JP2011102733A - Measuring device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To obtain a measuring device that obtains precise measurement results by detecting an accurate number of element waves by separating the respective element waves and does not require any high processing performance and wide storage regions. <P>SOLUTION: The measuring device includes: a plurality of sensors 1a-1m for receiving signals that are transmitted from a signal source and include a plurality of element waves propagated in mutually different paths; an angle measurement processing part 5 for estimating an unknown parameter including at least one of an incidence angle of a reception signal and distance to the signal source, based on reception signals received by the plurality of sensors 1a-1m; and a data length control part 7 for controlling observation data length, namely data length of reception signals input to the angle measurement processing part 5, such that signal correlation among a plurality of element waves approximates zero, based on the measurement processing results by the angle measurement processing part 5. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to a measuring device such as a horn measuring device that receives an incoming signal from a signal source by a plurality of sensors arranged side by side and estimates an incident angle (direction of the signal source) of the incoming signal, for example.

  In an angle measuring device using a plurality of sensors arranged side by side, the incident angle of an incoming signal is estimated based on an amplitude difference or a phase difference existing between observation data received by each sensor. An angle measuring device using a MUSIC (Multiple Signal Classification) method or the like is known as one that estimates an incident angle of an incoming signal using a plurality of sensors arranged side by side (see, for example, Non-Patent Document 1).

  In such an angle measuring device, a multipath propagation environment in which a plurality of routes are propagated is established before a signal transmitted from a signal source is received by each sensor. In this multipath propagation environment, when at least one of the signal source, the sensor, and the object (multipath generation source) that causes the multipath propagation environment due to reflection and diffraction of the signal moves independently, they are different from each other. A Doppler frequency difference is generated between the elementary waves of the signal propagated along the path and received by the sensor.

  Here, when the moving speed of any one of the signal source, the sensor, and the multipath generation source is large (the Doppler frequency difference is large), each elementary wave can be easily separated by the Doppler frequency difference. The angle measuring process can be easily executed. However, when the moving speed is low (the Doppler frequency difference is small), it is difficult to separate each elementary wave due to the Doppler frequency difference, and the accurate number of elementary waves cannot be detected, and a plurality of elementary waves are mixed. The angle measurement process is executed in the state. When a plurality of elementary waves with different Doppler frequencies are mixed and the incident angles of the plurality of elementary waves are close to each other, in the angle measurement processing result, with respect to the true value of the incident angle representing the azimuth of the signal source There was a problem that the bias error fluctuated with time.

  For example, when there is no Doppler frequency difference between the elementary waves, it is extremely difficult to separate the elementary waves, but in the angle measurement processing result, the bias error existing with respect to the true value of the incident angle is always constant. That is, since the output is steady without time fluctuation, the reliability of the angle measurement processing result is high. On the other hand, if there is a small Doppler frequency difference between the elementary waves, the reliability of the angle measurement result decreases because the bias error that exists with respect to the true value of the incident angle varies with time in the angle measurement result. To do.

  Note that one of the reasons why a plurality of elementary waves cannot be accurately separated is that the observation data length is not optimal with respect to the Doppler frequency difference. In general, in order to separate the Doppler frequency difference of fHz, an observation data length of at least 1 / f second is required. That is, in order to separate the 1 Hz Doppler frequency difference, an observation data length of at least 1 second is required. If the observation data length is shorter than 1 / f seconds, it is difficult to separate the individual waves.

  Moreover, it is considered that the angle measurement accuracy improves as the observation data length increases. This is because the signal-to-noise power ratio of the signal received by each sensor can be increased by increasing the observation data length. 2. Description of the Related Art Angle measuring devices having means for storing observation data are known as means for increasing observation data length to improve angle measurement accuracy (see, for example, Patent Documents 1 and 2).

