JP7044459B2 - Observation data analysis device, observation data analysis system, and observation data analysis method - Google Patents

Description

The present invention is an observation data analysis device that analyzes physical changes and physical quantities for deriving them from observation data obtained by observing physical changes such as landslides, and observation data analysis including this observation data analysis device. Regarding the system.

Currently, various systems for observing landslides have been devised from the viewpoint of disaster prevention. For example, the observation data recovery system described in Patent Document 1 includes a plurality of observation devices, a plurality of relay devices, an aggregation device, and a monitoring device.

The plurality of observation devices are linked to one of the plurality of relay devices. The observation devices to which each relay device is linked are different. Multiple relay devices are linked to the aggregate device.

The relay device acquires observation data of a plurality of linked observation devices. The relay device collects the acquired multiple observation data and sends them to the aggregation device.

The monitoring device is connected to the aggregation device via the Internet or the like. The monitoring device calculates the change in the baseline vector using a plurality of observation data aggregated in the aggregation device, and detects the displacement of the ground.

FIG. 11 is a diagram showing an observation interval of conventional observation data and a calculation interval of a baseline vector. As shown in FIG. 11, in the conventional observation system, the observation device outputs observation data such as the amount of change in the carrier phase integrated value with the observation time interval TI set to 30 seconds [sec]. Specifically, in the example of FIG. 11, the observation device outputs the observation data DP 1000, outputs the observation data DP 1001 after _a time TI (= 30 seconds [sec]), and then outputs the observation data DP ₁₀₀₁ . The observation device outputs the observation data in the order of DP 1001, DP 1002, DP ₁₀₀₃ , ... _{At intervals} of time TI (= 30 seconds [sec]). These observation data are transmitted to the relay device sequentially and temporarily stored in the relay device.

The relay device converts the observation data of the number of data used for calculating the baseline vector into one analysis data file and transmits it to the aggregation device. For example, in the case of FIG. 11, the relay device sets the transmission interval TT to 1 hour [hr] and transmits an analysis data file composed of a plurality of observation data stored in this 1 hour [hr] to the aggregation device. .. Specifically, in the case of FIG. 11, the relay device transmits an analysis data file consisting of observation data DP 1119 from observation data DP 1000 for _a transmission interval TT (= ₁ hour [hr]). After that, the analysis data file consisting of the observation data DP ₂ 119 is transmitted from the next observation data DP ₂₀₀₀ with a transmission interval TT (= 1 hour [hr]), and then the same transmission processing is continued. do.

The analysis device calculates a baseline vector using a plurality of observation data included in the analysis data file stored in the aggregation device.

Japanese Unexamined Patent Publication No. 2004-185459

However, in the conventional observation system, the baseline vector can be calculated only at the time interval for acquiring the number of observation data for which the baseline vector can be calculated, that is, the time interval for acquiring the analysis data file. In static positioning, as described above, the baseline vector is generally calculated from the observation data for 1 hour [hr], and in the conventional observation system, the baseline vector can be calculated only every 1 hour [hr]. could not. Therefore, it could not be used to detect the time change of displacement in more detail, such as in an emergency.

The observation data has an error due to the influence of the reception status of the positioning signal. 12 (A) and 12 (B) are diagrams showing changes in the baseline length due to changes in the reception status of the positioning signal. FIG. 12A shows the pattern of daily fluctuations in an easy-to-understand manner, and FIG. 12B shows the patterns of fluctuations in hours and minutes during the day in an easy-to-understand manner.

As shown in FIG. 12 (A), the baseline length fluctuates in a cycle of approximately one day. This results from an error due to the ionosphere and an error due to the troposphere. As shown in FIG. 12B, the baseline length further fluctuates in hours and minutes even during the day. This is caused by an error due to multipath caused by the installation environment of the observation device.

