JP7357217B2 - Estimation method, estimation device, and program - Google Patents

Description

The present disclosure relates to an estimation method, an estimation device, and a program using radio signals.

There is a technique for detecting a target using a wireless signal (see, for example, Patent Documents 1 and 2). For example, Patent Document 1 discloses a technique for determining an autocorrelation matrix of a received signal and determining the number of targets from the magnitude of the eigenvalue of the autocorrelation matrix. Further, for example, Patent Document 2 discloses a technique for calculating the number of targets by decomposing a correlation matrix of a received signal into eigenvalues and counting the number of eigenvalues that are equal to or larger than a threshold value.

Further, for example, Patent Document 3 discloses a technique for determining the position or state of a person to be detected by analyzing components including Doppler shift using Fourier transform.

Japanese Patent Application Publication No. 2009-281775 Japanese Patent Application Publication No. 2000-171550 Japanese Patent Application Publication No. 2015-117972

However, since the techniques disclosed in Patent Documents 1 and 2 are device position detection techniques, there is a problem that the position of a living body cannot be estimated. Furthermore, in the technique disclosed in Patent Document 3, instantaneous positioning errors may occur due to external noise, etc., and there is a problem in that it is not possible to know how reliable the positioning results are.

The present disclosure has been made in view of the above-mentioned circumstances, and includes an estimation method, an estimation device, and an estimation method capable of evaluating the reliability of estimation results such as the position of a living body existing in a target space using wireless signals. The purpose is to provide programs.

In order to achieve the above object, an estimation method, etc. according to an embodiment of the present disclosure includes a transmitter having at least one transmitting antenna element and a receiver having N receiving antenna elements (N is a natural number of 2 or more). An estimation method of an estimation device for estimating the position or direction of a living body existing in a target space using A transfer function calculation step of calculating a plurality of complex transfer functions representing propagation characteristics between each of the N receiving antenna elements; an extraction step of extracting a biological component in each of the N receiving antenna elements; and calculating a correlation matrix from the biological component in each of the N receiving antenna elements extracted in the extraction step. a calculation step of calculating one or more eigenvalues of the correlation matrix calculated in the correlation matrix calculation step, and biological number information indicating a value of the number of living organisms existing in the target space; a reliability estimation step of estimating the reliability of the estimation result when estimating the position or direction of the living body using the one or more eigenvalues calculated in the calculation step; and an estimating step of estimating the position or direction of the living body from the correlation matrix using a predetermined method.

Note that these general or specific aspects may be realized by an apparatus, a method, an integrated circuit, a computer program, or a computer-readable recording medium such as a CD-ROM. It may be realized by any combination of programs and recording media.

According to the estimation method and the like of the present disclosure, it is possible to evaluate the reliability of estimation results such as the position of a living body existing in a target space using a wireless signal.

FIG. 1 is a block diagram showing an example of the configuration of an estimation device according to the first embodiment. FIG. 2 is a conceptual diagram schematically showing the propagation of transmitted waves in the estimation device shown in FIG. 1. FIG. 3 is a diagram showing an example of the eigenvalue distribution in the first embodiment. FIG. 4 is a diagram showing an example of ratio information calculated from the eigenvalue distribution in the first embodiment. FIG. 5 is a block diagram illustrating an example of the configuration of the reliability estimation section in the first embodiment. FIG. 6 is a diagram showing another example of the eigenvalue distribution in the first embodiment. FIG. 7 is a diagram showing another example of ratio information calculated from the eigenvalue distribution in the first embodiment. FIG. 8 is a flowchart illustrating an example of the operation of the estimation device in the first embodiment. FIG. 9 is a flowchart showing another example of the operation of the estimation device in the first embodiment. FIG. 10 is a flowchart showing an example of detailed processing of step S6 in the first embodiment. FIG. 11 is a diagram illustrating an example of a positioning error distribution when using a multicarrier signal in a modification of the first embodiment. FIG. 12A is a distribution diagram showing whether the ratio of the first eigenvalue and the second eigenvalue at the same timing as the positioning error distribution shown in FIG. 11 is equal to or greater than a threshold value. FIG. 12B is a distribution diagram showing whether the distribution of the ratio of the second eigenvalue and the third eigenvalue at the same timing as the positioning error distribution shown in FIG. 11 is equal to or larger than the threshold value. FIG. 13 is a block diagram showing an example of the configuration of an estimation device according to the second embodiment. FIG. 14 is a block diagram illustrating an example of the configuration of an estimation device in Embodiment 3. FIG. 15 is a diagram showing an example of a graph of an evaluation function in the third embodiment.

(Findings that formed the basis of this disclosure)
The inventors conducted a detailed study on conventional techniques related to estimating the position of a living body using radio signals. The results have been described above, but since the techniques disclosed in Patent Documents 1 and 2 are device position detection techniques, there was a problem in that the position of a living body could not be estimated. Furthermore, in the technique disclosed in Patent Document 3, instantaneous positioning errors may occur due to external noise, etc., and there is a problem in that it is not possible to know how reliable the positioning results are.

As a result of extensive research into the above-mentioned problems, the inventors evaluated the reliability of estimation results such as the position of a living body existing in a target space by using radio signals received by a receiver. This discovery led to the present disclosure. That is, the inventors calculate a complex transfer function from a radio signal received by a receiver, and use an evaluation function calculated using a correlation matrix obtained from the calculated complex transfer function or an eigenvalue of the correlation matrix. We found that it is possible to evaluate the reliability of measurement results.