Hereinafter, the conventional angle measuring device described in Patent Documents 1 and 2 will be described with reference to FIG. FIG. 7 is a block diagram showing a conventional angle measuring device.
In FIG. 7, the angle measuring device includes a plurality of sensors 51 a to 51 m, a receiving unit 52, an observation data storage unit 53, a data statistical processing unit 54, an angle measurement processing unit 55, and an angle measurement result output unit 56.

  The plurality of sensors 51a to 51m are arranged side by side and each receive a signal transmitted from a signal source (not shown). The receiving unit 52 receives signals received by the plurality of sensors 51a to 51m as observation data vectors in a unit time zone. The observation data storage unit 53 stores the observation data vector of the unit time zone received by the reception unit 52.

  The data statistical processing unit 54 performs statistical processing based on the observation data vector in the unit time zone received by the reception unit 52 and the past observation data vector stored in the observation data storage unit 53. The angle measurement processing unit 55 performs angle measurement processing based on the statistical data statistically obtained by the data statistical processing unit 54. The angle measurement result output unit 56 outputs the angle measurement processing result measured by the angle measurement processing unit 55.

Here, the data statistical processing unit 54 of the angle measuring device shown in FIG. 7 has a processing configuration for executing statistical processing with a predetermined observation data length, or a processing configuration for continuously executing statistical processing. ing. Moreover, in the angle measuring device shown in FIG. 7, the unit time zone for acquiring observation data may be set to a short time. This is to reduce the delay time due to signal processing and to quickly cope with fluctuations in the signal source.
As described above, since the unit time zone itself is short, it has been studied to improve the angle measurement accuracy by increasing the observation data length used for the statistical processing.

JP 2003-318792 A Japanese Patent Laid-Open No. 2007-40806

H. Krim, M .; Viberg, “Two Decades of Array Signal Processing Research: The Parametric Approach”, IEEE Signal Processing Magazine, Vol. 13, no. 4, pp. 67-94, Jul 1996

However, the prior art has the following problems.
In the conventional angle measuring devices described in Patent Documents 1 and 2, the signal-to-noise power ratio of the signal received by each sensor can be increased by increasing the observation data length, but the signal correlation between the elementary waves Is present, there is a problem that the individual waves cannot be easily separated and a highly accurate angle measurement process result cannot be obtained. In addition, by increasing the observation data length, there is a problem that processing performance for processing long observation data and a storage area for storing long observation data are required.

  The present invention has been made to solve the above-described problems, and by separating each element wave and detecting the accurate element number, a highly accurate measurement result can be obtained and a high processing performance can be obtained. Another object is to obtain a measuring device that does not require a large storage area.

  The angle measuring device according to the present invention is based on a plurality of sensors each receiving a signal including a plurality of elementary waves transmitted from a signal source and propagating on different paths, and a reception signal received by the plurality of sensors. Based on the measurement processing means for estimating an unknown parameter including at least one of the incident angle of the signal and the distance to the signal source, and the measurement processing result in the measurement processing means, the signal correlation between the plurality of elementary waves approaches zero. And a data length control means for controlling the observation data length which is the data length of the received signal inputted to the measurement processing means.

According to the measuring apparatus according to the present invention, the data length control means propagates through different paths from the signal source and is received by the plurality of sensors based on the measurement processing result in the measurement processing means for estimating the unknown parameter. The observation data length, which is the data length of the received signal input to the measurement processing means, is controlled so that the signal correlation between the plurality of elementary waves approaches zero. Thus, each elementary wave can be easily separated by controlling the observation data length so that the signal correlation between the plurality of elementary waves approaches zero.
Therefore, by separating each element wave and detecting the accurate element number, a highly accurate measurement result can be obtained, and a measurement apparatus that does not require high processing performance and a wide storage area can be obtained.

