JP2019007844A - Timing signal generator, electronic apparatus equipped therewith, and timing signal generation method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To self-evaluate a clock error of an internal clock. SOLUTION: When a position of an antenna for receiving a signal from a satellite is set, a set position receiving unit receives a set position which is the set position. The Kalman filter unit performs Kalman filter processing based on the observed values obtained by observing the radio waves received from the satellite by the antenna. The state vector obtained by the above processing includes the clock error of the internal clock. By the above processing, the state vector of at least one of the position of the receiving point and the moving speed can be updated. The timing signal generation unit generates a timing signal based on the clock error included in the state vector estimated in the process. The clock error evaluation unit evaluates the certainty of the clock error included in the state vector based on the distance that the position of the receiving point moves from the set position in the state vector estimated by the process. [Selection diagram] Fig. 1

Description

  The present invention mainly relates to a timing signal generation apparatus that generates a timing signal using a signal from a satellite obtained via an antenna.

  Conventionally, a timing signal generation device has been used to obtain an accurate current time or to accurately synchronize with other devices. Such a timing signal generation device is provided in an electronic device, a wireless communication facility, an in-vehicle base station, or the like. In this type of timing signal generation device, a global navigation satellite system (GNSS) including an artificial satellite or the like is used to generate timing signals at accurate time intervals.

  The artificial satellite is equipped with an extremely accurate atomic clock synchronized with the GNSS reference time. The artificial satellite transmits a satellite signal on which radio wave transmission time based on an atomic clock, orbit information of the satellite, and the like are superimposed, toward the earth.

  A GNSS receiver that receives a satellite signal from an artificial satellite includes an internal clock constituted by, for example, a quartz clock. Since the position of the satellite can be acquired based on orbit information included in the satellite signal, if the exact position of the GNSS receiver is known, the propagation delay until the satellite signal from the satellite reaches the GNSS receiver. Time can be calculated.

  The GNSS receiver includes the position information / speed information of the satellite itself included in the satellite signal, the position information / speed information of the receiver itself of the GNSS receiver, the reception time of the satellite signal measured by the internal clock, the satellite Based on the radio wave transmission time included in the signal and the propagation delay time, a clock drift, which is a shift amount of the time update interval of the internal clock, and a clock bias, which is a time difference between the internal clock and the GNSS reference time, are obtained. Hereinafter, the clock drift and the clock bias may be described together as a clock error. The GNSS receiver generates a timing signal synchronized with the GNSS reference time by offsetting the timing output from the internal clock based on the above-described deviation time. The timing signal generation device generates and outputs a timing signal of a predetermined frequency with reference to the timing signal output by the GNSS receiver.

  In order to output an accurate timing signal in such a timing signal generation device, various proposals have been conventionally made. Patent documents 1 to 5 disclose this kind of technology.

JP 2014-137318 A JP 2014-126539 A Japanese Patent Laying-Open No. 2015-175812 JP-A-2015-158432 Special table 2014-507630 gazette

  However, the configurations of Patent Documents 1 to 5 have room for improvement in that the reliability of the clock error used for timing calculation has not been evaluated.

  The present invention has been made in view of the above circumstances, and an object of the present invention is to enable a timing signal generator to self-evaluate whether or not the clock error of the internal clock is reliable.

Means and effects for solving the problems

  The problems to be solved by the present invention are as described above. Next, means for solving the problems and the effects thereof will be described.

  According to the first aspect of the present invention, a timing signal generation device having the following configuration is provided. That is, the timing signal generation device includes a set position reception unit, a Kalman filter unit, a timing signal generation unit, and a clock error evaluation unit. The set position receiving unit receives a set position that is a set position (a position of a reception point) when the position of an antenna that receives a signal from a satellite is set from the inside or the outside. The Kalman filter unit performs Kalman filter processing based on an observation value obtained by observing radio waves received from the satellite by the antenna. The state vector obtained by the processing includes a clock error of the internal clock. The Kalman filter unit can update at least one of the state vectors of the position or the moving speed of the reception point by the processing. The timing signal generation unit generates a timing signal based on the clock error included in the state vector estimated by the Kalman filter unit. The clock error evaluation unit evaluates a probability of the clock error included in the state vector based on a distance that the position of the reception point has moved from the set position in the state vector estimated by the Kalman filter unit. To do.

  According to the second aspect of the present invention, the following timing signal generation method is provided. That is, when the position of the antenna that receives the signal from the satellite is set from the inside or the outside, the set position (the position of the reception point) that is the set position is received. Kalman filter processing is performed based on observation values obtained by observing radio waves received from the satellite by the antenna. The state vector obtained by the processing includes a clock error of the internal clock. Through the processing, at least one of the state vectors of the position or moving speed of the reception point is updated. A timing signal is generated based on the clock error included in the state vector estimated by the Kalman filter processing. In the state vector estimated by the Kalman filter process, the probability of the clock error included in the state vector is evaluated based on the distance that the position of the reception point has moved from the set position.