More specifically, in order to achieve the above object, an estimation method according to an embodiment of the present disclosure includes a transmitter having at least one transmitting antenna element and N receiving antenna elements (N is a natural number of 2 or more). An estimating method of an estimating device that estimates the position or direction of a living body existing in a target space using a receiver having: a transfer function calculating step of calculating a plurality of complex transfer functions representing propagation characteristics between the transmitting antenna element and each of the N receiving antenna elements, and a plurality of complex transfer functions calculated in the transfer function calculating step; , an extraction step of extracting a biological component due to the influence of a living body in each of the N receiving antenna elements; and a biological component in each of the N receiving antenna elements extracted in the extraction step. a correlation matrix calculation step of calculating a correlation matrix from the correlation matrix, a calculation step of calculating one or more eigenvalues of the correlation matrix calculated in the correlation matrix calculation step, and a calculation step of calculating a value of the number of living organisms existing in the target space. a reliability estimation step of estimating the reliability of an estimation result when estimating the position or direction of the living body using the number of living organisms information and the one or more eigenvalues calculated in the calculation step; and the estimation result. and an estimation step of estimating the position or direction of the living body using a predetermined method from the correlation matrix, depending on the reliability of the correlation matrix.

As a result, it is possible not only to estimate the position of a living body existing in the target space using wireless signals, but also to estimate the position etc. of a living body existing in the target space using wireless signals. It is possible to evaluate the reliability of

Here, for example, in the reliability estimation step, (1) calculate ratio information indicating the ratio or difference between adjacent eigenvalues when the one or more eigenvalues calculated in the calculation step are sorted in descending order of values; (2) When the value indicated by the biological number information is L (L is a natural number of 1 or more), and the ratio information corresponding to the L-th eigenvalue when sorted is a predetermined value or more, It is determined that the estimation result is highly reliable.

Thereby, the reliability of the estimation result can be evaluated using the eigenvalue of the correlation matrix.

Furthermore, in order to achieve the above object, an estimation method according to an embodiment of the present disclosure includes a transmitter having at least one transmitting antenna element and a receiver having N receiving antenna elements (N is a natural number of 2 or more). An estimation method of an estimation device for estimating the direction or position of a living body existing in a target space using a transfer function calculation step of calculating a plurality of complex transfer functions representing the propagation characteristics between the antenna and each of the N receiving antenna elements; an extraction step of extracting a biological component in each of the N receiving antenna elements; and a correlation matrix from the biological component in each of the N receiving antenna elements extracted in the extraction step. a step of calculating a correlation matrix; an estimation step of calculating an evaluation function using the correlation matrix; and an estimation step of estimating the direction or position of the living body using the calculated evaluation function; and the evaluation calculated in the estimation step. and a reliability estimation step of estimating the reliability of the estimation result when estimating the position or direction of the living body using the function.

Thereby, the reliability of the estimation result can be evaluated using the evaluation function calculated using the correlation matrix.

Here, for example, the evaluation function calculated in the estimation step is a MUSIC spectrum.

Further, for example, the evaluation function calculated in the estimation step is a spectrum of the Capon method.

Further, for example, the evaluation function calculated in the estimation step is a spectrum of a beamformer method.

Further, for example, in the reliability estimation step, if the estimated reliability is lower than a threshold value, it is determined that the reliability is low, and if the estimated reliability is greater than or equal to the threshold value, it is determined that the reliability is high. It may be determined.

For example, in the estimation step, if the reliability is determined to be low in the reliability estimation step, the estimation result of estimating the direction or position of the living body may be the point in time when the reliability is determined to be low. Among the estimation results for a period past the above, the estimation result at a past point in time determined to have high reliability in the reliability estimation step may be output.

This allows highly reliable estimation results to be used.

For example, in the estimation step, if the reliability is determined to be low in the reliability estimation step, the frequency that is the estimation result of estimating the direction or position of the living body and for which the reliability is determined to be low. Among the estimation results for frequency ranges other than the above, the estimation results for frequencies determined to have high reliability in the reliability estimation step may be output.

This allows highly reliable estimation results to be used.

Further, for example, in the estimation step, if the reliability is determined to be low in the reliability estimation step, the estimation result of estimating the direction or position of the living body, which is a result of estimating the direction or position of the living body, and which is the result of estimating the direction or position of the living body, and which is the result of estimating the direction or position of the living body, and which is the result of estimating the direction or position of the living body, which Among the estimation results, an estimation result at a past time and frequency determined to have high reliability in the reliability estimation step may be output.

This allows highly reliable estimation results to be used.

For example, if the reliability is determined to be low in the reliability estimation step, and the estimation result estimated in the estimation step is separated by a time or frequency equal to or more than a threshold value from the previous estimation result, the reliability Among the estimation results for a period past the time when the reliability was determined to be low, the estimation results for the past time when the reliability was determined to be high may be output.

This allows highly reliable estimation results to be used.

Further, for example, the method further includes a storing step of storing estimation results estimated in the past in the estimation step in a storage unit, and in the estimation step, when the reliability is determined to be low in the reliability estimation step, Calculate the average or median value of the past estimation results stored in the storage unit in the time range of the first interval, which is set longer than the reference interval according to the degree of reliability, and use it as the current estimation result. It may also be output.

This allows highly reliable estimation results to be used.

For example, the method further includes a storing step of storing estimation results estimated in the past in the estimation step in a storage unit, and when it is determined that the reliability is low in the reliability estimation step, the degree of reliability is The average or median value of the past estimation results stored in the storage unit may be calculated in the frequency range of the second interval, which is set longer than the reference interval according to the above, and output as the current estimation result. .

This allows highly reliable estimation results to be used.

Further, an estimation device according to an aspect of the present disclosure uses a transmitter having at least one transmitting antenna element and a receiver having N receiving antenna elements (N is a natural number of 2 or more) to obtain information in a target space. An estimation device for estimating the position or direction of an existing living body, the estimation device estimating the position or direction of an existing living body, the estimation device estimating the position or direction of the transmitting antenna element and each of the N receiving antenna elements from a received signal received for a predetermined period by each of the N receiving antenna elements. a transfer function calculation unit that calculates a plurality of complex transfer functions representing propagation characteristics between an extraction unit that extracts a biological component in each of the antenna elements; a correlation matrix calculation unit that calculates a correlation matrix from the biological component in each of the N receiving antenna elements extracted in the extraction unit; and the correlation matrix. a calculation unit that calculates one or more eigenvalues of the correlation matrix calculated in the calculation unit; living body count information indicating a value of the number of living organisms existing in the target space; and one or more of the one or more eigenvalues calculated in the calculation unit. a reliability estimating unit that estimates the reliability of the estimation result when estimating the position or direction of the living body using the eigenvalue of the correlation matrix; and an estimation unit that estimates the position or direction of the living body.

