JP6943746B2 - Data processing equipment, displacement observation system, data processing method, and data processing program - Google Patents

Description

The present invention relates to a technique for smoothing a plurality of analysis data composed of time series.

Positioning using GNSS (Global Navigation Satellite Systems) positioning signals such as GPS (Global Positioning System) has been put into practical use. As a highly accurate positioning method, there is positioning (interference positioning) using the carrier phase of the positioning signal.

For example, Patent Document 1 describes static positioning, which is a type of positioning using carrier phase. Patent Document 1 is a system for observing the displacement of a slope, and smoothes the analysis data (displacement data) obtained by static positioning. As a result, the variation included in the analysis data (displacement data) is suppressed.

As a kind of such smoothing process, there is a smoothing process using a weighting coefficient. Roughly speaking, a weighting coefficient for smoothing processing is set for each of the current analysis data and the past analysis data, and the smoothing value is the weighted current analysis data and the weighted past. It is calculated using the analysis data of.

In the past, in many cases, the weighting factor was determined by the residual between the smoothed value and the analysis data. More specifically, the weighting factors were adaptively processed to minimize the residuals.

Japanese Unexamined Patent Publication No. 2004-144623

However, the adaptive process that minimizes the residual has the effect of suppressing normal noise, but the weighting coefficient at this time is set so as to follow non-normal noise such as multipath. Therefore, the effect of suppressing non-normal noise is low.

On the other hand, in the smoothing process, if the weighting of the past analysis data is increased, the non-normal noise can be suppressed as well as the normal noise, but the followability to the fluctuation of the analysis data is lowered.

Therefore, an object of the present invention is to provide a data processing technique that effectively suppresses noise including non-normal noise and has high followability to fluctuations of time-series data.

The data processing apparatus of the present invention includes a forward smoothing processing unit, a reverse smoothing processing unit, and a smoothing method determining unit. The forward smoothing processing unit sequentially smoothes the observed time-series data before processing from the past to the present, and calculates the smoothing time-series data. The reverse smoothing processing unit sequentially smoothes the forward smoothing time series data from the present time to the past, and calculates the backward smoothing time series data. The smoothing method determination unit determines the integrated value of the difference value between the backward smoothing time series data and the unprocessed time series data within a predetermined period, and the backward smoothing time series data and the forward smoothing time series data. Integrated value of difference value within a predetermined period, temporal difference value of backward smoothing time series data, temporal difference value of time series data before processing, temporal difference value of forward smoothing time series data The responsiveness of the forward smoothing processing unit and the backward smoothing processing unit is determined according to at least one of the above.

In this configuration, the responsiveness of the smoothing process is adjusted according to the amount of change in the time series data, that is, the magnitude of the change, the speed, the acceleration, and the like.

According to the present invention, noise can be effectively suppressed and the followability to fluctuations in time series data can be improved.

It is a block diagram of the analysis data processing apparatus which concerns on embodiment of this invention. (A) is a block diagram which shows the 1st aspect of the smoothing method determination part which concerns on 1st Embodiment of this invention, and (B) is the smoothing method determination which concerns on 1st Embodiment of this invention. It is a block diagram which shows the 2nd aspect of a part. It is a graph for demonstrating the concept of determination of the smoothing coefficient which concerns on 1st Embodiment of this invention. (A) and (B) are diagrams showing the transition between the smoothed value by the analysis data processing apparatus according to the first embodiment of the present invention and the smoothed value of the comparative configuration, and (C) is the figure showing the transition between the smoothed value of the comparative configuration. It is a figure which shows the transition of the filter coefficient of the analysis data processing apparatus which concerns on 1st Embodiment of. It is a flowchart which shows the main processing of the analysis result processing method which concerns on 1st Embodiment of this invention. (A) is a block diagram showing a third aspect of the smoothing method determination unit, and (B) is a block diagram showing a fourth aspect of the smoothing method determination unit. 6 (A) and 6 (B) are graphs for explaining the concept of determining the filter coefficient and the function in the smoothing method determining unit of the configuration shown in FIG. 6 (B). It is a functional block diagram of the observation system including the analysis data processing apparatus which concerns on this embodiment.