  This makes it possible to self-evaluate the reliability of the clock error of the internal clock included in the state vector obtained for generating the timing signal. Further, since the timing signal is generated using only the reliable clock error calculation result, highly accurate timing can be obtained.

1 is a block diagram showing a schematic configuration of a timing signal generation device according to a first embodiment of the present invention. The flowchart which shows the process for implement | achieving the timing signal generation method which concerns on 1st Embodiment. The flowchart which shows the process for implement | achieving the timing signal generation method which concerns on 2nd Embodiment.

<First Embodiment>
First, the timing signal generation device 20 according to the first embodiment of the present invention will be described. FIG. 1 is a block diagram illustrating a schematic configuration of a timing signal generation device 20 according to the first embodiment.

  The timing signal generator 20 shown in FIG. 1 analyzes the satellite signals from the satellites 50, 50,... Received by the GNSS antenna 1 described later, and based on the orbit information and time information included in the satellite signals, A timing signal synchronized with the GNSS reference time can be generated.

  The timing signal generation device 20 of the present embodiment is provided in a drone equipped with, for example, a small radio base station facility. This drone is used to communicate with villages and buildings where disasters such as earthquakes and tsunamis occur, the traffic network is cut off, and people can no longer enter. This drone has the same function as a normal base station facility, and is used as a relay base station when using a mobile phone.

  This drone has a receiver for receiving satellite signals, an antenna, and various sensors for obtaining azimuth information, speed information, etc., and automatically flies and stays at a place designated by the user in advance. It is possible. The receiver and the antenna constitute part of the timing signal generation device 20. At this time, the drone that remains stationary in a stagnant state can be handled in the same way as a state where an antenna is installed at a fixed point, so that it is also possible to perform fixed point positioning for satellite signal reception. . Therefore, positioning can be performed at a set position set in advance by the user, and a timing signal with an accuracy equivalent to that of a normal base station that performs positioning at a fixed point can be generated. This timing signal can be used as a reference signal necessary for the base station.

  As shown in FIG. 1, the timing signal generation device 20 includes a GNSS antenna 1. The timing signal generation device 20 includes a signal demodulator 2, a code phase acquisition unit 3, a carrier frequency acquisition unit 4, a setting position reception unit 5, a Kalman filter unit 6, a timing signal generation unit 8, a clock error evaluation unit 11, and an evaluation output. Part 12 and the like.

  Specifically, the timing signal generation device 20 is configured as a known computer and includes a CPU, a ROM, a RAM, an oscillation circuit, and the like. The ROM or the like stores a program for realizing the timing signal generation method of the present invention. Then, by the cooperation of the above hardware and software, the timing signal generation device 20 is converted into a signal demodulation unit 2, a code phase acquisition unit 3, a carrier frequency acquisition unit 4, a set position reception unit 5, a Kalman filter unit 6, and a timing signal generation. Unit 8, clock error evaluation unit 11, evaluation output unit 12, and the like.

  The GNSS antenna 1 is an antenna that receives satellite signals in a predetermined frequency band transmitted from the satellites 50, 50,... Constituting the GNSS. Here, as the satellite, for example, artificial satellites operated in GPS, Galileo, GLONASS, and the like are conceivable, but for example, a quasi-zenith satellite may be included.

  The signal demodulator 2 extracts orbit information and time information from the satellite signal received by the GNSS antenna 1. This information is acquired by demodulating a satellite signal that is code-modulated with a pseudo-random noise code (PRN code) specific to each of the satellites 50, 50,. The obtained trajectory information and time information are output to the Kalman filter unit 6. Further, the signal demodulator 2 outputs the PRN code reproduced in the process of signal demodulation to the code phase acquisition unit 3 and outputs the reproduced carrier wave to the carrier frequency acquisition unit 4.

  The code phase acquisition unit 3 generates a replica PRN code based on the timing of an internal clock (for example, a quartz clock) of the timing signal generation device 20, and the replica PRN code and the PRN code input from the signal demodulation unit 2 Thus, the code phase of the satellite signal is obtained by obtaining the time deviation (chip time). The code phase acquisition unit 3 outputs the obtained code phase to the Kalman filter unit 6.

  The carrier frequency acquisition unit 4 acquires the carrier frequency input from the signal demodulation unit 2 by measuring the frequency. The carrier frequency acquisition unit 4 outputs the obtained carrier frequency to the Kalman filter unit 6.