Here, when the baseline vector is calculated at 1-hour [hr] intervals as in the conventional case, the baseline vector is obtained as shown by the ● mark in FIG. 12 (B). In this case, it is possible to estimate the daily fluctuation as shown by the dotted line in FIG. 12 (B). However, the number of calculated baseline vectors is small for fluctuations in time units as shown in the cycles TP1, TP2, and TP3 of FIG. 12B, and the calculation intervals of the baseline vectors are sufficient for the cycle of fluctuations. It is not short and the fluctuation cannot be estimated accurately.

Therefore, the error due to this fluctuation cannot be suppressed, and the calculation accuracy of the baseline vector has a limit corresponding to this error.

Further, in the conventional configuration, there is no configuration in which the error of the baseline vector is corrected by paying attention to such an error, and it may not be possible to calculate the baseline vector with high accuracy.

Therefore, an object of the present invention is to calculate a highly accurate baseline vector.

The observation data analysis device of the present invention includes a baseline vector calculation unit, a periodic noise extraction filter, and a subtractor. The baseline vector calculation unit calculates the baseline vector using the observation data including the carrier phase. The periodic noise extraction filter extracts the periodic noise contained in the baseline vector. The subtractor subtracts periodic noise from the baseline vector.

In this configuration, the periodic noise included in the baseline vector is extracted, and this periodic noise is subtracted from the baseline vector, so that the calculation accuracy of the baseline vector is improved.

Further, the observation data analysis device of the present invention includes a high-frequency noise reduction filter that removes high-frequency noise from the baseline vector from which periodic noise has been subtracted.

In this configuration, higher frequency noise is further removed from the baseline vector from which the periodic noise has been removed. As a result, the baseline vector is calculated with higher accuracy.

Further, the observation data analysis device of the present invention further includes a data acquisition unit, a data extraction unit, and a data combination unit. The data acquisition unit acquires an analysis data file including time-series observation data from the outside. The data extraction unit extracts a plurality of observation data included in the analysis data file. The data combination unit combines a number of time-continuous observation data required for calculating the baseline vector from a plurality of extracted observation data. At this time, the data combination unit includes a predetermined number of observation data on the side with the latest time in the plurality of observation data in the time series constituting the first combination, and a plurality of observation data in the time series constituting the second combination. A plurality of observation data are combined so that the predetermined number of observation data on the earliest side in the above is common observation data. The baseline vector calculation unit uses a plurality of observation data combined by the data combination unit.

In this configuration, the calculation interval of the baseline vector is shorter than the time required to acquire the number of observation data required for the calculation of the baseline vector. This improves real-time performance and makes it possible to extract and suppress higher frequency noise.

According to the present invention, the baseline vector can be calculated with high accuracy.

Functional block diagram of the observation data analysis device according to the embodiment of the present invention Functional block diagram of the observation data analysis system according to the embodiment of the present invention The figure which shows the combination method of the observation data of the observation data analysis apparatus which concerns on embodiment of this invention, and the calculation interval of a baseline vector. The figure which shows the calculation interval of the baseline vector in the observation data analysis apparatus which concerns on embodiment of this invention. Configuration diagram of the first aspect of the analysis unit of the observation data analysis device according to the embodiment of the present invention. The figure which shows the concept of the filter processing in the observation data analysis apparatus which concerns on embodiment of this invention. A flowchart showing a method of analyzing observation data according to an embodiment of the present invention. Configuration diagram of the second aspect of the analysis unit of the observation data analysis device according to the embodiment of the present invention. Configuration diagram of the third aspect of the analysis unit of the observation data analysis device according to the embodiment of the present invention. Configuration diagram of the fourth aspect of the analysis unit of the observation data analysis device according to the embodiment of the present invention. A diagram showing the observation interval of conventional observation data and the calculation interval of the baseline vector. Diagram showing changes in baseline length due to changes in positioning signal reception status

The observation data analysis device, the observation data analysis system, and the observation data analysis method according to the embodiment of the present invention will be described with reference to the drawings. The observation data analysis system shown in this embodiment is used, for example, in a landslide detection system or the like. However, the configuration and method of the present embodiment can be applied as long as the system analyzes a predetermined phenomenon using the observation data acquired at a preset time interval.