It is a block block diagram which shows the angle measuring device which concerns on Embodiment 1 of this invention. It is a block block diagram which shows the angle measuring device which concerns on Embodiment 2 of this invention. It is a block block diagram which shows the angle measuring device which concerns on Embodiment 3 of this invention. It is a block block diagram which shows the angle measuring device which concerns on Embodiment 4 of this invention. It is a block block diagram which shows the angle measuring device which concerns on Embodiment 5 of this invention. It is a block block diagram which shows the angle measuring device which concerns on Embodiment 6 of this invention. It is a block block diagram which shows the conventional angle measuring device.

Hereinafter, preferred embodiments of the measuring apparatus according to the present invention will be described with reference to the drawings. In the drawings, the same or corresponding parts will be described with the same reference numerals.
In the following embodiment, a case will be described in which this measuring device is an angle measuring device that estimates the incident angle of an incoming signal.

Embodiment 1 FIG.
1 is a block configuration diagram showing an angle measuring device according to Embodiment 1 of the present invention.
In FIG. 1, this angle measuring device includes a plurality of sensors 1a to 1m, a receiving unit 2, an observation data storage unit (observation data storage unit) 3, a data statistical processing unit 4, an angle measurement processing unit (measurement processing unit) 5, An angle measurement result output unit 6 and a data length control unit (data length control means) 7 are provided.

  The plurality of sensors 1a to 1m are arranged side by side. The plurality of sensors 1a to 1m respectively receive signals including a plurality of elementary waves that are transmitted from a signal source (not shown) and propagate through different paths. The receiving part 2 receives the received signal received by each of the plurality of sensors 1a to 1m as an observation data vector in a unit time zone. The observation data storage unit 3 stores the observation data vector of the unit time zone received by the reception unit 2 and outputs the stored observation data vector to the data statistical processing unit 4 in the unit time zone after the next time. .

  The data statistical processing unit 4 receives the incident angle (unknown parameter) of the received signal based on the observation data vector of the unit time zone received by the reception unit 2 and the past observation data vector stored in the observation data storage unit 3. Execute statistical processing to estimate The angle measurement processing unit 5 executes angle measurement processing based on the statistical data statistically obtained by the data statistical processing unit 4. The angle measurement result output unit 6 outputs the angle measurement processing result measured by the angle measurement processing unit 5.

  Based on the result of the angle measurement processing by the angle measurement processing unit 5, the data length control unit 7 propagates different paths from the signal source and receives signal correlation between the plurality of elementary waves received by the plurality of sensors 1a to 1m. The observation data length of the observation data vector input to the angle measurement processing unit 5 is controlled so as to approach zero. Specifically, the data length control unit 7 includes the observation data length of the observation data vector in the unit time zone received by the reception unit 2 and the observation data length of the past observation data vector stored in the observation data storage unit 3. The observation data length of the observation data vector output from the observation data storage unit 3 is controlled so that the total of the two becomes the desired observation data length.

  Here, in the angle measuring device shown in FIG. 1, the observation data length in the unit time zone is not particularly defined. Therefore, the data length control unit 7 can control the observation data length of the observation data vector that is statistically calculated by the data statistical processing unit 4 to be longer or shorter than the observation data length in the unit time zone. Note that the observation data storage unit 3 is not required when the data length control unit 7 controls the observation data length of the observation data vector that is statistics by the data statistical processing unit 4 to be shorter than the observation data length of the unit time zone. It becomes.

  Further, the data length control unit 7 has an area for storing the angle measurement processing result in the angle measurement processing unit 5 in time series. The data length controller 7 controls the observation data length of the observation data vector by arranging the measurement processing results in the unit time zone in time series and detecting the fluctuation as a frequency component. Note that the observation data length of the observation data vector is set using a reciprocal of the detected frequency component or a value that is a natural number multiple of the reciprocal of the detected frequency component.