  The setting position receiving unit 5 receives the position when the user (user) of the timing signal generation device 20 sets the position (reception point) of the GNSS antenna 1 to be installed in the timing signal generation device 20. To do. In other words, the set position reception unit 5 receives the position of the reception point input by the user to the timing signal generation device 20. Hereinafter, this position may be referred to as a “set position”. The position of the receiving point is set by operating an input unit such as a key and a dial (not shown) provided in the timing signal generation device 20 or by an external computer communicating with the timing signal generation device 20 and instructing it. Can do. Moreover, said setting position can also be obtained by reading from a map, for example, or performing surveying.

  Thus, in the timing signal generation device 20 of the present embodiment, the intended position of the reception point (position where the reception point should be) can be given from the outside in advance. The setting position reception unit 5 outputs the received setting position to the Kalman filter unit 6, the clock error evaluation unit 11, and the evaluation output unit 12.

  However, the set position is not necessarily a position given from the outside in advance. For example, the set position may be a position generated inside the timing signal generation device 20 instead. In addition, the position received by the set position receiving unit 5 is not necessarily the position that is generated each time.

  The Kalman filter unit 6 executes the Kalman filter process on the assumption that the set position received by the set position receiving unit 5 correctly represents the position of the GNSS antenna 1, so that the clock bias of the internal clock of the timing signal generation device 20 is included. And estimate the clock drift. Hereinafter, the clock bias may be described as “bias” and the clock drift may be described as “drift”.

  In the present embodiment, the bias means a time lag between the atomic clock included in the satellite 50 and the internal clock included in the timing signal generation device 20. Drift means the rate of change of the above time deviation. It can be said that both bias and drift are a kind of error (clock error) of the internal clock.

  In the Kalman filter unit 6, not only the bias and drift but also the position and moving speed of the reception point are estimated. The Kalman filter unit 6 repeatedly performs the Kalman filter process, and in principle, the estimated values of bias and drift are updated with the values obtained each time. However, if it is determined by the clock error evaluation unit 11 that the bias and drift values newly obtained by the Kalman filter process are not certain to be certain, the estimated values of the bias and drift are not updated, or only the bias was obtained last time. The value obtained by multiplying the drift value by the processing time interval is added and updated. In this embodiment, the Kalman filter processing is performed every second, but the processing time interval is arbitrary. Details of processing performed by the Kalman filter unit 6 will be described later.

  The timing signal generation unit 8 generates a timing signal using the clock error input from the Kalman filter unit 6. More specifically, the timing signal generation unit 8 of the present embodiment is based on the clock error (bias and drift described later) obtained by the Kalman filter unit 6 with respect to the timing of the quartz clock included in the timing signal generation device 20. By performing offset processing or the like, a 1-second periodic signal (1PPS: One Pulse per Second) synchronized with the GNSS reference time is output. In the present embodiment, the 1PPS signal corresponds to the timing signal.

  The clock error evaluation unit 11 determines whether or not the bias and drift values newly obtained by the Kalman filter process are certain to be certain, and how much the position of the reception point estimated together with the bias and drift by the Kalman filter unit 6 is from the set position. Evaluation is based on whether there is a gap. Details of this evaluation will be described later. The clock error evaluation unit 11 outputs this evaluation result to the Kalman filter unit 6.

  The evaluation output unit 12 obtains an index indicating how correct the timing signal output from the timing signal generation unit 8 is considered by a method substantially similar to the clock error evaluation performed by the clock error evaluation unit 11, for example, Notify the outside by outputting to a display or communicating.

  Next, the configuration of the Kalman filter unit 6 will be described in detail.

  In the present embodiment, the Kalman filter unit 6 includes reception point positions X, Y, Z, reception point movement speeds ΔX, ΔY, ΔZ, an internal clock bias B, and an internal clock drift D. The state vector x, which is a vector defined in (1), can be output.

ここで、添え字ｋは時刻を表す。式（１）において、Ａは、１つ前の時刻の状態から現在の状態を得るための状態遷移行列である。Ｂは、システム雑音のモデルを表す行列であり、ｖは、システム雑音ベクトルである。式（２）において、ｙは、実際の観測値を示す観測値ベクトルである。Ｃは、観測値ベクトルｙと状態ベクトルｘとの間の入出力関係を示す伝達行列である。ｗは、観測雑音ベクトルである。 In this Kalman filter, the state space model related to the state vector x is expressed by the following equations (1) and (2).

Here, the subscript k represents time. In Expression (1), A is a state transition matrix for obtaining the current state from the state at the previous time. B is a matrix representing a model of system noise, and v is a system noise vector. In equation (2), y is an observed value vector indicating an actual observed value. C is a transfer matrix indicating the input / output relationship between the observed value vector y and the state vector x. w is an observation noise vector.

  The Kalman filter unit 6 includes a prediction unit 16 and a filtering unit 17.