FIG. 1 is a functional block diagram of an observation data analysis device according to an embodiment of the present invention. FIG. 2 is a functional block diagram of the observation data analysis system according to the embodiment of the present invention.

First, an observation data analysis system including an analysis device 50 corresponding to the "observation data analysis device" of the present invention shown in FIG. 1 will be described with reference to FIG.

As shown in FIG. 2, the observation data analysis system 10 includes a plurality of observation devices 20, a line aggregation device 30, a server 40, and an analysis device 50.

The line aggregation device 30, the server 40, and the analysis device 50 are connected by a network 100. The line aggregation device 30 and the plurality of observation devices 20 are connected by wireless communication. The communication between the line aggregation device 30 and the plurality of observation devices 20 may be wired. Further, the line aggregation device 30 may have a function of acquiring observation data in the observation device 20 described later.

The plurality of observation devices 20 are arranged at individual observation positions. The observation device 20 is preferably arranged in an open sky environment as much as possible. The observation device 20 includes a GNSS receiver 21, a communication control unit 22, a GNSS antenna 23, and a communication antenna 24.

The GNSS receiver 21 is connected to the GNSS antenna 23 and the communication control unit 22. The GNSS receiver 21 captures and tracks the GNSS signal received by the GNSS antenna 23, and calculates the carrier phase. The GNSS receiver 21 outputs the carrier wave phase and its calculated time as observation data to the communication control unit 22.

GNSS is an abbreviation for Global Navigation Satellite System, and includes GPS (Global Positioning System), GLONASS, Galileo and the like. The positioning signal from the QZSS (quasi-zenith satellite system) satellite is also included in the GNSS signal of the present invention.

The communication control unit 22 generates an observation file including the carrier wave phase and the calculated time. The communication control unit 22 transmits the observation file via the communication antenna 24. Each time the communication control unit 22 generates an observation file, the generated observation file is sequentially transmitted.

The line aggregation device 30 includes a communication antenna 31, a line aggregation unit 32, a router 33, and a memory 34.

The line aggregation unit 32 receives the observation file from each observation device 20 and stores it in the memory 34.

The line aggregation unit 32 generates an analysis data file from a plurality of observation files stored in the memory 34. The number of observation files included in the analysis data file, that is, the number of observation data, is the minimum number necessary for calculating the baseline vector. The generation interval of the analysis data file is set in advance, and is, for example, the time required to acquire a plurality of observation files (observation data) constituting the analysis data file. The line aggregation unit 32 outputs the generated analysis data file to the router 33.

The router 33 connects the line aggregation unit 32 to the network 100. The router 33 transmits the analysis data file to the server 40.

The memory 34 may be omitted. At this time, the line aggregation unit 32 generates an analysis data file from a plurality of observation files temporarily stored by itself. Further, a data file for analysis is generated by converting the observation file from the observation device 20 into a file.

The server 40 stores the analysis data file from the line aggregation device 30. The server 40 is, for example, an FTP server. The type of server 40 is not limited to the FTP server, and may have at least a function of storing analysis data files.

The analysis device 50 has the configuration shown in FIG. Specifically, as shown in FIG. 1, the analysis device 50 includes a data acquisition unit 51, a data extraction unit 52, a data combination unit 53, and an analysis unit 54.

The data acquisition unit 51 has a function of connecting to the network 100. The data acquisition unit 51 requests the server 40 to transfer the analysis data file, and acquires the analysis data file from the server 40. If the analysis data file is stored in the predetermined directory of the server 40, the analysis data file may be automatically transferred to the data acquisition unit 51. The data acquisition unit 51 outputs the acquired analysis data file to the data extraction unit 52.

The data extraction unit 52 extracts a plurality of observation data from the analysis data file. At this time, the data extraction unit 52 extracts the observation time (calculation time of the carrier wave phase) of each observation data in association with the observation data. The data extraction unit 52 outputs the extracted observation data to the data combination unit 53.