  Although it has been described that an observation data length of at least 1 / f second is required to separate the Doppler frequency difference of fHz, this is because the signal correlation between a plurality of elementary waves is made zero. If the signal correlation between the elementary waves is zero, each elementary wave can be easily separated. When the observation data length is 1 / f second, the signal correlation is zero. When the observation data length exceeds 1 / f second, the signal correlation increases. When the observation data length is 2 / f second, the signal correlation is zero again. It becomes. Thus, when the observation data length is a natural number multiple of 1 / f seconds, the signal correlation becomes zero, and when the observation data length is other than that, the signal correlation becomes a value larger than zero. Therefore, the data length control unit 7 sets the observation data length with a value that is a natural number multiple of the reciprocal of the detected frequency component.

As described above, according to the first embodiment, the data length control unit propagates different paths from the signal source based on the measurement processing result of the measurement processing unit that estimates the unknown parameter, and uses a plurality of sensors. The observation data length, which is the data length of the received signal input to the measurement processing means, is controlled so that the signal correlation between the plurality of received waves approaches zero. Thus, each elementary wave can be easily separated by controlling the observation data length so that the signal correlation between the plurality of elementary waves approaches zero.
Therefore, by separating each element wave and detecting the accurate element number, a highly accurate measurement result can be obtained, and a measurement apparatus that does not require high processing performance and a wide storage area can be obtained.

In the first embodiment, the data length control unit 7 controls the observation data length of the observation data vector by arranging the measurement processing results in the unit time zone in time series and detecting the fluctuation as a frequency component. explained. However, the present invention is not limited to this, and the data length control unit 7 may control the observation data length of the observation data vector by arranging the observation data vectors in time series and detecting power interference in this time series. The Doppler frequency difference between the elementary waves causes power interference of received data.
Therefore, the observation data length of the observation data vector can be controlled by detecting the frequency component of power interference.

Embodiment 2. FIG.
In the first embodiment, the case where the observation data storage unit 3 stores the observation data vector has been described. However, the present invention is not limited to this, and the observation data storage unit 3 may store the correlation matrix of the observation data vector. . In the second embodiment, a case where the observation data storage unit 3 stores a correlation matrix of observation data vectors instead of the observation data vectors will be described.

FIG. 2 is a block diagram showing an angle measuring device according to Embodiment 2 of the present invention.
In FIG. 2, the angle measuring device includes a correlation matrix calculation unit 8 in addition to the angle measuring device shown in FIG. 1. The angle measuring device includes a correlation matrix storage unit (observation data storage means) 9 and a correlation matrix statistical processing unit 10 instead of the observation data storage unit 3 and the data statistical processing unit 4 shown in FIG. . Other configurations are the same as those in the first embodiment, and a description thereof will be omitted.

  The correlation matrix calculation unit 8 calculates a correlation matrix based on the observation data vector of the unit time zone received by the reception unit 2. The correlation matrix storage unit 9 stores the correlation matrix calculated by the correlation matrix calculation unit 8 and outputs the stored correlation matrix to the correlation matrix statistical processing unit 10 in the unit time zone after the next time.

  The correlation matrix statistical processing unit 10 estimates the incident angle (unknown parameter) of the received signal based on the correlation matrix calculated by the correlation matrix calculation unit 8 and the past correlation matrix stored in the correlation matrix storage unit 9. Perform statistical processing. In general, the angle measurement processing unit 5 can perform angle measurement processing even when a correlation matrix is input instead of the observation data vector.

Embodiment 3 FIG.
FIG. 3 is a block configuration diagram showing an angle measuring device according to Embodiment 3 of the present invention.
In FIG. 3, the angle measuring device includes a signal reproduction processing unit (signal reproduction processing means) 11 and a reproduction result output unit 12 in addition to the angle measuring device shown in FIG. Further, this angle measuring device includes a data length control unit 13 instead of the observation data storage unit 3 shown in FIG. Other configurations are the same as those in the first embodiment, and a description thereof will be omitted.

  The signal reproduction processing unit 11 reproduces a signal in a specific direction based on the angle measurement processing result in the angle measurement processing unit 5. The reproduction result output unit 12 outputs the signal reproduced by the signal reproduction processing unit 11. The data length control unit 13 controls the observation data length of the observation data vector based on the signal reproduced by the signal reproduction processing unit 11.