ここで、ｘの上に付されたハット記号は、その値が真の値とは異なることを強調して示すものである。上付きの−記号は、その値が予測値であることを意味している。 The prediction unit 16 implements a prediction step in the Kalman filter process. The Kalman filter process is repeatedly performed as described above, but the prediction unit 16 uses the state vector x _k−1 output from the filtering unit 17 in the previous process to be output in the current process. Predict _k . Although the Kalman filter is well known and will not be described in detail, the current predicted value of the state vector is represented by the following equation (3).

Here, the hat symbol added on x emphasizes that the value is different from the true value. A superscript-sign means that the value is a predicted value.

ここで、Ｑは、式（１）で示したシステム雑音ベクトルｖの分散を表す行列である。 Next, the prediction unit 16 predicts the error variance P _k from the true value of the state vector x _k output in the current process, using the error variance P _k−1 obtained in the previous process. The current predicted value of the error variance is expressed by the following equation (4).

Here, Q is a matrix representing the variance of the system noise vector v shown in Expression (1).

ここで、Ｒは、式（２）で示した観測雑音ベクトルｗの分散を表す行列である。 The filtering unit 17 implements a filtering step in the Kalman filter process. First, regarding the error (estimation error) with respect to the true value of the estimated value output from the Kalman filter unit 6, the filtering unit 17 calculates the Kalman gain G _k indicating the weight optimized so as to minimize the mean square error. It calculates | requires based on the following formula | equation (5).

Here, R is a matrix representing the variance of the observed noise vector w shown in Equation (2).

Subsequently, the filtering unit 17 uses the predicted value Cx ⁻ _k of the observed value in the current process predicted from the state vector x _k output in the previous process, and the actual observed value y _k. The true value of the state vector x _k is estimated in the process. Actually, the calculation is performed based on the following equation (6).

Finally, the filtering unit 17 obtains an error variance P _k for the true value of the state vector x _k obtained by the current processing based on the following equation (7).

  Thus, one Kalman filter process is completed, 1 is added to the time k, and the process returns to the prediction step by the prediction unit 16. Thus, the Kalman filter unit 6 alternately repeats the prediction step and the filtering step, and outputs the state vector estimated as a true value each time the filtering step is performed.

  Next, observation values input to the filtering unit 17 will be described. In this embodiment, there are two types of observation values, one is a pseudorange, and the other is a range rate.

  The observation of pseudorange will be described. The Kalman filter unit 6 updates the time obtained by multiplying the code phase (chip time) acquired by the code phase acquisition unit 3 by the speed of light and the relative speed between the satellite and the reception point obtained from the frequency acquired by the carrier frequency acquisition unit 4. The observed value of the pseudo distance is calculated by adding the distance obtained by multiplying by.

  Next, prediction of the observed value of the pseudo distance will be described. The observation value of the pseudorange at the time of this processing is the distance that the satellite and the reception point moved from the previous processing to the current processing with respect to the distance between the satellite and the receiving point at the time of the previous processing. It is possible to make a prediction by performing adjustment according to the above and then adding a value according to the bias of the internal clock of the timing signal generation device 20. The distance between the satellite and the reception point at the time of the previous process can be obtained from the position of the reception point included in the state vector output by the previous process and the position of the satellite obtained from the orbit information. . The distance traveled by the satellite between the previous processing and the current processing can be obtained from orbit information, and the distance traveled by the reception point is the movement of the reception point included in the state vector output by the previous processing. Can be predicted by speed. Furthermore, the bias of the internal clock at the time of the current process can be obtained using the bias and drift included in the state vector output in the previous process. Using the above, the Kalman filter unit 6 calculates a predicted value for predicting the observed value of the pseudorange.

  The pseudo-range observation value and the predicted value thus obtained are used in the calculation of the filtering unit 17 (the above formula (6)).

  The range rate observation will be described. Since the range rate is the rate of change of the distance between the satellite and the reception point, it can be obtained by measuring the Doppler shift of the carrier wave. Therefore, the Kalman filter unit 6 calculates the observed value of the range rate by adding a known correction amount to the relative velocity between the satellite and the reception point obtained from the frequency acquired by the carrier frequency acquisition unit 4.

  Next, prediction of the observation value of the range rate will be described. The observation value of the range rate at the time of this processing is the rate of change of the distance between the satellite and the reception point obtained based on the speed of the satellite and the speed of the reception point at the time of this processing. Further, it is possible to predict by adding a value corresponding to the drift of the internal clock of the timing signal generator 20. The speed of the satellite at the time of this processing can be obtained from orbit information, and the speed of the reception point included in the state vector output in the previous processing can be used as it is as the speed of the reception point. Furthermore, as the drift of the internal clock at the time of the current process, the drift included in the state vector output in the previous process can be used as it is. Using the above, the Kalman filter unit 6 calculates a predicted value for predicting the observed value of the range rate.