The data processing apparatus, the data processing method, and the data processing program according to the embodiment of the present invention will be described with reference to the drawings. FIG. 1 is a block diagram of an analysis data processing device according to an embodiment of the present invention.

As shown in FIG. 1, the analysis data processing device 10 includes a forward smoothing processing unit 11, a reverse smoothing processing unit 12, and a smoothing method determining unit 13. The forward smoothing processing unit 11, the reverse smoothing processing unit 12, and the smoothing method determining unit 13 are realized by, for example, hardware such as an arithmetic unit and a program executed in the hardware. The analysis data processing device 10 corresponds to the "data processing device" of the present invention. The forward smoothing processing unit 11 and the reverse smoothing processing unit 12 correspond to the "smoothing processing unit" of the present invention.

A plurality of analysis data D composed of time series are input to the forward smoothing processing unit 11, the reverse smoothing processing unit 12, and the smoothing method determining unit 13. The analysis data D is, for example, the displacement amount of the baseline vector calculated from the carrier phase (corresponding to the observed value) of the positioning signal. The analysis data D is not limited to the displacement amount of the baseline vector, and may be a value calculated from the observed values sequentially acquired at predetermined time intervals. Furthermore, the observed value itself may be used as analysis data. The analysis data D corresponds to the "time series data before processing" of the present invention.

The forward smoothing processing unit 11 executes forward smoothing processing of a plurality of analysis data D arranged in a time series. The forward smoothing process is a process of calculating the current forward smoothing value Dsa by using the past forward smoothing value and the current analysis data. The forward smoothing process is realized by a weighted averaging process, a Kalman filter process, or the like. The forward smoothing value Dsa corresponds to the "time series data of forward smoothing" of the present invention.

At this time, the forward smoothing processing unit 11 performs the forward smoothing processing by using the filter coefficient Ga or the function fa. The filter coefficient Ga or the filter function fa is determined by the smoothing method determination unit 13 by the process described later.

The forward smoothing processing unit 11 outputs the forward smoothing value Dsa to the reverse smoothing processing unit 12.

The reverse smoothing processing unit 12 executes the backward smoothing processing of a plurality of forward smoothing values Dsa arranged in a time series. The reverse smoothing process is performed using the latest forward smoothing process time as a reference time and a forward smoothing value Dsa earlier than this time. In the reverse smoothing process, the forward smoothing value Dsa is sequentially smoothed in the past direction, and the reverse smoothing value Dsb is sequentially calculated. The reverse smoothing process is realized by a weighted averaging process, a Kalman filter process, or the like, similarly to the forward smoothing process. The reverse smoothing value Dsb corresponds to the "time series data of reverse smoothing" of the present invention.

At this time, the reverse smoothing processing unit 12 performs the forward smoothing processing using the filter coefficient Gb or the function fb. The filter coefficient Ga or the filter function fa is determined by the smoothing method determination unit 13 by the process described later.

The reverse smoothing processing unit 12 outputs the reverse smoothing value Dsb to the outside as output data of the analysis data processing device 10, and also outputs the smoothing method determination unit 13.

The smoothing method determination unit 13 determines the responsiveness of each smoothing process, that is, the filter coefficients Ga and Gb, or the functions fa and fb, using the changed values of the analysis data. When the smoothing process is set using the filter coefficient, the smoothing method determining unit 13 outputs the filter coefficient Ga to the forward smoothing processing unit 11 and outputs the filter coefficient Gb to the backward smoothing processing unit 12. When the smoothing process is set by using the function, the smoothing method determining unit 13 outputs the function fa to the forward smoothing process unit 11 and outputs the function fb to the backward smoothing process unit 12. The smoothing method determination unit 13 may output both the filter coefficient and the function to the forward smoothing processing unit 11 and the backward smoothing processing unit 12. Specific processing in the smoothing method determination unit 13 will be described later.

FIG. 2A is a block diagram showing a first aspect of the smoothing method determining unit according to the first embodiment of the present invention, and FIG. 2B is a block diagram according to the first embodiment of the present invention. It is a block diagram which shows the 2nd aspect of the smoothing method determination part.