  Here, the frequency component of the power interference in the time series of the observation data vector described above can also be detected from the signal reproduced by the signal reproduction processing unit 11. Therefore, when the observation data length is not properly controlled and each elementary wave cannot be sufficiently separated, power interference remains in the signal reproduced by the signal reproduction processing unit 11. Therefore, the observation data length of the observation data vector can be controlled by detecting the frequency component of this power interference by the data length control unit 13. Further, by reproducing the signal by the signal reproduction processing unit 11, it is possible to avoid time variation of the received power.

Embodiment 4 FIG.
4 is a block diagram showing an angle measuring device according to Embodiment 4 of the present invention.
4, this angle measuring device includes a correlation matrix calculating unit (correlation matrix calculating means) 8 in addition to the angle measuring device shown in FIG. In addition, the angle measuring device includes a correlation matrix storage unit (observation data storage unit) 9, a correlation matrix statistical process instead of the observation data storage unit 3, the data statistical processing unit 4 and the data length control unit 7 shown in FIG. Section 10 and a data length control section 14. Other configurations are the same as those in the first embodiment, and a description thereof will be omitted.

  The correlation matrix calculation unit 8 calculates a correlation matrix based on the observation data vector of the unit time zone received by the reception unit 2. The correlation matrix storage unit 9 stores the correlation matrix calculated by the correlation matrix calculation unit 8 and outputs the stored correlation matrix to the correlation matrix statistical processing unit 10 in the unit time zone after the next time. The correlation matrix statistical processing unit 10 estimates the incident angle (unknown parameter) of the received signal based on the correlation matrix calculated by the correlation matrix calculation unit 8 and the past correlation matrix stored in the correlation matrix storage unit 9. Statistical processing is executed and a statistical correlation matrix is calculated.

  The data length control unit 14 calculates a difference between the statistical correlation matrix and the correlation matrix stored in the correlation matrix storage unit 9, and controls the observation data length of the observation data vector based on the difference. When the observation data length is in an optimal state, each elementary wave is decorrelated and the temporally varying component in the correlation matrix by the observation data vector is suppressed, so that the difference becomes a small value.

  Therefore, when the difference becomes smaller than a predetermined threshold value, the observation data length can be determined to be an optimal value, and the data length of the correlation matrix in the angle measurement processing unit 5 and the correlation matrix storage unit 9 is optimal. Can be controlled to any value. Note that the data length control unit 14 may test the difference in a time series and control the data length of the correlation matrix by setting the data length at which the difference is a minimum value as an optimum value.

Embodiment 5 FIG.
FIG. 5 is a block configuration diagram showing an angle measuring device according to Embodiment 5 of the present invention.
5, this angle measuring device includes a correlation value calculating unit (correlation value calculating means) 15 in addition to the angle measuring device shown in FIG. Further, this angle measuring device includes a data length control unit 16 instead of the data length control unit 7 shown in FIG. Other configurations are the same as those in the second embodiment, and a description thereof will be omitted.

  The correlation value calculation unit 15 calculates a correlation value between the elementary waves based on the statistical correlation matrix calculated by the correlation matrix statistical processing unit 10 and the angle measurement processing result by the angle measurement processing unit 5. The data length control unit 16 controls the observation data length of the observation data vector based on the correlation value between the elementary waves calculated by the correlation value calculation unit 15. Here, when the observation data length is set such that the number of elementary waves included in the incoming signal can be detected, the optimum observation data length can be determined from the estimated value of the signal correlation.

  Therefore, the data length control unit 16 can determine that the observation data length is the optimum value when the correlation value between the elementary waves is smaller than a predetermined threshold value, and the angle measurement processing unit 5 or the correlation matrix. The data length of the correlation matrix in the storage unit 9 can be controlled to an optimum value. The data length control unit 16 tests the correlation values between the elementary waves in a time series, and controls the data length of the correlation matrix by setting the data length at which the correlation value between the elementary waves is a minimum value as an optimum value. Also good.

