.1 chip and 0.05 chip. It seems that SGM cannot be applied. In such a case, the phase offset value group is intentionally replaced with (−0.1, −0.05, 0, 0.05, 0.1 chip), and the phase difference between each element of the phase offset value group Apply SGM as if. When such processing is performed, an approximation to a modified autocorrelation function is maintained while maintaining symmetry, but since SGM can be used, there is an effect that the amount of calculation can be reduced.

<Modification>
(1)
In the first embodiment, the I-phase multipliers 310 to 314 and the correlators 320 to 324 are used to obtain the correlation value group, but the Q-phase multipliers 315 to 319 and the correlators 325 to 329 are used. May be. In either case, when normalization is performed, processing is performed so as not to divide by 0 or a numerical value very close to 0.

Second Embodiment
(1) Overview of Multipath Detection Device In the first embodiment, a case has been described in which the multipath estimation device 51 estimates the signal strength of a multipath signal. The phase may be estimated by a multipath estimation device.

  Therefore, when compared with the multipath estimation device 51 or the multipath detection device 50 of the first embodiment, the multipath estimation device 51A or the multipath detection device 50A according to the second embodiment replaces the estimation unit 400 with an estimation unit. The difference is that 400A is provided.

  The estimation unit 400A, for each GNSS signal, the phase offset value group of the replica codes VE, E, P, L, and VL obtained from the correlation processing unit 30, the replica codes VE, E, P, L, and VL, and C / A The phase of the multipath signal is detected by using a correlation value group of I-phase and Q-phase consisting of correlation values with the code.

For example, if it is approximated by a quartic function, a general expression of the autocorrelation function of the I phase and the Q phase is given by the following expression (4).
R _k (t) = a _0k + a _1k · t + a _2k · t ² + a _3k · t ³ + a _4k · t ⁴ (4)

  In the equation (4), the subscript k is i or q, and represents whether the coefficient value is the coefficient value of the I-phase autocorrelation function or the Q-phase autocorrelation function.

Since each point of the above (t, R _k (t)) satisfies this general formula (4), (−0.1, RVE _k ), (−0.05, RE _k ), (0, RP _k ). ), (0.05, RL _k ), (0.1, RVL _k ), the coefficient values a _{0k to} a _4k of the five terms can be obtained. Here, RVE _k , RE _k , RP _k , RL _k , and RVL _k are I-phase correlation values when k = i, and are obtained from the correlators 320 to 324, respectively. Further, when k = q, it is a Q-phase correlation value, which is obtained from the correlators 325 to 329, respectively.

(2) Configuration of Multipath Estimation Device As described above, since the presence / absence and phase of a multipath signal is detected using a coefficient of a predetermined term of a fourth-order or higher-order polynomial, estimation is performed as shown in FIG. The unit 400A includes an I-phase correlator output acquisition unit 411, a Q-phase correlator output acquisition unit 421, coefficient calculation units 412 and 422, and a multipath phase calculation unit 415.

The I-phase correlator output acquisition unit 411 acquires correlation values RVE _i , RE _i , RP _i , RL _i , and RVL _i output from the five correlators 320 to 324. Similarly, the Q-phase correlator output acquisition unit 421 acquires correlation values RVE _q , RE _q , RP _q , RL _q , and RVL _q output from the five correlators 325 to 329.

The coefficient calculation unit 412 includes a correlation value group including five correlation values RVE _i , RE _i , RP _i , RL _i , RVL _i given from the I-phase correlator output acquisition unit 411, and each of these correlation value groups. For example, each coefficient value a _{0i of} a fourth-order polynomial is used by using a phase offset value group consisting of phase offset values -0.1, -0.05, 0, 0.05, 0.1 chips corresponding to elements. Find ~ _a4i . Since the method of _obtaining these coefficient values a _{0i to} a _4i is the same as that in the first embodiment, the description thereof is omitted.