  The observed value of the range rate and the predicted value thus obtained are used in the calculation of the filtering unit 17 (the above formula (6)), similarly to the observed value of the pseudorange and the predicted value thereof.

  The pseudo distance and the range rate are observed every second, for example, and the observed value is input to the filtering unit 17 together with the predicted value each time.

  Next, the clock error evaluation unit 11 will be described.

  As described above, when the drone is stationary while staying at a preset position, it can be handled in the same manner as when the GNSS antenna 1 is installed at a fixed point. Therefore, the Kalman filter unit 6 performs the Kalman filter process on the assumption that the reception point is stationary at the set position received by the set position receiving unit 5.

  The Kalman filter processing performed in the Kalman filter unit 6 includes a position X, Y, Z of the reception point, a moving speed ΔX, ΔY, ΔZ of the reception point, a bias B of the internal clock, and a drift D of the internal clock. It is designed so that the filtering unit 17 outputs the vector x.

  If the above assumption is true, the reception point positions X, Y, and Z output from the filtering unit 17 substantially maintain the above-described set positions, and the moving speeds ΔX, ΔY, and ΔZ are zero or Should be a small value. Therefore, when the position of the reception point output by the filtering unit 17 hardly changes from the set position and the moving speed is zero or a small value, the above assumption is correct, and the clock error output at the same time. It can be said that (Bias B and Drift D) are also in a good state depending on their smallness, or are more than certain.

  On the other hand, when the position of the reception point output by the filtering unit 17 has changed greatly from the set position, or when the moving speed is high, there is a high possibility that the previous assumption was not true. Alternatively, there is a high possibility that a large observation error due to multipath or the like was included. Therefore, in this case, it can be said that the accuracy of the clock error (bias B and drift D) output simultaneously is in a bad state depending on the magnitude.

  Therefore, the clock error evaluation unit 11 monitors the position and moving speed of the reception point output from the filtering unit 17.

  The clock error evaluating unit 11 causes the Kalman filter unit 6 to update the state vector when the position of the reception point does not change more than a predetermined distance from the set position and the moving speed is less than the predetermined speed. As a result, the Kalman filter unit 6 uses the updated state vector in the next Kalman filter process, and outputs the clock errors B and D included in the updated state vector to the timing signal generation unit 8.

  On the other hand, the clock error evaluation unit 11 does not allow the Kalman filter unit 6 to update the state vector when the position of the reception point changes by a predetermined distance or more from the set position or the moving speed is a predetermined speed or more. Only the bias B is updated using the drift D included in the previous state vector. As a result, the Kalman filter unit 6 uses a state vector that is not directly affected by the current update in the next Kalman filter process. The clock errors B and D output from the Kalman filter unit 6 to the timing signal generation unit 8 are adapted to state vector values that are not directly affected by the current update.

  As described above, the clock error evaluation unit 11 according to the present embodiment ensures that the clock errors B and D output from the filtering unit 17 of the Kalman filter unit 6 are not less than a predetermined value on the assumption that the reception point is stationary at the set position. It can be evaluated by using the position and the moving speed of the reception point output at the same time. Based on this evaluation result, the method of updating the clock errors B and D output from the Kalman filter unit 6 to the timing signal generation unit 8 is switched, thereby improving the accuracy of the timing signal output from the timing signal generation unit 8 as a whole. be able to.

  Further, in the present embodiment, the evaluation output unit 12 determines the deviation of the position of the reception point from the set position with respect to the state vector based on the clock errors B and D output from the Kalman filter unit 6 to the timing signal generation unit 8. Based on the magnitude and the magnitude of the moving speed of the reception point, the degree of correctness of the timing signal output from the timing signal generator 8 can be evaluated, and the result can be output to the outside. There are various methods for expressing the evaluation. For example, it is conceivable to display in a plurality of stages such as high, medium, and low. In the present embodiment, since the timing signal generation device 20 is provided in the drone, the evaluation result of the correctness of the timing signal is wirelessly transmitted from the drone to the operation device or the like operated by the user and displayed on the operation device. It is preferable. Thereby, the user can grasp the accuracy of the currently calculated timing signal, and can move the drone or interrupt or stop the timing calculation as necessary.

  Next, processing performed by the Kalman filter unit 6 and the like will be described with reference to FIG. FIG. 2 is a flowchart showing processing for realizing the timing signal generation method in the first embodiment.

  The process is started in a state where the drone is flying at a predetermined position while stopping at a predetermined position. First, the state vector is initialized in the Kalman filter unit 6 (step S101). Of the initial values of the state vectors, the received position positions X, Y, and Z use the set positions received by the set position receiving unit 5 as they are, and the moving speeds ΔX, ΔY, and ΔZ of the received points are set to zero. .