Further, a program according to an aspect of the present disclosure uses a transmitter having at least one transmitting antenna element and a receiver having N receiving antenna elements (N is a natural number of 2 or more) to exist in a target space. A program for causing a computer to execute an estimation method of an estimation device for estimating the position or direction of a living body, the program comprising a transfer function calculation step of calculating a plurality of complex transfer functions representing propagation characteristics between the two receiving antenna elements; and a biological component due to the influence of the living body from the plurality of complex transfer functions calculated in the transfer function calculation step. an extraction step of extracting biological components in each of the N receiving antenna elements; and a correlation calculating a correlation matrix from the biological components in each of the N receiving antenna elements extracted in the extraction step. a matrix calculation step, a calculation step of calculating one or more eigenvalues of the correlation matrix calculated in the correlation matrix calculation step, biological number information indicating a value of the number of living organisms existing in the target space, and the calculation a reliability estimation step of estimating the reliability of the estimation result when estimating the position or direction of the living body using the one or more eigenvalues calculated in the step; and according to the reliability of the estimation result, and an estimating step of estimating the position or direction of the living body from the correlation matrix using a predetermined method.

Note that the present disclosure can be realized not only as a device, but also as an integrated circuit including processing means included in such a device, or as a method in which the processing means constituting the device is a step, or as a method in which the processing means constituting the device is a step. It can also be realized as a program for a computer to execute, or as information, data, or signals indicating the program. These programs, information, data, and signals may be distributed via a recording medium such as a CD-ROM or a communication medium such as the Internet.

Embodiments of the present disclosure will be described in detail below with reference to the drawings. Note that each of the embodiments described below represents a preferred specific example of the present disclosure. The numerical values, shapes, materials, components, arrangement positions and connection forms of the components, steps, order of steps, etc. shown in the following embodiments are examples, and do not limit the present disclosure. Furthermore, among the constituent elements in the embodiments below, constituent elements that are not described in the independent claims representing the most important concept of the present disclosure will be described as arbitrary constituent elements constituting a more preferable embodiment. Note that, in this specification and the drawings, components having substantially the same functional configurations are designated by the same reference numerals and redundant explanation will be omitted.

(Embodiment 1)
In the following, referring to the drawings, we will observe the correlation matrix of the received signal using a transmitter and receiver with a SIMO (Single Input Multiple Output) configuration, and use the eigenvalues of the correlation matrix to calculate We will explain how to estimate the direction of a living body that exists in a body.

[Configuration of estimation device 1]
FIG. 1 is a block diagram showing an example of the configuration of an estimation device 1 in the first embodiment. FIG. 1 also shows a living body 50 that is a detection target of the estimation device 1 shown in FIG.

The estimation device 1 in the first embodiment exists in a target space using a transmitter 10 having one transmitting antenna element and a receiver 11 having N receiving antenna elements (N is a natural number of 2 or more). The direction of one or more living organisms 50 is estimated.

As shown in FIG. 1, the estimation device 1 in the first embodiment includes a transmitter 10, a receiver 11, a biological information calculation section 12, a calculation section 13, a storage section 14, and a reliability estimation section 15. , and a direction estimation unit 16.

[Transmitter 10]
The transmitter 10 includes a transmitting antenna section 101 and a transmitting section 102.

Transmission antenna section 101 has one transmission antenna element.

The transmitter 102 generates a high frequency signal. The transmitting unit 102 transmits a transmission signal, which is a generated high-frequency signal, from one transmitting antenna element included in the transmitting antenna unit 101. In this embodiment, the transmitter 102 will be described as transmitting, for example, a 2.45 GHz sine wave, but the transmitter 102 is not limited to this. Transmitting section 102 may transmit a transmission signal of a different frequency, or may transmit a transmission signal generated using a different modulation method.

[Receiver 11]
The receiver 11 includes a receiving antenna section 111 and a receiving section 112.

The receiving antenna section 111 has N receiving antenna elements. In the present embodiment, the reception antenna section 111 will be described below as having MR antenna elements #1 to _#MR ( _MR is an integer of 2 or more) _, as shown in FIG. 1, for example. In the receiving antenna section 111, each of the M _R receiving antenna elements receives a received signal including a signal transmitted from one transmitting antenna element and a signal reflected by the living body 50 when the living body 50 is present. do.

The receiving unit 112 observes (that is, receives) the received signal for a predetermined period of time using each of the N receiving antenna elements. More specifically, as shown in FIG. 1, the receiving unit 112 converts high frequency signals received for a predetermined period by each of the _MR receiving antenna elements into a low frequency signal that can be processed. The receiving unit 112 transmits the converted low frequency signal to the biological information calculating unit 12.

[Biological information calculation unit 12]
As shown in FIG. 1, the biological information calculation section 12 includes a complex transfer function calculation section 121, an extraction section 122, and a correlation matrix calculation section 123, and calculates biological information from the signal transmitted by the receiver 11. .

<Complex transfer function calculation unit 121>
The complex transfer function calculating unit 121 calculates a complex transfer function representing the propagation characteristics between the transmitting antenna element and each of the N receiving antenna elements from the received signal received for a predetermined period by each of the N receiving antenna elements. Perform multiple calculations. More specifically, the complex transfer function calculation unit 121 expresses the propagation characteristics between one transmitting antenna element and each of the M _R receiving antenna elements from the low frequency signal transmitted from the receiver 11. Calculate the complex transfer function. A more specific explanation will be given below using FIG. 2.