Embodiment 6 FIG.
The method of separating each elementary wave by storing the observation data vector is effective when the number of elementary waves included in the incoming signal is stationary. Therefore, when the number of elementary waves changes, the stored observation data vector may not be used. In such a case, it is necessary to execute control such as discarding the stored observation data vector. In the sixth embodiment, a case where the number of elementary waves is changed will be described.

FIG. 6 is a block diagram showing an angle measuring device according to Embodiment 6 of the present invention.
6, in addition to the angle measuring device shown in FIG. 1, this angle measuring device includes a first wave number estimation processing unit (first wave number estimation processing means) 17 and a second wave number estimation processing unit (second wave number estimation processing). Means) 18. In addition, this angle measuring device includes a data length control unit 19 instead of the data length control unit 7 shown in FIG. Other configurations are the same as those in the first embodiment, and a description thereof will be omitted.

  The first wave number estimation processing unit 17 estimates the wave number based on the observation data vector of the unit time zone received by the receiving unit 2. The second wave number estimation processing unit 18 estimates the wave number based on the statistical data statistically obtained by the data statistical processing unit 4. The data length control unit 19 compares the wave numbers estimated by the first wave number estimation processing unit 17 and the second wave number estimation processing unit 18, respectively, and when the wave number estimated by the first wave number estimation processing unit 17 is large, Control is performed so that the observation data length of the observation data vector stored in the observation data storage unit 3 becomes zero.

  This is because there is a possibility that the change in the number of elementary waves included in the incoming signal cannot be accommodated due to the influence of the past observation data vector. In such a case, it is conceivable that the stored observation data vector is discarded and the observation data vector is newly stored. On the other hand, when the wave number estimated by the second wave number estimation processing unit 18 is large, no problem occurs.

  In the first to sixth embodiments, the data length control unit performs time series when the observation data storage unit 3 and the correlation matrix storage unit 9 store past observation data vectors for each unit time zone. The method of deleting from the old observation data vector may be adopted.

  In the first to sixth embodiments, the data length control unit may control the observation data length of the observation data vector by weighting the observation data vector for each unit time zone. Specifically, the data length control unit controls the observation data length or thins out the observation data vector to be stored by setting the weight for the old observation data vector in time series to zero or a value close to zero. The amount of data stored in the observation data storage unit 3 is reduced by giving a weight.

  In the first to sixth embodiments, the data length control unit sets the forgetting coefficient and controls the influence of the old observation data vector, for example, when calculating the correlation matrix by sequential averaging. May be controlled. Here, when shortening the observation data length, a coefficient with a large forgetting degree is set, and when increasing the observation data length, a coefficient with a small forgetting degree is set.

  In the first to sixth embodiments, the R.P. T.A. Williams, S.W. Prasad, A .; K. Mahalanabis, L.M. H. Sibul, “An Improved Spatial Smoothing Technique for Bearing Estimate in a Multipath Environment, Acoustic Transactions on Acoustics, ProSep. 36, no. 4, pp. 425-432, Apr 1988 (hereinafter referred to as “Non-Patent Document 2”), a pre-processing unit that executes the spatial average preprocessing or the improved spatial average preprocessing may be inserted.

  This process can be performed not only by averaging the subsets in the correlation matrix but also by concatenating the subsets in the observed data vector as time series data. In addition, the process of averaging the subsets in the correlation matrix is applied only to the estimation means that uses the correlation matrix in the angle measurement processing unit 5, but the process of concatenating the subsets in the observation data vector as time-series data is performed. Any estimation means can be applied to the corner processing unit 5.