FIG. 3 is a graph for explaining the concept of determining the smoothing coefficient according to the first embodiment of the present invention. In FIG. 3, the horizontal axis is the elapsed time (observation time), and the vertical axis is the displacement. In FIG. 3, the fine solid line is the analysis data. The thick solid line is the forward smoothing value Dsa, and the broken line is the backward smoothing value Dsb.

In the embodiment shown in FIG. 2A, the smoothing method determination unit 13 includes a difference value calculation unit 131, an integrated value calculation unit 132, and a coefficient setting unit 133. In the embodiment shown in FIG. 2B, the smoothing method determination unit 13 includes a difference value calculation unit 131, an integrated value calculation unit 132, and a function setting unit 134. In the aspect shown in FIG. 2A and the aspect shown in FIG. 2B, the difference value calculation unit 131 and the integrated value calculation unit 132 have the same configuration.

The forward smoothing value Dsa and the reverse smoothing value Dsb are input to the difference value calculation unit 131. As shown in FIGS. 2 (A) and 2 (B), the analysis data D may be input to the difference value calculation unit 131.

The difference value calculation unit 131 calculates the difference value ΔD between the forward smoothing value Dsa and the reverse smoothing value Dsb. At this time, as shown in FIG. 3, the difference value calculation unit 131 sets the time ts (m) in the past by a predetermined time with respect to the latest time ts (i), and from the time ts (m) to the time ts (i). ), The difference value ΔD is calculated at each time in Pts.

In the mode in which the analysis data D is input, the difference value calculation unit 131 may calculate the difference value ΔD between the analysis data D and the reverse smoothing value Dsb instead of the forward smoothing value Dsa. In this case, the difference value ΔD may be smoothed.

The difference value calculation unit 131 outputs the difference value ΔD at each time to the integrated value calculation unit 132.

The integrated value calculation unit 132 integrates the difference value ΔD of each time with respect to the period Pts from the time ts (m) to the time ts (i), and calculates the integrated value Sts. The integrated value Sts corresponds to the "change value" of the present invention.

The calculation process of the difference value ΔD in the difference value calculation unit 131 and the calculation process of the integrated value Sts in the integrated value calculation unit 132 are sequentially performed at time intervals for changing the setting of the smoothing coefficient. The time interval for changing the setting of the smoothing coefficient is usually longer than the acquisition interval of the plurality of analysis data D, but may be the same.

The coefficient setting unit 133 sets the filter coefficients Ga and Gb according to the integrated value Sts. The filter coefficients Ga and Gb are coefficients for setting the weight of the analysis data D with respect to the prior smoothed values Dsa and Dsb.

For example, when the time constant of the filter is designed to be large when the filter coefficients Ga and Gb are large, if the filter coefficients Ga and Gb are large, the weight of the analysis data D becomes relatively large and the responsiveness becomes fast. The degree of smoothing becomes small. On the contrary, if the filter coefficients Ga and Gb are small, the weight of the analysis data D is relatively small, the weights of the preliminary smoothing values Dsa and Dsb are relatively large, the responsiveness is slowed, and the degree of smoothing is low. growing.

If the integrated value Sts is large, the coefficient setting unit 133 sets the filter coefficients Ga and Gb to large values. If the integrated value Sts is small, the coefficient setting unit 133 sets the filter coefficients Ga and Gb to small values. That is, when the forward smoothing value Dsa and the backward smoothing value Dsb deviate significantly, the filter coefficients Ga and Gb are set to large values. On the other hand, when the forward smoothing value Dsa and the backward smoothing value Dsb are close to each other, the filter coefficients Ga and Gb are set to small values.

By using such settings of the filter coefficients Ga and Gb, as shown in FIG. 3, during the period when the change in the analysis data is small, the forward smoothing value Dsa and the backward smoothing value Dsb are stable in the past. It is greatly affected by the analysis data. Therefore, the forward smoothing value Dsa and the backward smoothing value Dsb are stable because the influence of noise including non-normal noise is suppressed.

On the other hand, as shown in FIG. 3, during the period when the analysis data fluctuates, the forward smoothing value Dsa and the backward smoothing value Dsb change so as to be greatly influenced by the analysis data at each time. Therefore, the followability of the forward smoothing value Dsa and the backward smoothing value Dsb with respect to the fluctuation of the analysis data is improved.