FIG. 2 is a conceptual diagram schematically representing the propagation of transmitted waves in the estimation device 1 shown in FIG. 1. As shown in FIG. 2, a part of the transmission wave transmitted from the transmission antenna element of the transmission antenna section 101 is reflected by the living body 50 and reaches the reception array antenna of the reception antenna section 111. Here, the reception antenna section 111 is a reception array antenna consisting of _MR reception antenna elements, and is a linear array with an element interval d. Further, the direction of the living body 50 viewed from the front of the receiving antenna section 111 is assumed to be θ. It is assumed that the distance between the living body 50 and the receiving antenna section 111 is sufficiently large, and the reflected wave originating from the living body that arrives at the receiving antenna section 111 can be regarded as a plane wave.

Then, the complex transfer function calculation unit 121 calculates a complex reception signal vector representing the propagation characteristics between the transmission antenna element and each of the _{MR reception antenna elements from the complex reception signal vector observed using the MR reception} antenna elements. A transfer function vector can be calculated.

<Extraction unit 122>
The extracting unit 122 extracts the biological components in each of the N receiving antenna elements, which are the biological components due to the influence of the living body 50, from the plurality of complex transfer functions calculated by the complex transfer function calculating unit 121. More specifically, the extraction unit 122 records the complex transfer function calculated by the complex transfer function calculation unit 121 in time series, and extracts the biological component due to the influence of the living body 50 from among the changes in the complex transfer function. Here, as a method for extracting the biological component due to the influence of the living body 50, there is a method of extracting only the component corresponding to the vibration of the living body after conversion to the frequency domain such as Fourier transform, or a method of extracting only the component corresponding to the vibration of the living body after conversion to the frequency domain such as Fourier transform, or A method of calculating a difference, etc. can be used. By executing these methods, the complex transfer function that passes through a fixed object other than the living body 50 is removed, and only the complex transfer function component that passes through the living body 50 remains. Note that since there are a plurality of receiving antenna elements, the number of components of the complex transfer function corresponding to the receiving antenna elements passing through the living body 50 is also plural. These are collectively defined as a biological component channel vector.

<Correlation matrix calculation unit 123>
The correlation matrix calculation unit 123 calculates a correlation matrix from the biological components in each of the N reception antenna elements extracted by the extraction unit 122. More specifically, the correlation matrix calculating unit 123 calculates a correlation matrix R _i of a biological component channel vector composed of a plurality of complex transfer function components passing through the living body 50 which is a biological component.

[Calculation section 13]
The calculation unit 13 calculates one or more eigenvalues of the correlation matrix calculated by the correlation matrix calculation unit 123. More specifically, the calculation unit 13 calculates the eigenvalue of the correlation matrix R _i calculated by the correlation matrix calculation unit 123. The calculation unit 13 causes the storage unit 14 to store the calculated eigenvalues of the correlation matrix R _i .

The calculation unit 13 calculates the eigenvalues by, for example, decomposing the correlation matrix R _i into eigenvalues. Generally, each of the eigenvalue and the eigenvector represents a propagation path of radio waves (transmission waves) from the transmission antenna section 101 to the reception antenna section 111, that is, one path. Normally, there are various paths such as direct waves or reflections from fixed objects such as walls, and each path corresponds to each eigenvalue and eigenvector. However, in the present embodiment, since the correlation matrix R _i is calculated from the biological components of each of the N receiving antenna elements by the correlation matrix calculation unit 123, each eigenvalue calculated by the calculation unit 13 is 50 and the path corresponding to the noise.

[Storage unit 14]
The storage unit 14 is composed of an HDD (Hard Disk Drive), a memory, or the like, and stores data and the like used in the processing of the estimation device 1.

For example, the storage unit 14 stores calculation results of the calculation unit 13, data used for estimation processing by the reliability estimation unit 15, and the like. Furthermore, the storage unit 14 stores estimation results estimated in the past by the direction estimation unit 16.

[Reliability estimation unit 15]
When the reliability estimating unit 15 estimates the direction of the living body 50 using living body count information indicating the value of the number of living bodies 50 existing in the target space and one or more eigenvalues calculated by the calculation unit 13, Estimate the reliability of the estimation results. Here, the living body number information is information given in advance, and indicates information that L living bodies 50 are present in the target space that is the measurement target range.

More specifically, the reliability estimation unit 15 calculates ratio information indicating the ratio or difference between adjacent eigenvalues when the plurality of eigenvalues calculated by the calculation unit 13 are sorted by value size. Then, when the value indicated by the biological number information is L (L is a natural number of 1 or more), the reliability estimation unit 15 estimates The results are determined to be highly reliable.

In other words, the reliability estimation unit 15 evaluates the reliability of the estimation result performed by the direction estimation unit 16 from the characteristics of the eigenvalue distribution obtained by sorting the plurality of eigenvalues calculated by the calculation unit 13 by value size. do.

FIG. 3 is a diagram showing an example of the eigenvalue distribution in the first embodiment.

The eigenvalue distribution is a distribution obtained when a plurality of eigenvalues calculated by the calculation unit 13 are sorted by size. The eigenvalue distribution may be a table in which a plurality of eigenvalues are sorted and arranged row by row, as shown in FIG. In the example shown in FIG. 3, four eigenvalues of 20, 18, 0.15, and 0.12 are sorted and arranged in the first row of the eigenvalue distribution table. More specifically, in the first row of the table shown in FIG. 3, the largest eigenvalue 20 is placed in the first eigenvalue column, the second largest eigenvalue 18 is placed in the second eigenvalue column, and the third eigenvalue The third largest eigenvalue, 0.15, is placed in the column , and the fourth largest eigenvalue, 0.12, is placed in the fourth eigenvalue column. Note that the second and third rows of the table shown in FIG. 3 are also the same, so their explanation will be omitted.