  Moreover, in the said Embodiment 1-6, the angle measurement process part 5 which performs an angle measurement process based on the statistical data statistics by the data statistical process part 4 is the maximum likelihood method described in the said nonpatent literature 1, Capon method, MUSIC (Multiple Signal Classification) method, ESPRIT (Estimation of Signal Parameters via Rotational Invariance techniques), MODE (method Of Direction Estimation), SAGE (Space-Alternating Generalized Expectation-maximization algorithm) and VESPA (Virtual ESPRIT Algorith An estimation method based on any of m) may be used. These estimation methods are known to have a high resolution, and are advantageous when the incident angles are close as is a problem in the present invention.

  In the first to sixth embodiments, the angle measurement processing unit 5 performs the blind signal separation processing based on one of ESPRIT, VESPA, and independent component analysis as preprocessing, and then performs measurement based on the separated signal. Corner processing may be performed. These blind signal separation processes are known to have extremely high signal resolution, but as an application condition, the signals are required to be uncorrelated. Therefore, the high resolution performance of the blind signal separation process can be effectively utilized by optimally controlling the observation data length by the data length control unit and making each elementary wave uncorrelated. Further, the high resolution performance in the estimation methods such as the maximum likelihood method, the Capon method, the MUSIC method, ESPRIT, MODE, SAGE, and VESPA is limited to the aperture lengths of the plurality of sensors 1a to 1m. The high resolution performance is not limited by the opening length.

  Further, even if each elementary wave cannot be separated by executing blind signal separation processing based on any one of ESPRIT, VESPA, and independent component analysis, the output information of blind signal separation processing is described in Non-Patent Document 2. In some cases, the individual waves can be separated by executing the spatial average preprocessing or the improved spatial average preprocessing.

  In addition, the angle measurement processing after the blind signal separation processing based on any one of ESPRIT, VESPA, and independent component analysis in the angle measurement processing unit 5 is based on a monopulse estimation method or DBF (Digital Beam Forming) with low resolution. Processing may be processing, and processing based on any of the maximum likelihood method, Capon method, MUSIC method, ESPRIT, MODE, SAGE, and VESPA, which is a high-resolution processing method. This is because the signal is decomposed in the blind signal separation process, and the angle measurement process does not require high resolution performance. However, when there is a possibility that the individual waves cannot be separated by executing the blind signal separation process, it is desirable to select a processing method with high resolution performance.

  In the first to sixth embodiments, the angle measuring device has been described as an example. However, the present invention can be applied to the case where the distance measuring device is affected by the Doppler frequency difference between the elementary waves. The same effect can be obtained. That is, since the coping method for the influence due to the Doppler frequency difference between the elementary waves is the same in both the angle measuring device and the distance measuring device, the present invention can also be applied to the distance measuring device.

  1a to 1m sensor, 3 observation data storage unit (observation data storage unit), 5 angle measurement processing unit (measurement processing unit), 7, 13, 14, 16, 19 data length control unit (data length control unit), 8 correlation Matrix calculation unit (correlation matrix calculation unit), 9 correlation matrix storage unit (observation data storage unit), 11 signal reproduction processing unit (signal reproduction processing unit), 15 correlation value calculation unit (correlation value calculation unit), 17 first wave number An estimation processing unit (first wave number estimation processing means), 18 a second wave number estimation processing unit (second wave number estimation processing means).

Claims (19)