It is better to set the period of the aperiodic noise to a long time constant that can suppress the period of the aperiodic noise during the period when the change of the analysis data D is small, and fix the time constant in the period. As a result, a more stable smoothing value can be obtained in a period of little change.

The function setting unit 134 sets the functions fa and fb according to the integrated value Sts. The functions fa and fb set the relationship between the weights of the prior smoothed values Dsa and Dsb and the weights of the analysis data D. The functions fa and fb are set by, for example, a linear function, a quadratic function, or the like. By appropriately setting the functions fa and fb, the weight of the analysis data D can be made relatively larger or smaller than the weight of the smoothed values Dsa and Dsb after the prior smoothing process. The relationship between the integrated value Sts and the functions fa and fb is stored, for example, in a correspondence table.

If the integrated value Sts is large, the function setting unit 134 sets a function that reduces the weights of the prior smoothed values Dsa and Dsb and increases the weight of the analysis data D. If the integrated value Sts is small, the coefficient setting unit 133 sets a function that increases the weights of the prior smoothed values Dsa and Dsb and decreases the weight of the analysis data D. That is, when the forward smoothing value Dsa and the backward smoothing value Dsb deviate significantly, a function is set that reduces the weights of the preliminary smoothing values Dsa and Dsb and increases the weight of the analysis data D. On the other hand, when the forward smoothing value Dsa and the backward smoothing value Dsb are close to each other, a function is set to increase the weights of the preliminary smoothing values Dsa and Dsb and decrease the weight of the analysis data D.

By using the settings of the functions fa and fb, as shown in FIG. 3, during the period when the change in the analysis data is small, the forward smoothing value Dsa and the backward smoothing value Dsb are obtained from a plurality of stable analyzes in the past. Greatly affected by data. Therefore, the forward smoothing value Dsa and the backward smoothing value Dsb are stable because the influence of noise including non-normal noise is suppressed.

On the other hand, as shown in FIG. 3, during the period when the analysis data fluctuates, the forward smoothing value Dsa and the backward smoothing value Dsb change so as to be greatly influenced by the analysis data at each time. Therefore, the followability of the forward smoothing value Dsa and the backward smoothing value Dsb with respect to the fluctuation of the analysis data is improved.

4 (A) and 4 (B) are diagrams showing the transition between the smoothed value by the analysis data processing apparatus according to the first embodiment of the present invention and the smoothed value of the comparative configuration. The horizontal axis of FIGS. 4 (A) and 4 (B) is the elapsed time, and the vertical axis of FIGS. 4 (A) and 4 (B) is the displacement. In FIGS. 4 (A) and 4 (B), the fine solid line is the analysis data, the thick solid line is the smoothed value of the present application, and the broken line is the smoothed value of the comparative configuration. The comparative configuration A shown in FIG. 4 (A) is configured so that the influence of the past analysis data is greater than that of the current analysis data. The comparative configuration B shown in FIG. 4 (B) is configured so that the influence of the current analysis data is greater than that of the past analysis data.

FIG. 4C is a diagram showing a transition of the filter coefficient of the analysis data processing apparatus according to the first embodiment of the present invention. In FIG. 4C, the horizontal axis is the elapsed time and the vertical axis is the filter coefficient. Further, in the figure of FIG. 4C, the functions adopted at each time are described. The position of each time on the horizontal axis of FIGS. 4 (A) and 4 (B) and the position of each time on the horizontal axis of FIG. 4 (C) are the same.

As shown in FIGS. 4 (A) and 4 (B), in the analysis data processing device 10 of the present embodiment, noise including non-normal noise of the analysis data is suppressed, and the followability to the fluctuation of the analysis data is improved. high. On the other hand, in the comparative configuration A shown in FIG. 4 (A), the noise of the analysis data is suppressed, but the followability to the fluctuation of the analysis data is low. Further, in the comparative configuration B shown in FIG. 4B, non-normal noise may be followed.

Further, as shown in FIG. 4C, the analysis data processing device 10 according to the present embodiment can appropriately set the smoothing coefficient and the function according to the fluctuation of the analysis data.