The data combination unit 53 groups a plurality of time-series observation data for each number required for calculating the baseline vector (for example, 1 hour [hr] minutes), and outputs the grouped data to the analysis unit 54. do. At this time, adjacent groups on the time axis include common observation data. That is, the data combination unit 53 sets the first group and the second group that are continuous in time, and the predetermined number of observation data on the latest time side in the first group and the most in the second group. Combine the observation data so that it is common with the predetermined number of observation data on the early time side. In other words, a first group and a second group that are continuous in time are set, and the data combination unit 53 observes the earliest time observation data in the first group and the earliest time observation in the second group. Combine the observation data so that the time difference from the data is shorter than the time required to acquire the number of observation data used to calculate the baseline vector.

The analysis unit 54 calculates a baseline vector using a plurality of observation data constituting one group. At this time, the baseline vector is calculated for each observation device 20 by using the phase difference between the carrier wave phase of the reference station and the carrier wave phase which is the observation data of each observation device 20. The position of the reference station may be known at the start of observation. For example, the electronic reference point of the Geographical Survey Institute may be used. In addition, at the start of observation or before the start of observation, single positioning is performed a plurality of times by a specific line aggregation device 30 or observation device 20, and the average value of the position coordinates obtained by the plurality of single positioning is used as the position of the reference station. May be good. At this time, as the specific line aggregation device 30 or the observation device 20, one that does not change the position coordinates, for example, one that is installed at a position where the ground does not change is used.

FIG. 3 is a diagram showing a method of combining observation data of the observation data analysis device according to the embodiment of the present invention and a calculation interval of a baseline vector. Note that FIG. 3 shows the observation data of one observation device. However, this process is performed in the same way with other observation devices.

First, the data combination unit 53 outputs the observation data DP 1119 from the observation data DP 1000 constituting the _first analysis data file as a group GR1 composed of observation data for _one hour [hr]. The analysis unit 54 calculates the baseline vector BL001 using the observation data group of the group GR1.

Next, the data combination unit 53 receives observation data from the observation data DP ₁ 119 that constitutes the _first analysis data file and the observation data observation data DP 2000 that constitutes the _next analysis data file. _DP2001 is output as group GR2 consisting of observation data for 1 hour [hr]. The analysis unit 54 calculates the baseline vector BL002 using the observation data group of the group GR2.

After that, the data combination unit 53 shifts the observation data of the earliest time constituting the group by two (1 minute [min] in the example of FIG. 3), and observes each for 1 hour [hr]. A group GRn (n is an integer of 3 or more) composed of data is configured and output. Along with this, the analysis unit 54 calculates the baseline vector BLn from the observation data constituting the group GRn for each group GRn.

With such a configuration, the group GR can be configured and the baseline vector BL can be calculated every time (1 minute [min]) for observing the two observation data. Therefore, the baseline vector BL can be calculated by constructing 60 group GRs in the time required to acquire the number of observation data required for calculating the baseline vector.

As a result, the baseline vector can be calculated more frequently than the conventional baseline vector calculation method. As a result, the time change of the displacement based on the baseline vector can be detected at a shorter time interval. Therefore, the number of baseline vector calculations per unit time can be improved, and the calculation accuracy of the baseline vector can be improved by the calculation of averaging.

FIG. 4 is a diagram showing the calculation interval of the baseline vector in the observation data analysis device according to the embodiment of the present invention. In FIG. 4, the horizontal axis shows time and the vertical axis shows the baseline length of the baseline vector. Further, in FIG. 4, the baseline vectors calculated by the analysis device 50 are plotted every 10th, and in reality, the baseline vectors are obtained at shorter (1/10) time intervals.

By using the analyzer 50 of the present embodiment, the baseline vector (baseline length) can be obtained at 1-minute intervals. Therefore, as shown in FIG. 4, periodic noise of periodic TP1 = about 2 hours [hr], periodic TP2. A baseline vector (baseline length) is obtained at sufficiently short intervals with respect to periodic noise of about 2.6 hours [hr] and periodic noise of periodic TP3 = about 2 hours [hr]. As a result, the time function of the baseline vector (baseline length) becomes a function including each periodic noise.