A plurality of sensors each receiving a signal including a plurality of elementary waves transmitted from a signal source and propagating through different paths;
Measurement processing means for estimating an unknown parameter including at least one of an incident angle of the reception signal and a distance to the signal source based on reception signals received by the plurality of sensors;
Based on the measurement processing result in the measurement processing means, the observation data length which is the data length of the received signal input to the measurement processing means is controlled so that the signal correlation between the plurality of elementary waves approaches zero. Data length control means;
A measuring apparatus comprising:   In addition to storing observation data of unit time periods received by the plurality of sensors, the apparatus further comprises observation data storage means for outputting the observation data stored in the unit time period after the next time to the measurement processing means. The measuring apparatus according to claim 1.   The measurement apparatus according to claim 2, wherein the observation data storage unit stores a correlation matrix of the observation data. The data length control means controls the observation data length with a data length longer than the data length of the observation data of the unit time zone received by the plurality of sensors,
The measurement apparatus according to claim 2, wherein the measurement processing unit estimates the unknown parameter according to the observation data length. The data length control means controls the observation data length with a data length shorter than the data length of the observation data in the unit time zone received by the plurality of sensors,
The measurement apparatus according to claim 1, wherein the measurement processing unit estimates the unknown parameter according to the observation data length.   The data length control means has an area for storing the measurement processing result in time series, and controls the observation data length based on a variation in time series of the measurement processing result. The measuring apparatus according to any one of claims 1 to 5.   The said data length control means controls the said observation data length based on the power interference in the time series of the observation data of the unit time slot received by these sensors. The measuring apparatus according to any one of the above. Further comprising signal reproduction processing means for performing signal reproduction processing based on the measurement processing result,
The said data length control means controls the said observation data length based on the power interference of the signal reproduced | regenerated by the said signal reproduction | regeneration processing means, The any one of Claim 1-5 characterized by the above-mentioned. The measuring device described. Correlation matrix calculation means for calculating a correlation matrix based on observation data of unit time zones received by the plurality of sensors,
The observation data storage means stores the correlation matrix calculated by the correlation matrix calculation means,
The data length control means calculates a difference between the correlation matrix calculated by the correlation matrix calculation means and the correlation matrix stored in the observation data storage means, and tests in time series, under the condition that the difference is minimal. The measurement apparatus according to claim 2, wherein the observation data length is controlled. A correlation value calculating means for calculating a correlation value between the plurality of elementary waves;
6. The data length control unit according to claim 1, wherein the data length control unit tests the correlation value in time series and controls the observation data length under a condition that the correlation value is minimized. The measuring device according to item. First wave number estimation processing means for estimating the wave numbers of the plurality of elementary waves based on observation data of unit time zones received by the plurality of sensors;
Second wave number estimation processing means for estimating the wave numbers of the plurality of elementary waves based on observation data whose observation data length is controlled,
The data length control means controls the observation data length based on a comparison result obtained by comparing the wave number estimated by the first wave number estimation processing means and the wave number estimated by the second wave number estimation processing means. The measuring apparatus according to any one of claims 1 to 5, wherein:   The said data length control means deletes the observation data preserve | saved at the said observation data preservation | save means from the old observation data in time series, The one of Claim 2 to 11 characterized by the above-mentioned. measuring device.   The data length control unit controls the observation data length by weighting observation data received by the plurality of sensors for each unit time zone. The measuring device according to claim 1.   The measurement apparatus according to any one of claims 1 to 11, wherein the data length control unit controls the observation data length by setting a forgetting factor in a sequential averaging process.   The measuring apparatus according to any one of claims 1 to 14, wherein a pre-processing unit for executing a spatial average pre-processing or an improved spatial average pre-processing is inserted in a pre-stage of the measurement processing unit. .   2. The measurement processing means estimates the unknown parameter by an estimation method based on any one of a maximum likelihood method, a Capon method, a MUSIC method, ESPRIT, MODE, SAGE, and VESPA. The measuring device according to claim 1.   The measurement processing means performs a blind signal separation process based on any one of ESPRIT, VESPA, and independent component analysis as preprocessing. The measuring device according to item.   The measurement apparatus according to claim 17, wherein the measurement processing unit performs spatial average preprocessing or improved spatial average preprocessing on the signal separated by the blind signal separation processing.   The measurement processing means is any one of a monopulse estimation method, DBF, maximum likelihood method, Capon method, MUSIC method, ESPRIT, MODE, SAGE, and VESPA with respect to the signal separated by the blind signal separation processing. The measurement apparatus according to claim 17, wherein an estimation method based on the method is applied.

Images