Similarly, the coefficient calculation unit 422 includes a correlation value group including five correlation values RVE _q , RE _q , RP _q , RL _q , and RVL _q given from the Q-phase correlator output acquisition unit 421, and these correlations. Phase offset values corresponding to each element of the value group -0.1, -0.05, 0, 0.05, and a phase offset value group consisting of 0.1 chips, The coefficient values a _{0q to} a _4q are obtained.

In the coefficient calculation units 412 and 422, each coefficient value a _{0i to} a _4i and a _{0q to} a _4q can be obtained by a simple method of product of a constant matrix and a correlation value using SGM. In the case of a fourth-order polynomial, the portion that is solved as shown in the above equation (2) and becomes a fifth-order square matrix is a coefficient matrix determined by the SGM.

The multipath phase calculation unit 3 determines the presence / absence of a multipath signal from the coefficient values a _{0i to} a _4i and the coefficient values a _{0k to} a _{4k of} the polynomial calculated by the coefficient calculation units 412 and 422, and has a multipath signal. In some cases, the phase of the multipath signal is estimated.

(3) Discrimination of presence / absence of multipath signal and calculation of phase Specifically, for example, when the autocorrelation function is approximated by a fourth-order polynomial, coefficient values a _1i and a _1q of the first-order terms are used. It is preferable to detect the presence or absence of a multipath signal. This is because, when approximated by a fourth-order polynomial, as described above, when the multipath signal is superimposed, it appears prominently as changes in the first-order coefficient values a _1i and a _1q .

The presence / absence of a multipath signal is detected using a geometric mean value GM of coefficient values a _1i and a _1q of a fourth-order polynomial given by the following equation (5).

When coefficient values a _1i and a _1q of a fourth-order polynomial are used, the phase θ is obtained by the following equation (6).

  By calculating Equation (6) in the multipath phase calculation unit 415, it is possible to estimate the phase difference between the carrier tracking phase plane and the phase of the multipath signal.

  The procedure until the detection of the presence / absence of the multipath signal and the estimation of the phase are summarized as shown in the flowchart of FIG.

  The procedure from step S1 to step S3 shown in FIG. 10 is the same as the procedure for detecting the presence or absence of the multipath signal of the first embodiment shown in FIG.

The correlation value processing unit 30 performs correlation processing between the I-phase C / A code and the replica code group (replica codes VE, E, P, L, and VL), which is a correlation value group (output from the correlators 320 to 324). RVE _i , RE _i , RP _i , RL _i , RVL _i ) are generated (step S4). Further, the correlation value processing unit 30 performs correlation processing between the Q-phase C / A code and the replica code group, so that correlation value groups (RVE _q , RE _q , RP _q , RL _q) that are outputs of the correlators 325 to 329 are obtained. , RVL _q ) is generated (step S4).

These phase offset value groups (−0.1, −0.05, 0, 0.05, 0.1 chip) and correlation value groups (RVE _i , RE _i , RP _i , RL _i , RVL _i , RVE). _q , RE _q , RP _q , RL _q , RVL _q ) are used to approximate the I-phase and Q-phase autocorrelation functions to polynomials, and the estimation unit 400A of the processor 35 determines the I-phase and Q-phase polynomials. The presence / absence of a multipath signal is detected using the first coefficient value and the second coefficient value of the term (step S5). Further, the phase is estimated using the first coefficient value and the second coefficient value of the predetermined terms of the obtained I-phase and Q-phase polynomials (step S6).

  That is, a series of procedures from step S2 to step S5 is a method for detecting the presence or absence of a multipath signal. In step S6, the phase of the multipath signal is estimated.

(4) Use of Phase Difference Estimation Results The multipath phase calculation unit 415 to the positioning calculation unit 355A detect the presence / absence of a multipath signal and, when it is determined that there is a multipath signal, the carrier tracking plane and the multipath The phase difference of the signal is given. Here, the carrier tracking plane is a complex plane that gives an initial phase in which the product of the output Q of the multiplier 302 and the prompt replica code P in FIG. 8 is 0, that is, the output of the multiplier 317 is 0.