In the above description, an embodiment in which the analysis data processing device 10 is configured by a plurality of functional units is shown. However, one arithmetic processing unit may perform the processing executed by the plurality of functional units described above. In this case, the arithmetic processing unit programs and stores the functions performed by the plurality of functional units described above. The arithmetic processing unit reads and executes this program.

FIG. 5 is a flowchart showing the main processing of the analysis result processing method according to the first embodiment of the present invention.

The arithmetic processing unit acquires a plurality of analysis data composed of time series (S101). The arithmetic processing unit determines the smoothing method as described above according to the change values of the plurality of analysis data (S102). For example, in the above example, the arithmetic processing apparatus determines the filter coefficients Ga, Gb or the functions fa, fb using the integrated value of the difference value between the analysis data D or the forward smoothing value Dsa and the backward smoothing value Dsb. do. The arithmetic processing unit uses this smoothing method to execute the smoothing process of the analysis data (S103).

In the above description, the aspect of using the integrated value of the difference value between the analysis data D or the forward smoothing value Dsa and the backward smoothing value Dsb is shown, but the analysis data D, the forward smoothing value Dsa, or the backward smoothing value Dsa is used. It is also possible to determine the smoothing method by the temporal difference value (differential value) of any of the smoothing values Dsb.

The smoothing method determination unit of the analysis data processing apparatus can also be realized by the configuration of each of the following embodiments. FIG. 6A is a block diagram showing a third aspect of the smoothing method determination unit. FIG. 6B is a block diagram showing a fourth aspect of the smoothing method determination unit. FIG. 7 is a graph for explaining the concept of determining the filter coefficient and the function in the configuration smoothing method determination unit shown in FIGS. 6 (A) and 6 (B). In FIG. 7, the horizontal axis is the elapsed time (observation time), and the vertical axis is the displacement. In FIG. 7, the fine solid line is the analysis data. The thick solid line is the forward smoothing value Dsa, and the broken line is the backward smoothing value Dsb.

In the smoothing method determination unit shown in FIG. 6 (A), the difference value calculation unit 131 and the integrated value calculation unit 132 are omitted from the smoothing method determination unit shown in FIG. 2 (A), and the differential value calculation unit is omitted. 135 has been added. Further, in the smoothing method determination unit shown in FIG. 6B, the difference value calculation unit 131 and the integrated value calculation unit 132 are omitted from the smoothing method determination unit shown in FIG. 2B, and the differential value is differentiated. Calculation unit 135 has been added. Due to these changes, the processing of the coefficient setting unit 133 and the processing of the function setting unit 134 are different.

The differential value calculation unit 135 calculates the differential value at the latest time (corresponding to ts (i) in FIG. 7) at which the forward smoothing value Dsa is calculated. For example, when the differential value calculation unit 135 calculates the differential value with respect to the forward smoothing value Dsa, the difference between the forward smoothing value Dsa at the latest time and the forward smoothing value Dsa earlier than this time is used. The differential value dDsa / dt is calculated by calculating the value and dividing by the time difference. At this time, the differential value calculation unit 135 uses the forward smoothing value Dsa (i) at the latest time ts (i) and the past forward smoothing value Dsa (i-N) to obtain the differential value dDsa (i) /. It is preferable to calculate dt. N is an integer, and ts (i-N) means a time N before the latest time ts (i). N is a value that can be set as appropriate, and by reducing N, the delay of the smoothed value from the analysis data can be reduced when the change in the analysis data is large.

The calculation process of the differential value dDsa (i) / dt in the differential value calculation unit 135 is sequentially performed at time intervals for changing the setting of the smoothing coefficient. The time interval for changing the setting of the smoothing coefficient is usually longer than the acquisition interval of the plurality of analysis data D, but may be the same.

In the configuration of FIG. 6A, the differential value calculation unit 135 outputs the differential value dDsa / dt of the forward smoothing value Dsa to the coefficient setting unit 133. This differential value dDsa / dt corresponds to the "change value" of the present invention.

The coefficient setting unit 133 sets the filter coefficients Ga and Gb according to the differential value dDsa / dt.