  Next, the prediction unit 16 executes a prediction step of the Kalman filter process using the state vector at the previous time (step S102).

  Subsequently, the filtering unit 17 inputs the observation values of the pseudorange and the range rate, and executes the filtering step of the Kalman filter process (step S103). Thereby, a new state vector is obtained.

  Next, the clock error evaluation unit 11 receives the reception points based on the reception point positions X, Y, and Z included in the new state vector obtained in step S103 and the movement speeds ΔX, ΔY, and ΔZ of the reception points. It is determined whether or not the point has moved from the set position (step S104).

  If the movement of the position of the reception point is small, it is considered that the estimation of the current state vector (including estimation of clock errors B and D) is likely. Therefore, in this case, the Kalman filter unit 6 updates the state vector with the new state vector obtained in step S103 (step S105). On the other hand, if the position of the reception point has moved greatly, the process of step S105 is not performed (that is, the state vector obtained in step S103 is not used), and the bias is applied using the drift D in the previous state vector. Only B is updated (step S106).

  Regardless of how the state vector is updated, the Kalman filter unit 6 outputs the clock errors B and D included in the state vector to the timing signal generation unit 8 (step S107).

  Further, the evaluation output unit 12 determines the clock error B, output in step S107, based on the position X, Y, Z of the reception point included in the state vector and the moving speed ΔX, ΔY, ΔZ of the reception point. The probability of D (in other words, the correctness of the timing signal generated by the timing signal generator 8) is evaluated and output (step S108). Thereafter, the process returns to step S102.

  Next, a case where the GNSS antenna 1 receives a plurality of satellite signals will be described.

  Up to now, for convenience of explanation, the case where the signal from one satellite 50 is received by the GNSS antenna 1 has been described, but actually, the signal from the plurality of satellites 50, 50,. It is common to receive. In this case, the Kalman filter unit 6 prepares a state vector for each satellite 50 and performs Kalman filter processing in parallel. As a result, as many state vectors as the number of satellites that have received satellite signals can be obtained.

  However, in this case, in step S107 in FIG. 2, it becomes a problem which state vector clock error B, D is selected from among a plurality and is output to the timing signal generation unit 8.

  As a criterion for this selection, it is conceivable to use the residual between the observed value of the pseudorange and the predicted value. In brief, the observation of the pseudorange and the predicted value that predicted the observed value have been described earlier, but the residual between the observed value of the pseudorange and the predicted value (hereinafter referred to as the observation prediction residual). (Sometimes called a difference)) selects the state vector that is closest to zero.

  This selection criterion is effective to some extent, but if the accuracy of the clock error of the internal clock is not good, the observation error caused by some reason (for example, multipath) will accidentally cancel the error due to the time shift of the internal clock. As a result, there is a possibility that a satellite whose pseudo-range observation prediction residual is small may be selected, leading to a decrease in timing signal accuracy.

  In this regard, the Kalman filter unit 6 of the present embodiment uses the clock errors B and D as timing signals from a composite viewpoint that takes into account the evaluation results of the clock error evaluation unit 11 in addition to the above-described pseudo-distance observation prediction residuals. A state vector to be output to the generation unit 8 is selected. There are various specific selection methods. For example, the pseudorange residual score is set so as to decrease as the observation prediction residual of the pseudorange decreases, and the evaluation result by the clock error evaluation unit 11 becomes smaller as better. It is conceivable that a clock error evaluation score is determined so that a state vector whose sum of both scores is closest to zero is selected. Thereby, since a more suitable satellite can be substantially selected, the accuracy of the timing signal can be improved.

  As described above, the timing signal generation device 20 of the present embodiment includes the setting position reception unit 5, the Kalman filter unit 6, the timing signal generation unit 8, and the clock error evaluation unit 11. When the position of the GNSS antenna 1 that receives a signal from the satellite 50 (the position of the reception point) is set from the inside or the outside, the set position receiving unit 5 receives the set position that is the set position. The Kalman filter unit 6 performs Kalman filter processing based on observation values obtained by observing radio waves received from the satellite 50 by the GNSS antenna 1. The state vector obtained by the above processing includes the clock errors B and D of the internal clock, the positions X, Y and Z of the reception points, and the moving speeds ΔX, ΔY and ΔZ. The timing signal generator 8 generates a timing signal based on the clock errors B and D included in the state vector estimated by the Kalman filter unit 6. Based on the distance (how far the position of the reception point has moved from the set position) in the state vector estimated by the Kalman filter unit 6, the clock error evaluation unit 11 includes the clock error B included in the state vector. , D is evaluated for certainty.