  The positioning calculation unit 355A uses the same method for detecting the presence / absence of a multipath signal as in the positioning calculation unit 355 of the first embodiment.

  The positioning calculation unit 355A can be configured to increase the accuracy of the positioning calculation using, for example, the phase of the multipath signal. Since the multipath signal also changes, the influence of the multipath signal on the positioning calculation changes with time. Although it is impossible to estimate the effect of multipath signals that change over time only by the presence or absence of multipath signal superposition, if the phase of the multipath signal with respect to the carrier tracking plane is known, the effect of multipath signals that change over time It is possible to guess the change of In the positioning calculation unit 355A, by using the phase of the multipath signal, for example, it is possible to perform processing such as changing the weighting as time elapses, thereby enabling more accurate positioning calculation.

<Features>
In the multipath estimation apparatus 51A, as with the multipath estimation apparatus 51 of the first embodiment, the register 330 (an example of a replica code generation unit) uses a register 330 (an example of a replica code generation unit) for the replica code generated by the code generator 331 to prompt the tracking point. P, replica code E whose phase is advanced by 0.05 (an example of the first phase) with respect to this prompt replica code P, replica code L whose phase is delayed by 0.05 (first phase), and 0.1 A replica code VE whose phase is advanced by (an example of the second phase) and a replica code VL whose phase is delayed by 0.1 (second phase) are generated. By the multipliers 310 to 314 and the correlators 320 to 324 of the correlation processing unit 30, the I-phase C / A code (an example of the pseudo noise code) and the replica codes VE, E, P, L, and VL are respectively Correlation value groups (RVE _i , RE _i , RP _i , RL _i , RVL _i ) are generated. Furthermore, the multipliers 315 to 319 and the correlators 325 to 329 of the correlation processing unit 30 of the GNSS receiver 10 and Q-phase C / A codes (an example of orthogonal pseudo noise codes) and the replica codes VE, E, P, Q-phase correlation values (RVE _q , RE _q , RP _q , RL _q , RVL _q ) are generated from L and VL, respectively.

  Then, the estimation unit 400A of the processor 35 can detect whether or not the signal strength of the multipath signal superimposed on the GNSS signal is small enough to be used by the positioning calculation unit 355A. In the estimation unit 400A, the presence or absence of a multipath signal is detected by using a coefficient value of a predetermined term of the polynomial by approximating a fourth-order or higher order polynomial. Can be detected with resources.

  As described above, the positioning calculation unit 355A performs the positioning calculation using the GNSS signal on which the multipath signal is not superimposed, thereby calculating the pseudo distance with high accuracy and performing the positioning with high accuracy.

  FIG. 11 is a graph showing the relationship between the delay of the multipath signal and the phase difference θ. FIG. 11 shows a result of using a simulator to generate a multipath signal whose signal strength is half that of a direct wave and calculating a phase difference in the GNSS receiver 10. In FIG. 11, the distinction when the phase difference between the direct wave and the multipath signal is 0 degree, 90 degrees, 180 degrees and 270 degrees is the same as in FIGS.

  When FIG. 11 is compared with FIG. 7, the phase difference of 90 degrees and 270 degrees, which is difficult to detect in FIG. 7, is clearly separated from the case where there is a multipath signal and the case where there is no multipath signal. It can be seen that the presence or absence of a signal can be detected.

  FIG. 12 is a graph showing the relationship between the delay of the multipath signal and the phase difference of the multipath signal with respect to the direct wave. It can be seen that the phase difference error is relatively large until the multipath delay amount is about 20 m, but the phase difference can be detected stably when the multipath delay amount exceeds about 40 m. In the used GNSS receiver 10, such a result is obtained. However, if a more accurate GNSS receiver is used, the error of the phase difference can be reduced even when the multipath delay amount is smaller. Conceivable.

  By estimating the phase difference of the multipath signal with respect to the direct wave, it is possible to improve the multipath detection accuracy and reduce the carrier phase error. Since it requires less software resources (calculation amount), it can also be used for inexpensive consumer receivers.