It should be noted that the daily periodic noise shown in the conventional problem has a longer period than the periodic noises of these periods TP1, TP2, and TP3. Therefore, the periodic noise in units of one day is included in the time function of this baseline vector (baseline length) in the same manner as the periodic noise of these periodic TP1, TP2, TP3.

This makes it possible to suppress periodic noise from the calculated baseline vector by the method shown below.

FIG. 5 is a configuration diagram of a first aspect of the analysis unit of the observation data analysis device according to the embodiment of the present invention. 6 (A), 6 (B), and 6 (C) are diagrams showing the concept of filtering in the observation data analysis device according to the embodiment of the present invention. In FIGS. 6 (A), 6 (B), and 6 (C), the horizontal axis is the frequency [Cycle / day], and the vertical axis is the power of each frequency component constituting the baseline vector.

The analysis unit 54 includes a baseline vector calculation unit 540, a filter 551, a delay circuit (Delay) 552, a subtractor 553, and a filter 554. The filter 551 is the "periodic noise extraction filter" of the present invention.

The baseline vector calculation unit 540 calculates the baseline vector y (t) from the observation data constituting the group GR. The baseline vector calculation unit 540 outputs the baseline vector y (t) to the filter 551 and the subtractor 553.

The filter 551 extracts periodic noise included in the baseline vector y (t). Specifically, the filter 551 is a high-pass filter and has a cutoff frequency Th (HPF) (see FIG. 6A). The filter 551 attenuates a component having a frequency lower than the cutoff frequency Th (HPF) contained in the baseline vector y (t), and passes a component having a frequency higher than the cutoff frequency Th (HPF). This filtered component is the high frequency component yH (t) of the baseline vector y (t). The delay unit 552 delays the high frequency component yH (t) of the baseline vector y (t) by a predetermined time. The delay unit 552 outputs the delayed-processed baseline vector yHD (t) to the subtractor 553. The delay time is set in the subtractor 553 so that the time difference between the high frequency component yH (t) of the baseline vector y (t) and the baseline vector y (t) is one sidereal day.

The subtractor 553 subtracts the baseline vector y (t) and the delayed-processed baseline vector yD (t) (y (t) -yHD (t)) to obtain the first correction baseline vector yc1 (t). , Output to LPF554.

If there is no displacement of the observation device 20, the baseline vector is composed of only frequency components of approximately 0 [Cycle / day] and contains almost no other frequency components. Moreover, the frequency of the baseline vector is extremely low even if it is displaced due to gradual ground movement or the like. That is, the baseline vector when displacement occurs contains only extremely low frequency components.

On the other hand, the periodic noise has the periodicity of one hour unit and several hour unit shown in the above-mentioned prior art and the problem, and is a high frequency as compared with the main frequency component of the baseline vector when there is no displacement. ..

Therefore, as shown in FIGS. 6A and 6B, periodic noise is extracted from the baseline vector y (t) in the first correction baseline vector yc1 (t) output from the subtractor 553. It will be.

The filter 554 removes high frequency noise that is constantly superimposed. Specifically, the filter 554 is a low-pass filter and has a cutoff frequency Th (LPF) (see FIG. 6C). The filter 554 passes a frequency component lower than the cutoff frequency Th (LPF) included in the first correction baseline vector yc1 (t) and attenuates a frequency component higher than the cutoff frequency Th (LPF). , The second correction baseline vector yc2 (t) is output. By this processing, as shown in FIG. 6C, high frequency noise included in the first correction baseline vector yc1 (t) can be suppressed. This high frequency noise is caused by noise such as thermal noise included in the carrier phase, for example.

As described above, by providing the analysis unit 54 with the above-mentioned filter, unnecessary components such as periodic noise and high frequency noise included in the baseline vector can be suppressed, and the baseline vector can be calculated with high accuracy.

It is also possible to omit one of the filters 551 and 554 depending on the required calculation accuracy of the baseline vector, the setting of the filter (setting of the cutoff frequency), and the like.

Further, in the above description, the analysis unit 54 has shown an aspect of performing up to the calculation of the baseline vector. However, the analysis unit 54 can also calculate the displacement of the installation position of the observation device 20 using this baseline vector. As a result, displacement can be observed with high accuracy, and for example, ground displacement and landslide can be detected with high accuracy.