For example, if the differential value dDsa / dt is large, the coefficient setting unit 133 sets the filter coefficients Ga and Gb to large values. In other words, set the time constant for filtering the input short. If the differential value dDsa / dt is small, the coefficient setting unit 133 sets the filter coefficients Ga and Gb to small values. In other words, set a long time constant for filtering the input. That is, when the change in the forward smoothing value Dsa is large, the filter coefficients Ga and Gb are set to large values. On the other hand, when the change in the forward smoothing value Dsa is small, the filter coefficients Ga and Gb are set to small values.

By using such settings of the filter coefficients Ga and Gb, the forward smoothing value Dsa and the backward smoothing value Dsb are greatly affected by a plurality of stable analysis data in the past during a period in which the change in the analysis data is small. .. Therefore, the forward smoothing value Dsa and the backward smoothing value Dsb are stable because the influence of noise including non-normal noise is suppressed.

On the other hand, during the period when the analysis data fluctuates, the forward smoothing value Dsa and the backward smoothing value Dsb change so as to be greatly influenced by the analysis data at each time. Therefore, the followability of the forward smoothing value Dsa and the backward smoothing value Dsb with respect to the fluctuation of the analysis data is improved.

In the configuration of FIG. 6B, the differential value calculation unit 135 outputs the differential value dDsa / dt of the forward smoothing value Dsa to the function setting unit 134. This differential value dDsa / dt corresponds to the "change value" of the present invention.

The function setting unit 134 sets the functions fa and fb according to the differential value dDsa / dt.

For example, if the differential value dDsa / dt is large, the function setting unit 134 sets a function that reduces the weights of the prior smoothed values Dsa and Dsb and increases the weight of the analysis data D. In other words, set the time constant for the input of the function short. If the integrated value Sts is small, the coefficient setting unit 133 sets a function that increases the weights of the prior smoothed values Dsa and Dsb and decreases the weight of the analysis data D. In other words, set a long time constant for the input of the function. When the change in the analysis data D is large, a function is set in which the weights of the prior smoothed values Dsa and Dsb are reduced and the weight of the analysis data D is increased. On the other hand, when the change in the analysis data D is small, a function is set in which the weights of the prior smoothed values Dsa and Dsb are increased and the weight of the analysis data D is decreased.

By using such settings of the functions fa and fb, the forward smoothing value Dsa and the backward smoothing value Dsb are greatly affected by a plurality of stable analysis data in the past during a period in which the change in the analysis data is small. Therefore, the forward smoothing value Dsa and the backward smoothing value Dsb are stable because the influence of noise including non-normal noise is suppressed.

On the other hand, during the period when the analysis data fluctuates, the forward smoothing value Dsa and the backward smoothing value Dsb change so as to be greatly influenced by the analysis data at each time. Therefore, the followability of the forward smoothing value Dsa and the backward smoothing value Dsb with respect to the fluctuation of the analysis data is improved.

In the above description, the first derivative value of the forward smoothing value Dsa is used, but the nth derivative value of the forward smoothing value Dsa (n is an integer of 2 or more) may be used. Further, the m-th order differential value of the reverse smoothing value Dsb (m is an integer of 1 or more) may be used (for example, the first-order differential value dDsb / dt of the reverse smoothing value Dsb as shown in FIG. 7). Further, the m-th order differential value of the analysis data D (m is an integer of 1 or more) may be used (for example, the first-order differential value dD / dt of the analysis data D as shown in FIG. 7). Further, a smoothing value such as a simple moving average of the differential value may be used.

Such an analysis data processing device can be applied to the following system using static positioning. FIG. 8 is a functional block diagram of an observation system including the analysis data processing device according to the present embodiment. As shown in FIG. 8, the observation system 1 includes an observation station 21, an observation station 22, a reference station 30, a line aggregator 40, a server 50, and an analysis device 60. The observation system 1 corresponds to the "displacement observation system" of the present invention.

In FIG. 8, there are two observation stations and one reference station, but the present invention is not limited to this. The number of these observation stations and the number of reference stations are appropriately determined according to the observation location and the like. Further, although FIG. 8 shows an embodiment in which GPS is used, other positioning systems of GNSS can also be used.