<Modification>
(1)
In the first embodiment and the second embodiment, the case where the multipath estimation devices 51 and 51A are applied to the GNSS receiver 10 has been described as an example. However, the multipath estimation device of the present invention uses GNSS. Any device that calculates pseudoranges can be applied to devices other than GNSS receivers.

(2)
Examples of the GNSS receiver according to the first and second embodiments include a GPS (Global Positioning System) receiver and a Galileo (Galileo Positioning System).

(3)
In the first embodiment and the second embodiment, the register 330 is used as the replica code generation unit. However, the replica code generation unit is configured by a device other than the register, for example, using a plurality of delay lines having different delay amounts. Also good.

(4)
In the first embodiment and the second embodiment, the phase calculation unit 355 does not use the GNSS signal on which the multipath signal is superimposed in the estimation units 400 and 400A as the process for optimizing the positioning calculation. However, the process of optimizing is not limited to this. For example, the weighting may be performed so as to reduce the influence on the positioning result in the positioning calculation of the GNSS signal on which the multipath signal is superimposed. Alternatively, the signal strength of the multipath signal may be determined by receiving a plurality of threshold values, and weighting may be performed so that the higher the signal strength of the superimposed GNSS signal is, the smaller the influence on the calculation result of the positioning calculation is. .

(5)
In the GNSS receiver 10 of the first embodiment and the second embodiment, the example in which the signal processing unit 31 of the baseband unit 3 is configured by hardware has been described. However, the function of the baseband unit 3 is realized by software. Also good. In order to realize the function of the baseband unit 3 by software, the processor 35 that has read a program from the ROM 351 realizes the function of the baseband unit 3.

  In the first embodiment, the function of the estimation unit 400 is described. In the second embodiment, the function of the estimation unit 400A is realized by software. However, the estimation unit 400 and the estimation unit 400A include components such as transistors. It can also be realized by a combined integrated circuit.

(6)
The estimation unit 400A of the second embodiment estimates the phase of the multipath signal together with the estimation of the signal strength of the multipath signal, and performs estimation of the multipath signal with higher accuracy by both functions. Although described, only the function of estimating the phase of the multipath signal of the estimation unit 400A may be used. For example, the configuration may be such that the presence or absence of a multipath signal is detected by a method similar to the conventional method, and the phase of the multipath signal is estimated by the estimation unit 400A.

DESCRIPTION OF SYMBOLS 1 Antenna 2 RF part 3 Baseband part 10 GNSS receiver 30 Correlation processing part 50, 50A Multipath detection apparatus 51, 51A Multipath estimation apparatus 325, 325A Positioning calculation part 330 Register 331 Code generation part 400, 400A Multipath estimation part

JP 2009-159261 A

Claims (11)