  Moreover, in the timing signal generation device 20 of the present embodiment, the timing signal is generated as follows. When the position of the GNSS antenna 1 that receives a signal from the satellite (the position of the reception point) is set from the inside or the outside, the set position that is the set position is received. Kalman filter processing is performed based on observation values obtained by observing radio waves received from the satellite 50 by the GNSS antenna 1. The state vector obtained by the above processing includes the clock errors B and D of the internal clock, the positions X, Y and Z of the reception points, and the moving speeds ΔX, ΔY and ΔZ. A timing signal is generated based on the clock errors B and D included in the state vector estimated by the processing. In addition, in the state vector estimated by the Kalman filter processing, the probability of the clock errors B and D included in the state vector is determined based on the distance (how far the position is moved) from the set position. evaluate.

  As a result, it becomes possible to self-evaluate the reliability of the clock errors B and D of the internal clock included in the state vector obtained for generating the timing signal. As a result, the accuracy of the timing signal can be improved.

  Further, in the timing signal generation device 20 of the present embodiment, when the clock error evaluation unit 11 determines that the clock errors B and D included in the state vector estimated by the Kalman filter unit 6 are more than predetermined, the Kalman filter unit 6 The clock errors B and D output to the timing signal generator are updated.

  As a result, the timing signal generator 8 generates the timing signal based on the probable calculation results of the clock errors B and D, so that highly accurate timing can be obtained continuously.

  In addition, the timing signal generation device 20 of this embodiment includes an evaluation output unit 12. The evaluation output unit 12 is a timing signal generation unit based on the distance that the position of the reception point has moved from the set position in the state vector that is the basis of the clock errors B and D output from the Kalman filter unit 6 to the timing signal generation unit 8. 8 evaluates the correctness of the timing signal output and outputs the result.

  As a result, the user can easily know the accuracy of the timing signal.

  In the timing signal generation device 20 of the present embodiment, the Kalman filter unit 6 performs Kalman filter processing on each of the signals received by the GNSS antenna 1 from the plurality of satellites 50, 50,. The clock error evaluation unit 11 evaluates the likelihood of the clock errors B and D for the state vector estimated by the Kalman filter unit 6 with respect to the signals from the respective satellites. The Kalman filter unit 6 uses the evaluation result of the clock error evaluation unit 11 to select which satellite from which the clock errors B and D obtained for the signal are output to the timing signal generation unit 8.

  Thereby, when signals from a plurality of satellites are received, a timing signal can be generated based on the clock errors B and D based on signals from more appropriate satellites.

  In the present embodiment, the drone includes the timing signal generation device 20.

  Thereby, in a drone as an electronic device, a highly accurate timing signal generated in consideration of the clock errors B and D can be obtained, and the drone can be operated at an accurate timing.

Second Embodiment
Next, a timing signal generation device according to a second embodiment of the present invention will be described. FIG. 3 is a flowchart showing a process for realizing the timing signal generation method according to the second embodiment. In the description of the present embodiment, the same or similar members as those of the above-described embodiment may be denoted by the same reference numerals in the drawings, and description thereof may be omitted.

  In the Kalman filter unit 6 of the second embodiment, the processing shown in the flowchart of FIG. 3 is performed. This process is the same as the process of the first embodiment shown in FIG. 2 except for step S204 and step S205.

  Since the processing from step S201 to step S203 corresponds to the processing from step S101 to step S103 of the first embodiment, description thereof will be omitted.

  In step S204, the clock error evaluation unit 11 determines the reception point positions X, Y, and Z included in the new state vector obtained in step S203 and the movement speeds ΔX, ΔY, and ΔZ of the reception points. The probability of the clock errors B and D is quantitatively evaluated, and the weight for updating the state vector is calculated according to the evaluation result. This weight is set to a value between 0 and 1 and is calculated so as to decrease as the reception point moves greatly from the above-mentioned set position in the new update vector (in other words, the clock errors B and D are suspicious). The

  In step S205, the Kalman filter unit 6 obtains a difference between the new state vector obtained in step S203 and the previous state vector, and multiplies this difference by the weight obtained in step S204 to obtain the previous state vector. Is added to update the state vector. As a result, when the reception point has not moved much, the influence of the new state vector obtained in step S203 is strengthened, and when the reception point has moved greatly, the influence of the new state vector is weakened. In addition, the state vector is updated. Since the state vector includes clock errors B and D as described above, updating the state vector means updating the clock errors B and D.

  Since the processing of step S206 and step S207 corresponds to the processing of step S107 and step S108 of the first embodiment, description thereof will be omitted.

  Even with such a configuration, it is possible to obtain a highly accurate timing signal by outputting clock errors B and D that are entirely reliable to the timing signal generation unit 8.

  As described above, in the timing signal generation device according to the present embodiment, weighting is performed according to the probability of the clock error evaluated by the clock error evaluation unit 11, and this weight is used by the Kalman filter unit 6 as a timing signal. Used to update clock errors B and D output to the generator.