In the above description, the mode in which the calculation of the baseline vector by the analysis device is executed by dividing it into a plurality of functional parts is shown. However, the analysis process of the observation data shown in FIG. 7 may be programmed and executed by an information processing device such as a computer. FIG. 7 is a flowchart showing a method of analyzing observation data according to the embodiment of the present invention.

The information processing apparatus acquires an analysis data file including time-series observation data (S101). At this time, the information processing apparatus acquires a number of analysis data files according to the time to be analyzed. The analysis data file contains the amount of observation data required for one calculation of the baseline vector. This process corresponds to the "data acquisition process" of the present invention.

The information processing apparatus extracts each observation data from the analysis data file (S102). At this time, the information processing apparatus extracts the observation data in a state associated with the observation time.

The information processing apparatus extracts the time-series observation data as one group, which is necessary for calculating the baseline vector (S103). This process corresponds to the "data extraction step" of the present invention. At this time, the step of determining a plurality of observation vectors constituting the group as described above corresponds to the "data combination step" of the present invention. The information processing apparatus calculates the baseline vector using the observation data of the group (S104). This step corresponds to the "baseline vector calculation step" of the present invention.

If the information processing apparatus has calculated a desired number of baseline vectors (S105: YES), the information processing apparatus extracts periodic noise (S107). This step corresponds to the "periodic noise extraction step" of the present invention. The information processing apparatus subtracts periodic noise from the baseline vector (S108). This step corresponds to the "subtraction step" of the present invention. By executing such processing, as described above, unnecessary components such as periodic noise can be suppressed, and the baseline vector can be calculated with high accuracy.

If the information processing apparatus has not calculated a desired number of baseline vectors (S105: NO), the information processing apparatus continues the calculation process of the baseline vectors. Specifically, the information processing apparatus shifts the position of the observation data at the earliest time of the group in the plurality of observation data in the time series to form a new group (S106). The shift amount at this time is defined by a time shorter than the time required to acquire the number of observation data required for calculating the baseline vector. The information processing apparatus calculates the baseline vector using the observation data of the newly configured group (S104).

By performing such processing, the baseline vector can be calculated at a time interval shorter than the time required to acquire the number of observation data required for calculating the baseline vector.

In the above description, the number of observation data required for calculating the baseline vector is used as one analysis data file, but the number of observation data constituting the analysis data file is limited to this. It suffices if it contains a plurality of observation data.

Further, the analysis unit of the analysis device may have each of the following aspects.

FIG. 8 is a configuration diagram of a second aspect of the analysis unit of the observation data analysis device according to the embodiment of the present invention. The analysis unit 54A shown in FIG. 8 is obtained by adding the threshold value determination units 555 and 556 to the analysis unit 54 of the first aspect shown in FIG. Other configurations of the analysis unit 54A are the same as those of the analysis unit 54, and the description of this same part will be omitted.

The analysis unit 54A includes a threshold value determination unit 555 and 556. The threshold determination unit 555 performs known adaptive processing, and monitors the suppression state of the periodic noise of the first correction baseline vector yc1 (t) output from the subtractor 553, while monitoring the cutoff frequency Th (HPF). ) Appropriately and feed back to the filter 551. Similarly, the threshold value determination unit 556 also performs known adaptive processing, and monitors the suppression state of the high frequency noise of the second correction baseline vector yc2 (t) output from the filter 554, while monitoring the cutoff frequency Th ( LPF) is adjusted appropriately and fed back to the filter 554.

With such a configuration, the threshold values of the filters 551 and 554 are appropriately adjusted according to the situation, so that the baseline vector can be calculated with higher accuracy.

FIG. 9 is a configuration diagram of a third aspect of the analysis unit of the observation data analysis device according to the embodiment of the present invention. The analysis unit 54B shown in FIG. 9 is obtained by adding a filter 557 to the analysis unit 54 of the first aspect shown in FIG. Other configurations of the analysis unit 54B are the same as those of the analysis unit 54 except for the setting of parameters, and the description of this same part will be omitted.