The observation station 21, the observation station 22, and the reference station 30 are connected to the line aggregator 40 by wire or wirelessly. The line aggregator 40, the server 50, and the analysis device 60 are connected via the network 500.

The observation station 21 includes a GPS antenna 211 and a GPS receiver 212. The GPS antenna 211 receives positioning signals (GPS signals) transmitted from each of the plurality of positioning satellites SAT1, SAT2, SAT3, and SAT4, and outputs them to the GPS receiver 212. The GPS receiver 212 observes the carrier phase of the positioning signals from the plurality of positioning satellites SAT1, SAT2, SAT3, and SAT4, respectively. Further, the GPS receiver 212 positions the position of its own device by independent positioning using the code phase or the like. The GPS receiver 212 transmits these carrier phase and the positioning result to the line aggregator 40.

The observation station 22 includes a GPS antenna 221 and a GPS receiver 222. The GPS antenna 221 receives positioning signals (GPS signals) transmitted from each of the plurality of positioning satellites SAT1, SAT2, SAT3, and SAT4, and outputs them to the GPS receiver 222. The GPS receiver 222 observes the carrier phase of the positioning signals from the plurality of positioning satellites SAT1, SAT2, SAT3, and SAT4, respectively. In addition, the GPS receiver 222 positions its own device by independent positioning using the code phase or the like. The GPS receiver 222 transmits these carrier phase and the positioning result to the line aggregator 40.

The reference station 30 includes a GPS antenna 301 and a GPS receiver 302. The GPS antenna 301 receives positioning signals (GPS signals) transmitted from each of the plurality of positioning satellites SAT1, SAT2, SAT3, and SAT4, and outputs them to the GPS receiver 302. The GPS receiver 302 observes the carrier phase of the positioning signals from the plurality of positioning satellites SAT1, SAT2, SAT3, and SAT4, respectively. Further, the GPS receiver 302 positions its own device by independent positioning using the code phase or the like. The GPS receiver 302 transmits these carrier phase and the positioning result to the line aggregator 40.

The line aggregator 40 transmits the carrier phase and the positioning result from each of the observation station 21, the observation station 22, and the reference station 30 to the server 50. The server 50 is realized by, for example, an FTP server.

The analysis device 60 includes a calculation unit 61 and a storage unit 62. The calculation unit 61 has the function of the analysis data processing device 10 described above. Further, the arithmetic unit 61 acquires the carrier phase and the positioning result from each of the observation station 21, the observation station 22, and the reference station 30 from the server 50, and calculates a baseline vector connecting each station. The calculation unit 61 sequentially calculates the baseline vector at the preset observation time interval of the baseline vector. Further, the calculation unit 61 sequentially calculates the displacement amount of this baseline vector. For example, the displacement amount of this baseline vector corresponds to the "analysis data" of the present invention.

For example, in the case of a landslide observation system, this corresponds to the amount of ground fluctuation at the positions where the observation station 21 and the observation station 22 are installed. The amount of such ground fluctuation does not change stably unless there is ground fluctuation due to an earthquake, heavy rain, or the like. On the other hand, when ground fluctuations occur, the changes occur in a wide range depending on the factors. Therefore, such changes can be observed with high accuracy by using the carrier phase, and further, by using the configuration of the analysis data processing device 10 of the present invention, the influence of noise can be effectively suppressed and fluctuated. It is possible to realize high followability of the observed data. Further, when a positioning signal is used, non-normal noise such as a multipath error is generated, and such non-normal noise can be suppressed by using the configuration of the analysis data processing device 10 of the present invention.

In the above description, the mode of using the analysis data by static positioning is shown, but the above-mentioned configuration and processing are also used for the analysis data by kinematic positioning such as RTK (real-time kinematic) and precision independent positioning (PPP, etc.). Can be applied.

Further, in the above description, the embodiment including the line aggregator 40 is shown, but this may be omitted. In this case, the observation stations 21, 22 and the reference station 30 may output the carrier wave phase and the positioning result to the server 50. Further, the server 50 can be omitted. In this case, the line aggregator 40, the observation stations 21, 22 and the reference station 30 may output the carrier wave phase and the positioning result to the analysis device 60. The analysis device 60 stores these in the storage unit 62 and uses them for the above-mentioned processing.