  Thereby, when the obtained clock errors B and D are likely, the influence at the time of updating is strengthened, and when not so, the influence at the time of updating is weakened, so that highly accurate timing can be flexibly obtained.

  The preferred embodiments and modifications of the present invention have been described above, but the above configuration can be modified as follows, for example.

  In the above embodiment, the pseudorange and range rate are observed, but either one may be omitted. Further, instead of obtaining the observed value of the range rate by Doppler shift, the observed value of the range rate may be obtained by changing the pseudo distance.

  The expression of the state vector in the Kalman filter process is not limited to the above. For example, either one of the position of the reception point and the moving speed may be omitted. Also, the drift of the internal clock may be omitted from the expression of the state vector.

  Orbit information used in Kalman filter processing or the like may be acquired from another information source instead of directly acquiring from satellite signals. For example, the trajectory information stored in the non-volatile storage unit may be used for hot start to make positioning possible in a short time immediately after the power is turned on. For example, the timing signal generation device may be configured to be connectable to the Internet, and the trajectory information may be acquired based on so-called GNSS assist information acquired from the Internet.

  Instead of the 1PPS signal, a signal having a shorter or longer interval may be used as the timing signal. The timing signal may be an arbitrary form of pulse.

  The timing signal generation device of the present invention is not limited to a drone for disaster, but can be used for various electronic devices that need to operate at an accurate timing. As the electronic device, for example, a wireless communication device can be considered. Further, if the apparatus is used in a stationary state, the timing signal generating device may be provided in a movable electronic device (for example, an in-vehicle electronic device).

  In the above embodiment, the GNSS antenna 1 is detachably attached to the timing signal generation device 20, but instead of this, an external antenna electrically connected to the timing signal generation device 20 may be used. it can.

1 GNSS antenna (antenna)
5 Setting Position Receiving Unit 6 Kalman Filtering Unit 9 Clock Error Evaluation Unit 12 Evaluation Output Unit 20 Timing Signal Generation Device 50 Satellite

Claims (7)

A setting position receiving unit that receives a setting position that is the set position when the position of an antenna that receives a signal from a satellite is set from the inside or the outside;
A Kalman filter process is performed based on an observation value obtained by observing the radio wave received from the satellite by the antenna, the state vector obtained by the process includes a clock error of an internal clock, and the reception point is obtained by the process. A Kalman filter unit capable of updating at least one of the state vectors or the moving speeds of
A timing signal generating unit that generates a timing signal based on the clock error included in the state vector estimated by the Kalman filter unit;
In the state vector estimated by the Kalman filter unit, a clock error evaluation unit that evaluates the probability of the clock error included in the state vector based on the distance that the position of the reception point has moved from the set position;
A timing signal generating device comprising: The timing signal generation device according to claim 1,
When the clock error evaluation unit determines that the clock error included in the state vector updated by the Kalman filter unit is smaller than a predetermined value, the Kalman filter unit updates the clock error output to the timing signal generation unit. A featured timing signal generator. The timing signal generating device according to claim 1 or 2,
Weighting is performed according to the accuracy of the clock error evaluated by the clock error evaluation unit,
The timing signal generating apparatus, wherein the weight is used for updating the clock error output from the Kalman filter unit to the timing signal generating unit. The timing signal generation device according to any one of claims 1 to 3,
The timing signal output from the timing signal generator based on the distance that the position of the reception point has moved from the set position in the state vector that is the basis of the clock error output from the Kalman filter unit to the timing signal generator A timing signal generation device comprising an evaluation output unit that evaluates the correctness of the output and outputs the result. The timing signal generation device according to any one of claims 1 to 4,
The Kalman filter unit performs the Kalman filter process for each of the signals received by the antenna from a plurality of the satellites,
The clock error evaluation unit evaluates the probability of the clock error for the state vector estimated by the Kalman filter unit with respect to the signal from each satellite,
The timing signal generation device, wherein the Kalman filter unit selects, based on the evaluation result of the clock error evaluation unit, which signal to output the clock error obtained for the signal to the timing signal generation unit.   An electronic apparatus comprising the timing signal generation device according to any one of claims 1 to 5. When the position of the antenna that receives the signal from the satellite is set from the inside or the outside, the set position that is the set position is received,
A Kalman filter process is performed based on an observation value obtained by observing a radio wave received from the satellite by the antenna, and a state vector obtained by the process includes a clock error of an internal clock. Update the state vector of at least one of the moving speeds,
Generating a timing signal based on the clock error included in the state vector estimated by the Kalman filtering;
In the state vector estimated by the Kalman filter processing, the probability that the probability of the clock error included in the state vector is evaluated based on the distance the position of the reception point has moved from the set position Signal generation method.

Images