FIG. 4 is a diagram showing an example of ratio information calculated from the eigenvalue distribution in the first embodiment. FIG. 4 also shows the reliability of the estimation results determined from the ratio information.

Ratio information is an example of a feature of the eigenvalue distribution, and is obtained by calculating the ratio or difference between adjacent eigenvalues in the eigenvalue distribution. As shown in FIG. 4, the ratio information may be a table showing the results of calculating the ratios of adjacent eigenvalues in the eigenvalue distribution for each row shown in FIG. More specifically, in the first row of the table shown in FIG. 4, the ratio 1.11 when the first eigenvalue is divided by the second eigenvalue in the first eigenvalue distribution shown in FIG. A ratio of 120 when divided by the third eigenvalue and a ratio of 1.25 when the third eigenvalue is divided by the fourth eigenvalue are arranged.

The reliability of the estimation result is determined by whether, when the value indicated by the biological number information is L, the value indicated by the ratio information of the L-th eigenvalue counted from the largest eigenvalue is equal to or greater than a predetermined value. Ru. In the example shown in Figure 4, the value indicated by the number of living organisms information is 2 (two people), so the ratio information for the second eigenvalue counting from the largest eigenvalue is obtained by dividing the second eigenvalue by the third eigenvalue. The determination is made based on whether the ratio is equal to or greater than a predetermined value, such as 10, for example. In the first row of the table shown in FIG. 4, the ratio when the second eigenvalue is divided by the third eigenvalue is 120, which is larger than the predetermined value, so it is determined that the reliability of the estimation result is high. Note that the second and third rows of the table shown in FIG. 4 are also the same, so their explanation will be omitted.

Note that the eigenvalue distribution is not limited to being expressed as a table; for example, it can be expressed as a graph in which the vertical axis is the size of the eigenvalue and the horizontal axis is a number indicating the order counted from the largest eigenvalue when sorted. It's okay. Similarly, the ratio information of the eigenvalue distribution is not limited to being expressed as a table; for example, it can be expressed as a graph in which the vertical axis is the ratio of eigenvalues and the horizontal axis is a number indicating the order counted from the largest eigenvalue when sorted. may be expressed.

FIG. 5 is a block diagram showing an example of a detailed configuration of the reliability estimating unit 15 in the first embodiment.

In this embodiment, the reliability estimation section 15 includes an eigenvalue distribution calculation section 151, a feature determination section 152, and a reliability determination section 153, as shown in FIG.

<Eigenvalue distribution calculation unit 151>
The eigenvalue distribution calculation unit 151 calculates an eigenvalue distribution obtained when the plurality of eigenvalues calculated by the calculation unit 13 are sorted by size. For example, the eigenvalue distribution calculation unit 151 may calculate an eigenvalue distribution as shown in FIG.

Here, FIG. 6 is a diagram showing another example of the eigenvalue distribution in the first embodiment. The eigenvalue distribution shown in FIG. 6 is calculated by the eigenvalue distribution calculation unit 151 shown in FIG. In Figure 6, the vertical axis indicates the size of the eigenvalue, and the horizontal axis indicates the number of eigenvalues counted from the largest eigenvalue when multiple eigenvalues are sorted by size, with the largest eigenvalue being the first. . Note that FIG. 6 shows an example of a graph when M eigenvalues are sorted by size, and values corresponding to noise such as thermal noise are further shown by dotted lines.

Note that the eigenvalue distribution calculation unit 151 may calculate ratio information of the eigenvalue distribution, which will be described later, as the eigenvalue distribution.

<Feature determination unit 152>
The feature determining unit 152 determines the characteristics of the eigenvalue distribution calculated by the eigenvalue distribution calculating unit 151. More specifically, the feature determination unit 152 calculates ratio information indicating the ratio or difference between adjacent eigenvalues in the eigenvalue distribution calculated by the eigenvalue distribution calculation unit 151 as a feature of the eigenvalue distribution. For example, the feature determination unit 152 may calculate ratio information of the eigenvalue distribution as shown in FIG. 7 as the feature of the eigenvalue distribution shown in FIG.

FIG. 7 is a diagram showing another example of ratio information calculated from the eigenvalue distribution in the first embodiment. The graph of ratio information of the eigenvalue distribution shown in FIG. 7 is calculated by the feature determination unit 152 shown in FIG. In Figure 7, the vertical axis shows the size of the eigenvalue ratio, and the horizontal axis shows the eigenvalue ratio when multiple eigenvalues are sorted by size and the ratio between the largest eigenvalue and the next largest eigenvalue is taken as the first. Show the number. That is, the ratio information of the eigenvalue distribution shown in FIG. 7 shows the distribution of the ratios of adjacent eigenvalues when the plurality of eigenvalues calculated by the calculation unit 13 are sorted by size. Here, the ratio is calculated by dividing the eigenvalue λ _i corresponding to the eigenvalue ratio number i (i is a natural number) as the numerator and the eigenvalue λ _i+1 adjacent to the corresponding eigenvalue λ _i as the denominator, that is, λ _i /λ _i+1 (i is natural numbers).

Note that the feature determination unit 152 is not limited to determining the ratio information of the eigenvalue distribution as a feature of the eigenvalue distribution calculated by the eigenvalue distribution calculation unit 151, and may determine the amount of change in the eigenvalue in the eigenvalue distribution.

<Reliability determination unit 153>
The reliability determining unit 153 determines the reliability of the estimation result using the characteristics of the eigenvalue distribution determined by the feature determining unit 152. More specifically, when the value indicated by the living body count information is L, the reliability determination unit 153 determines whether the value indicated by the ratio information of the L-th eigenvalue counted from the largest eigenvalue is greater than or equal to a predetermined value. The reliability of the estimation result is determined based on whether or not the estimation result is true.