The analysis unit 54B includes a filter 557. The filter 557 removes sudden noise such as thermal noise and spike noise. Specifically, the filter 557 is a low-pass filter. The cutoff frequency of the filter 557 is higher than the cutoff frequency of the filter 554. The filter 557 corresponds to the "high frequency noise reduction filter" of the present invention. The step of executing this process in the information processing apparatus corresponds to the "high frequency noise removing step" of the present invention.

By providing such a filter 557, the influence of sudden noise can be suppressed, and the baseline vector can be calculated with higher accuracy.

Although FIG. 9 shows an embodiment in which the filter 551 which is an HPF and the filter 557 which is an LPF are connected in series, a BPF filter which realizes these characteristics at the same time may be used.

A threshold value setting unit for each filter shown in the analysis unit 54A may be added to the analysis unit 54B according to the present embodiment.

FIG. 10 is a configuration diagram of a fourth aspect of the analysis unit of the observation data analysis device according to the embodiment of the present invention. The analysis unit 54C shown in FIG. 10 is obtained by adding a filter 554C to the analysis unit 54 of the first aspect shown in FIG. Other configurations of the analysis unit 54C are the same as those of the analysis unit 54, and the description of this same part will be omitted.

The analysis unit 54C includes a filter 554C. The filter 554C removes stationary high frequency components. Specifically, the filter 554C is the same low-pass filter as the filter 554 and has a threshold value different from that of the filter 554. The filter 554C is connected to the subtractor 553. The filter 554C filters the input first correction baseline vector yc1 (t) and outputs the third correction baseline vector yc3 (t).

In such a configuration, a plurality of baseline vectors from which high frequency components of different frequency bands are removed are output. As a result, when the rate of change of the observed amount differs depending on the situation, such as when the displacement speed and displacement amount of the ground differ depending on the ground condition (moisture content, etc.), these baseline vectors are observed. More accurate detection (detection of landslides, etc.) can be realized.

It is also possible to add a threshold setting unit for each filter shown in the analysis unit 54A to the analysis unit 54C according to the present embodiment, and to apply the configuration of the filter shown in the analysis unit 54B.

In the above-described embodiment, the observation system is provided with one line aggregation device 30. Then, the observation data of all the observation devices 20 are acquired by the line aggregation device 30. However, a plurality of relay devices may be connected (wireless or wired) to the line aggregation device 30, and the plurality of relay devices may receive observation files from different observation devices 20. In other words, the plurality of observation devices 20 select one relay device from the plurality of relay devices and transmit the observation file. The plurality of relay devices transmit the received observation file to the line aggregation device 30. In this case, the relay device does not need to have a memory.

Further, communication between the relay devices may be enabled, and the observation file may be transmitted to the line aggregation device 30 via a plurality of relay devices. In other words, a plurality of relay devices may be connected in a string. Further, these may be combined to form a tree structure by a plurality of relay devices and a plurality of observation devices 20 starting from the line aggregation device 30. That is, any other configuration may be used as long as the system configuration can store the analysis data files of the plurality of analysis devices 20 in the server 40 via a communication network such as a network. Further, the function of the observation device 20 may be added to the relay device.

Further, in the above description, the number of observation files included in the analysis data file transmitted by the line aggregation device 30, that is, the number of observation data is the minimum number necessary for calculating the baseline vector, but the number is not limited to this. do not have. The data file for analysis may be composed of at least a plurality of observation files. However, when the number of observation files included in the analysis data file is larger than the number is smaller, the frequency with which the line aggregation device 30 communicates using the network 100 can be reduced.

Further, in the above description, the embodiment in which the FTP server is used as the server 40 is shown, but it has a function of storing the analysis data file from the line aggregation device 30, and the analysis device 50 stores the analysis data file. As long as it has a function that can be used, it is not limited to the FTP server.