A multipath estimation device that estimates a multipath signal of a GNSS signal modulated using a pseudo noise code,
A prompt replica code having a phase corresponding to the phase of the pseudo noise code, a first early replica code advanced by a first phase with respect to the prompt replica code, a first late replica code delayed by the first phase, and a second phase A replica code generator for generating a second early replica code that has advanced, and a second late replica code that is delayed by the second phase;
A correlation processing unit that calculates a correlation value by performing a correlation process on each replica code generated by the replica code generation unit and the pseudo noise code;
The signal strength of the multipath signal and / or the coefficient value of a predetermined term in a fourth-order or higher order polynomial approximated by using the phase amount of each replica code and each correlation value corresponding to the phase of each replica code A multipath estimation apparatus comprising: an estimation unit that performs estimation related to a phase. The correlation processing unit generates each I-phase correlation value from the correlation processing between each replica code generated by the replica code generation unit and the pseudo noise code, and is orthogonal to the pseudo noise code. Generate each correlation value of the Q phase from the correlation processing of the pseudo noise code and each replica code generated by the replica code generation unit,
The estimation unit includes a first coefficient value of a fourth or higher order polynomial approximated using each replica code generated by the replica code generation unit and each I-phase correlation value generated by the correlation processing unit, A second coefficient value of a fourth or higher order polynomial approximated using each replica code generated by the replica code generation unit and each correlation value of the Q phase generated by the correlation processing unit is obtained, and the first coefficient value And inferring the phase of the multipath signal from a comparison of the second coefficient value with
The multipath estimation apparatus according to claim 1. The estimation unit substitutes the correlation value into a variable of a predetermined first-order polynomial when the phase difference between adjacent replica codes is equal, and obtains a coefficient value of the predetermined term.
The multipath estimation apparatus according to claim 1 or 2. The multipath estimation apparatus according to any one of claims 1 to 3,
A multipath detection device comprising: a detection unit that detects the presence or absence of the multipath signal using the coefficient value. The detection unit has an absolute value of a coefficient value of a specific odd-order term in the polynomial having an order of 4th order or higher and an even number, or an absolute value of a calculation result based on coefficient values of a plurality of odd-order terms exceeds a preset threshold value. It is determined that the multipath signal is present.
The multipath detection device according to claim 4.   6. The multipath detection according to claim 5, wherein the detection unit determines that the multipath signal is present when an absolute value of a coefficient value of a third-order term in the polynomial of fourth order exceeds the threshold value. apparatus. In the detection unit, the absolute value of the coefficient value of a specific even-order term in the polynomial having an order of 4 or more and an even number, or the absolute value of the calculation result by the coefficient values of a plurality of even-order terms is smaller than a preset threshold value. Sometimes it is determined that the multipath signal is present,
The multipath detection device according to claim 4. An antenna for receiving a GNSS signal modulated with a pseudo-noise code;
A high-frequency processing unit that down-converts the GNSS signal, performs AD conversion, and outputs digital reception data including the pseudo-noise code;
A positioning unit that performs positioning calculation from a pseudo distance obtained based on the pseudo noise code;
The multipath detection device according to any one of claims 4 to 7, which detects presence or absence of the multipath signal from the digital reception data output by the high-frequency processing unit.
With
The positioning unit is a GNSS receiver that optimizes the positioning calculation in accordance with signal strength and / or phase of the multipath signal detected by the multipath detection device. A GNSS receiver according to claim 8;
An information device terminal comprising: an information device that uses positioning information obtained by positioning calculation performed by the GNSS receiver. A multipath estimation method for estimating a multipath signal of a GNSS signal modulated using a pseudo noise code,
A prompt replica code having a phase corresponding to the phase of the pseudo noise code, a first early replica code advanced by a first phase with respect to the prompt replica code, a first late replica code delayed by the first phase, and a second phase A replica code generating step for generating a second advanced early replica code and a second late replica code delayed by the second phase;
A correlation processing step of calculating each correlation value by correlating each replica code generated by the replica code generation unit and the pseudo noise code;
The signal strength of the multipath signal and / or the coefficient value of a predetermined term in a fourth-order or higher order polynomial approximated by using the phase amount of each replica code and each correlation value corresponding to the phase of each replica code A multipath estimation method comprising: an estimation step for performing estimation relating to a phase. A multipath estimation program for estimating a multipath signal of a GNSS signal modulated using a pseudo noise code,
A prompt replica code having a phase corresponding to the phase of the pseudo noise code, a first early replica code advanced by a first phase with respect to the prompt replica code, a first late replica code delayed by the first phase, and a second phase A replica code generating function for generating an advanced second early replica code, and a second late replica code delayed by the second phase;
A correlation processing function for calculating each correlation value by correlating each replica code generated by the replica code generation unit and the pseudo noise code;
The signal strength of the multipath signal and / or the coefficient value of a predetermined term in a fourth-order or higher order polynomial approximated by using the phase amount of each replica code and each correlation value corresponding to the phase of each replica code A multipath estimation program for causing a computer to realize an estimation function for estimating a phase.

Images