For example, when the value indicated by the biological number information is L, the feature determination unit 152 calculates the amount of change in the Lth and L+1st eigenvalues in the eigenvalue distribution, that is, the ratio of the eigenvalues, as ratio information for the Lth eigenvalue. . Then, the feature determining unit 152 determines whether the reliability of the estimation result is high by determining whether the eigenvalue ratio of the calculated L-th eigenvalue is greater than or equal to a predetermined value.

In the example shown in FIG. 7, since the value indicated by the biological number information is L, the feature determination unit 152 determines whether the eigenvalue ratio of the L-th eigenvalue surrounded by a solid line circle is equal to or greater than a predetermined value. In the example shown in FIG. 7, as can be seen from the fact that the eigenvalue ratio of the Lth eigenvalue surrounded by a solid line circle stands out, the feature determination unit 152 determines that the eigenvalue ratio of the Lth eigenvalue is equal to or greater than a predetermined value. It is determined that the estimation result is highly reliable.

In this way, the feature determining unit 152 can use the features of the eigenvalue distribution determined by the feature determining unit 152 to determine whether the L-th and L+1-th eigenvalues in the eigenvalue distribution are rapidly decreasing. Therefore, it can be determined whether the estimation result of the direction estimation unit 16 is highly reliable.

This is because if there are L living bodies 50 in the target space, there will be L propagation paths of radio waves through the living bodies 50, and there will be L eigenvalues that are significantly larger than noise such as thermal noise. .

In other words, when there are L living organisms 50 in the target space, if the number of eigenvalues that are significantly larger than noise is less than L, it means that there are living organisms 50 whose direction cannot be estimated. Furthermore, when there are L living organisms 50 in the target space, if the number of eigenvalues that are significantly larger than noise is L+1 or more, the number of eigenvalues that are significantly larger than the noise is greater than the number of living organisms 50 that actually exist. This means that the image (living body 50) is visible. In other words, if there are L living organisms 50 in the target space, there should be only L eigenvalues that are significantly large, so there should be a difference in size between the Lth eigenvalue and the L+1th eigenvalue. . Therefore, the feature determining unit 152 can evaluate the reliability of the estimation result of the direction estimating unit 16 by evaluating the magnitude of the difference between the L-th eigenvalue and the L+1-th eigenvalue.

Note that the reliability determination unit 153 may output the calculated eigenvalue ratio as it is as a determination result of the reliability of the estimation result of the direction estimation unit 16.

[Direction estimation unit 16]
The direction estimation unit 16 estimates the direction of the living body 50 using a predetermined method from the correlation matrix according to the reliability estimated by the reliability estimation unit 15. More specifically, the direction estimating unit 16 uses the correlation matrix calculated by the biological information calculating unit 12 to determine the direction of the living body 50 as viewed from the receiving antenna unit 111 according to the reliability estimated by the reliability estimating unit 15. Estimate the direction.

Here, the direction estimation unit 16 may estimate the direction of the living body 50 using a MUSIC (Multiple Signal Classification) method, a beamformer method, or the like.

Note that the direction estimation unit 16 may change its output depending on whether the reliability estimated by the reliability estimation unit 15 is high or low. For example, if the reliability estimated by the reliability estimation unit 15 is low, the direction estimation unit 16 discards the estimation result of estimating the direction of the living body 50 from the correlation matrix in which the eigenvalues used for reliability estimation are calculated, The previous estimation result may be output. More specifically, when the reliability estimation unit 15 determines that the reliability is low, the direction estimation unit 16 estimates the direction of the living body 50 from the time when the reliability is determined to be low. Of the estimation results for the past period, the estimation results at a past point in time determined to be highly reliable by the reliability estimation unit 15 may be output.

Further, instead of discarding the estimation result, the direction estimation unit 16 may output the estimation result together with information such as a mark indicating that the reliability is low.

Further, the direction estimation unit 16 discards the estimation result when the reliability estimated by the reliability estimation unit 15 is low and the distance is greater than a predetermined threshold from the previous estimation result performed by the direction estimation unit 16. However, the previous estimation result may be output. In other words, if the reliability estimation unit 15 determines that the reliability is low, and the current estimation result is separated by a time equal to or more than the threshold value from the previous estimation result, the direction estimation unit 16 determines that the reliability is low. Among the estimation results for a period past the determined time point, the estimation result for a past time point determined to be highly reliable may be output.

Further, the direction estimation unit 16 sets an interval that is set according to the reliability estimated by the reliability estimation unit 15, calculates the time average or median of the past estimation results for the interval, and calculates the estimation result. You can also output it as In other words, when the reliability estimating unit 15 determines that the reliability is low, the direction estimating unit 16 sets the time range of the first interval that is set longer than the reference interval according to the determined degree of reliability. , the average or median value of past estimation results stored in the storage unit 14 may be calculated and output as the current estimation result.

Note that when the direction estimation unit 16 performs direction estimation using a method such as the MUSIC method, the beamformer method, or the Capon method, the direction estimation unit 16 obtains a spectrum of the evaluation function. In such a case, the direction estimation unit 16 may use the maximum value of each spectrum to evaluate the reliability of the estimation result, and may output the maximum value according to the evaluated reliability.

[Operation of estimation device 1]
Next, the operation of the estimation device 1 configured as above will be explained using the drawings.

FIG. 8 is a flowchart showing an example of the operation of the estimation device 1 in the first embodiment. FIG. 9 is a flowchart showing another example of the operation of the estimation device 1 in the first embodiment. Elements similar to those in FIG. 8 are given the same reference numerals.

First, the estimation device 1 observes a received signal for a predetermined period (S1). More specifically, the estimation device 1 observes, for a predetermined period, a received signal including a reflected signal transmitted from one transmitting antenna element and reflected by the living body 50 present in the target space.