Further, the carrier phase and the positioning result may be aggregated by any of the observation stations 21, 22 and the reference station 30, and the displacement amount of the baseline vector and the baseline vector may be calculated. In this case, the station that has calculated the displacement amount of the baseline vector may output the displacement amount of the baseline vector to the server 50.

Further, in the above description, the embodiment in which the analysis device 60 is provided separately from the observation stations 21 and 22 and the reference station 30 is shown, but the analysis device 60 is any one of the observation stations 21, 22 and the reference station 30. It may be built in.

1: Observation system 10: Analysis data processing device 11: Forward smoothing processing unit 12: Reverse smoothing processing unit 13: Smoothing method determination unit 21, 22: Observation station 30: Reference station 40: Line aggregator 50: Server 60: Analytical device 61: Calculation unit 62: Storage unit 131: Difference value calculation unit 132: Integrated value calculation unit 133: Coefficient setting unit 134: Function setting unit 135: Differential value calculation unit 211, 221 and 301: GPS antenna 212, 222, 302: GPS receiver 500: network

Claims (7)

A forward smoothing processing unit that sequentially smoothes the observed time-series data before processing from the past to the present and calculates the time-series data of forward smoothing.
A reverse smoothing processing unit that sequentially smoothes the forward smoothing time series data from the present time to the past and calculates the backward smoothing time series data.
The integrated value of the difference value between the backward smoothing time series data and the unprocessed time series data within a predetermined period, and the difference value between the backward smoothing time series data and the forward smoothing time series data. Integrated value within a predetermined period, temporal difference value of the backward smoothing time series data, temporal difference value of the time series data before processing, temporal difference value of the forward smoothing time series data A smoothing method determining unit that determines the responsiveness of the forward smoothing processing unit and the reverse smoothing processing unit according to at least one of the above.
A data processing device equipped with. The data processing device according to claim 1.
The smoothing method determination unit
When the amount of change in the time-series data before processing is large, the followability of the time-series data of the forward smoothing to the time-series data before processing is enhanced, and the amount of change in the time-series data before processing is increased. If it is small, the responsiveness is determined so as to enhance the stability of the time series data of the forward smoothing.
Data processing device. The data processing device according to claim 1 or 2.
The responsiveness is set by a filter coefficient or function.
Data processing device. The data processing device according to any one of claims 1 to 3.
The time series data before processing is
The amount of displacement of the baseline vector of the coordinates based on the carrier phase of the positioning signal,
Data processing device. The configuration of the data processing device according to claim 4 and
A plurality of stations connected to the data processing device, each of which receives the positioning signal and observes the carrier phase, respectively.
Displacement observation system equipped with. The observed time-series data before processing is sequentially smoothed from the past to the present, and the time-series data for forward smoothing is calculated.
The time-series data of the forward smoothing is sequentially smoothed from the present time to the past, and the time-series data of the backward smoothing is calculated.
The integrated value of the difference value between the backward smoothing time series data and the unprocessed time series data within a predetermined period, and the difference value between the backward smoothing time series data and the forward smoothing time series data. Integrated value within a predetermined period, temporal difference value of the backward smoothing time series data, temporal difference value of the time series data before processing, temporal difference value of the forward smoothing time series data The responsiveness of the forward smoothing process and the backward smoothing process is determined according to at least one of.
Data processing method. A process of sequentially smoothing the observed time-series data before processing from the past to the present, and a process of calculating the time-series data of forward smoothing.
The process of sequentially smoothing the time-series data of the forward smoothing from the present time to the past and calculating the time-series data of the backward smoothing,
The integrated value of the difference value between the backward smoothing time series data and the unprocessed time series data within a predetermined period, and the difference value between the backward smoothing time series data and the forward smoothing time series data. Integrated value within a predetermined period, temporal difference value of the backward smoothing time series data, temporal difference value of the time series data before processing, temporal difference value of the forward smoothing time series data The process of determining the responsiveness of the forward smoothing process and the backward smoothing process according to at least one of
Is a data processing program that causes the arithmetic processing unit to execute.

Images