Further, in the above description, the aspect in which the server 40 and the analysis device 50 are separated from each other is shown, but these may be integrated. In this case, the integrated device may be able to provide the calculated baseline vector, the displacement state of the ground based on the analysis result of the baseline vector, and the like to the outside.

Further, by equipping the observation device 20, the line aggregation device 30, and the relay device with predetermined computing power, the analysis processing of the baseline vector executed by the analysis device 50 can be performed by the observation device 20, the line aggregation device 30, and the line aggregation device 30. It is also possible to let the relay device do it. In this case, the server 40 may not be used.

Further, in the above description, the case of applying to landslide detection by static positioning has been shown, but the above-mentioned filter processing for the baseline vector can also be applied to RTK (real-time kinematic) positioning, kinematic positioning, and PPP (precision single positioning). It is possible to apply.

10: Observation data analysis system 20: Observation device 21: GNSS receiver 22: Communication control unit 23: GNSS antenna 24: Communication antenna 30: Line aggregation device 31: Communication antenna 32: Line aggregation unit 33: Router 34: Memory 40: Server 50: Analysis device 51: Data acquisition unit 52: Data extraction unit 54, 54A, 54B, 54C: Analysis unit 100: Network 540: Baseline vector calculation unit 551: Filter 552: Delay unit 553: Subtractor 554,554C : Filter 555, 556: Threshold determination unit 557: Filter

Claims (4)

Baseline vector calculation unit that calculates the baseline vector using observation data including carrier phase,
A periodic noise extraction filter that extracts periodic noise included in the baseline vector, and
A subtractor that subtracts the periodic noise from the baseline vector,
A data acquisition unit that acquires an analysis data file containing the observation data in time series from the outside,
A data extraction unit that extracts the plurality of observation data included in the analysis data file, and a data extraction unit.
A data combination unit for combining the number of time-continuous observation data required for calculating the baseline vector from the plurality of extracted observation data is provided.
The data combination unit is the most of the predetermined number of observation data on the latest time side in the plurality of observation data of the time series constituting the first combination and the plurality of observation data of the time series constituting the second combination. Combine multiple observation data so that the predetermined number of observation data on the earlier time side becomes common observation data.
The baseline vector calculation unit calculates the baseline vector for each combination of the plurality of observation data using the plurality of observation data combined by the data combination unit.
The periodic noise extraction filter extracts the periodic noise by using the baseline vector calculated for each combination of the plurality of observation data combined in the data combination unit.
Observation data analysis device. The observation data analysis device according to claim 1 .
A cutoff frequency determination unit for the periodic noise extraction filter is provided, which determines the cutoff frequency of the periodic noise extraction filter while monitoring the suppression state of the periodic noise.
Observation data analysis device. The observation data analysis device according to claim 1 or 2 ,
A plurality of observation devices that receive a positioning signal and output the observation data including the carrier phase, and
A relay device that generates the analysis data file using the observation data from a plurality of observation devices and outputs the analysis data file, and a relay device.
A server for storing the analysis data file is provided.
The observation data analysis device acquires the plurality of analysis data files from the server.
Observation data analysis system. Data acquisition process to acquire analysis data files including time-series observation data from the outside,
A data extraction step for extracting the plurality of observation data included in the analysis data file, and
A data combination step of combining the number of time-continuous observation data required for calculating the baseline vector from the plurality of extracted observation data, and
A baseline vector calculation step for calculating the baseline vector using the observation data including the carrier phase, and
A periodic noise extraction step for extracting periodic noise included in the baseline vector, and
A subtraction step of subtracting the periodic noise from the baseline vector,
Have,
The data combination step is the most in the predetermined number of observation data on the latest time side in the plurality of observation data in the time series constituting the first combination and the plurality of observation data in the time series constituting the second combination. Combine multiple observation data so that the predetermined number of observation data on the earlier time side becomes common observation data.
In the baseline vector calculation step, the baseline vector is calculated for each combination of the plurality of observation data using the plurality of observation data combined in the data combination step.
In the periodic noise extraction step, the periodic noise is extracted using the baseline vector calculated for each combination of the plurality of observation data combined in the data combination step.
Observation data analysis method.

Images