Next, the estimation device 1 calculates a complex transfer function from the received signal observed in step S1 (S2). More specifically, the estimation device 1 calculates a complex signal representing the propagation characteristics between the transmitting antenna element and each of the N receiving antenna elements from a received signal received for a predetermined period by each of the N receiving antenna elements. Calculate multiple transfer functions. Since the details are as described above, the explanation here will be omitted. The same applies below.

Next, the estimation device 1 extracts a biological component that is a biological component due to the influence of the living body 50 from the complex transfer function calculated in step S2 (S3). More specifically, the estimation device 1 extracts a biological component due to the influence of the living body 50 and a biological component in each of the N reception antenna elements from the plurality of complex transfer functions calculated in step S2.

Next, the estimation device 1 calculates a correlation matrix of the biological components extracted in step S3 (S4). More specifically, the estimation device 1 calculates a correlation matrix from the biological components in each of the N reception antenna elements extracted in step S3.

Next, the estimation device 1 calculates the eigenvalue of the correlation matrix calculated in step S4 (S5). More specifically, the estimation device 1 calculates one or more eigenvalues of the correlation matrix calculated in step S4.

Next, the estimation device 1 estimates the reliability of the estimation result using the eigenvalues calculated in step S5 (S6). More specifically, the estimation device 1 determines the direction of the living organisms 50 using living organism count information indicating the number of living organisms 50 existing in the target space and one or more eigenvalues calculated in step S4. Estimate the reliability of the estimation result.

Next, the estimating device 1 estimates the direction of the living body 50 from the correlation matrix calculated in step S4 according to the reliability estimated in step S6 (S7). Here, the estimation device 1 may estimate the living body 50 using the MUSIC method or the like from the correlation matrix calculated in step S4.

Note that the estimation device 1 may change the output depending on whether the reliability estimated in step S6 is high or low. A detailed explanation will be given below using FIG. 9.

That is, in step S7, the estimation device 1 may first determine whether the reliability estimated in step S6 is high (S7A). More specifically, the estimation device 1 may determine whether the reliability estimated in step S6 is high or low based on whether the reliability estimated in step S6 is greater than or equal to a predetermined value.

In step S7A, if it is determined that the reliability is high (Yes in S7A), the estimation device 1 may estimate the direction of the living body 50 from the correlation matrix calculated in step S4, and output the estimation result ( S7B).

On the other hand, if it is determined in step S7A that the reliability is low (No in S7A), the estimation device 1 may skip the current estimation result estimated from the correlation matrix calculated in step S4, or may skip the current estimation result estimated from the correlation matrix calculated in step S4. The estimation result may be discarded and the estimation result estimated last time may be output (S7C).

FIG. 10 is a flowchart showing an example of detailed processing of step S6 in the first embodiment.

In step S6, the estimation device 1 first calculates the eigenvalue distribution (S61). More specifically, the estimation device 1 calculates an eigenvalue distribution obtained by sorting the plurality of eigenvalues calculated in step S5 by size. For example, the estimation device 1 may calculate a graph of the eigenvalue distribution as shown in FIG.

Next, the estimation device 1 determines the characteristics of the eigenvalue distribution calculated in step S61 (S62). More specifically, the estimation device 1 calculates ratio information indicating the ratio or difference between adjacent eigenvalues in the eigenvalue distribution calculated in step S61 as a feature of the eigenvalue distribution. For example, the estimation device 1 may calculate a graph of ratio information of the eigenvalue distribution as shown in FIG. 7 as a feature of the eigenvalue distribution shown in FIG.

Next, the estimation device 1 estimates the reliability of the estimation result using the characteristics of the eigenvalue distribution determined in step S62 (S63). More specifically, when the value indicated by the living body count information is L, the estimation device 1 determines whether the value indicated by the ratio information of the L-th eigenvalue counted from the largest eigenvalue is greater than or equal to a predetermined value. to estimate the reliability of the estimation results. That is, in the eigenvalue distribution calculated in step S61, the estimation device 1 uses the value L of the number of living organisms indicated by the number of living organisms information to determine whether the Lth and L+1th eigenvalues are decreasing rapidly. The reliability of the estimation results is estimated.

(Modified example)
In the first embodiment described above, the transmitter 10 transmits a 2.45 GHz sine wave as an example, but the present invention is not limited to this. The transmitter 10 may transmit a multicarrier signal such as an OFDM (Orthogonal Frequency Division Multiplexing) signal as a transmission wave (radio wave). A case in which the transmitter 10 transmits a multicarrier signal such as an OFDM signal as a transmission wave will be described below as a modified example.

[Configuration of estimation device 1 of this modification]
The estimation device 1 of this modification differs from the estimation device 1 of the first embodiment in that it handles multicarrier signals. Hereinafter, the differences from Embodiment 1 will be mainly explained using FIG. 1.

[Transmitter 10]
As described above, the transmitter 10 of this modification generates a multicarrier signal, such as an OFDM signal, and transmits it from one transmitting antenna element included in the transmitting antenna section 101.

[Receiver 11]
The receiver 11 of this modification includes a receiving antenna section 111 and a receiving section 112, as in the first embodiment. The receiver 11 of this modification differs from the first embodiment in the signal that the receiver 112 receives.

Receiving section 112 observes the multicarrier signal for a predetermined period with receiving antenna section 111 having N receiving antenna elements. More specifically, the receiving unit 112 converts the multicarrier signal received for a predetermined period by each of the M _R receiving antenna elements into a low frequency signal that can be processed, and converts the multicarrier signal into a subcarrier signal. Disassemble each. For example, if the multicarrier signal is composed of s subcarriers, the receiving section 112 decomposes the signals into s low frequency signals and transmits them to the biological information calculating section 12.

[Biological information calculation unit 12]
The biological information calculating section 12 calculates s pieces of biological information from the s low frequency signals transmitted by the receiving section 112. Specifically, the biological information calculation unit 12 performs calculation processing of s pieces of biological information in parallel (performs in s parallel), and finally calculates s correlation matrices.

[Calculation section 13]
The calculation unit 13 calculates s sets of eigenvalues based on the s correlation matrices calculated by the biological information calculation unit 12.

[Reliability estimation unit 15]
The reliability estimation unit 15 uses the s sets of eigenvalues calculated by the calculation unit 13 to estimate the reliability of the s sets of estimation results when the direction estimation unit 16 estimates the direction of the living body 50. When the reliability estimation unit 15 estimates the direction of the living body 50 using the living body number information indicating the value of the number of living bodies 50 existing in the target space and the s set of eigenvalues calculated by the calculation unit 13, Estimate the reliability of the s sets of estimation results. Here, the living body number information is information given in advance, and indicates information that L living bodies 50 are present in the target space that is the measurement target range.

More specifically, the reliability estimation unit 15 calculates s sets of ratio information of adjacent eigenvalues when the eigenvalues calculated by the calculation unit 13 are sorted by value size. Then, in each of the ratio information of the s group, when the value indicated by the living body count information is L, the reliability estimation unit 15 estimates The results are determined to be highly reliable.

Note that the reliability estimation method for each subcarrier is the same as in Embodiment 1, so a detailed explanation will be omitted.

In this way, the reliability estimation unit 15 uses the characteristics of the eigenvalue distribution obtained from the eigenvalues of each of the s sets calculated by the calculation unit 13 to ensure that the estimation results of the s sets by the direction estimation unit 16 are highly reliable. It can be determined whether

This is because, as in the first embodiment, if the direction of the living organisms 50 existing in the target space can be successfully estimated, significantly large eigenvalues appear as many as the number of living organisms 50. On the other hand, if the direction of the living organisms 50 existing in the target space cannot be well estimated due to noise, significantly large eigenvalues appear equal to or greater than the number of living organisms 50. In other words, since the positioning error and eigenvalues around a certain positioning point (path) in the target space have similar tendencies, the eigenvalues can be used to determine the It can be seen whether the direction etc. of the living body 50 can be estimated successfully.

FIG. 11 is a diagram illustrating an example of a positioning error distribution when using a multicarrier signal in a modification of the first embodiment. The horizontal axis shows subcarriers, and the vertical axis shows time.

The positioning error distribution shown in FIG. 11 is obtained by estimating the position of the living body 50 using the MUSIC method for all subcarriers at multiple times using an OFDM signal. The points plotted in black in FIG. 11 indicate that the error is below a certain threshold.

FIG. 12A is a distribution diagram showing whether the ratio of the first eigenvalue and the second eigenvalue at the same timing as the positioning error distribution shown in FIG. 11 is equal to or greater than a threshold value. FIG. 12B is a distribution diagram showing whether the distribution of the ratio of the second eigenvalue and the third eigenvalue at the same timing as the positioning error distribution shown in FIG. 11 is equal to or larger than the threshold value. Note that, as in the first embodiment, the first eigenvalue means the largest eigenvalue, the second eigenvalue means the second largest eigenvalue, and the third eigenvalue means the third largest eigenvalue. Moreover, the same timing means that the timing is the same in the time direction and the subcarrier direction shown in FIG.

In FIG. 12A, points where the ratio of the first eigenvalue to the second eigenvalue is equal to or less than a certain threshold are plotted in black on the same vertical and horizontal axes as in FIG. 11. Comparing FIG. 12A and FIG. 11, in FIG. 12A, in the region in FIG. 11 that corresponds to the region where the ratio of the first eigenvalue to the second eigenvalue is more than the threshold value (that is, the same region), It can be seen that the positioning error tends to be below a certain threshold.

Further, in FIG. 12B, points where the ratio of the second eigenvalue to the third eigenvalue is equal to or less than a certain threshold are plotted in black on the same vertical and horizontal axes as in FIG. 11. Comparing FIG. 12B and FIG. 11, in FIG. 12B, the region in FIG. 11 that corresponds to the region where the ratio of the second eigenvalue to the third eigenvalue is more than the threshold (that is, the same region), It can be confirmed that the positioning error tends to be above a certain threshold.

Therefore, as can be seen from FIGS. 11 to 12B, the reliability estimation unit 15 uses the distribution of the eigenvalue ratio as a feature of the eigenvalue distribution obtained from the s set of eigenvalues calculated by the calculation unit 13 to estimate the direction. It can be determined whether the estimation result of the unit 16 is highly reliable.

[Direction estimation unit 16]
The direction estimation unit 16 estimates the direction of the living body 50 using a predetermined method from the correlation matrix according to the reliability estimated by the reliability estimation unit 15. More specifically, the direction estimation unit 16 determines the direction of the living body 50 based on the s correlation matrices calculated by the biological information calculation unit 12 and the s reliability estimated by the reliability estimation unit 15. presume. Here, the direction estimation unit 16 estimates the direction of the living body 50 by using the MUSIC method or the like on the correlation matrix of each subcarrier.

In this modification, the direction estimation unit 16 combines the estimation result of the direction of the living body 50 and the reliability estimated by the reliability estimation unit 15 using a predetermined method to obtain a final estimation result. Here, for example, the direction estimation unit 16 may output the one with the highest reliability among the s estimation results as the final estimation result, or the direction estimation unit 16 may output the one with the highest reliability as the final estimation result, or the estimation result whose reliability value is greater than or equal to the threshold The average or median value may be output as the final estimation result.

Note that the direction estimation unit 16 selects the reliability estimation unit from among the estimation results of the direction of the living body 50 for a period in the past from the time when the reliability estimation unit 15 determines that the reliability is low. The estimation result at a past point in time determined to be highly reliable by No. 15 may be output.

For example, referring to FIG. 12A, the direction estimating unit 16 selects a time point determined to be highly reliable among the estimation results in the time direction other than a certain time point determined to be low reliability by the reliability estimating unit 15. The estimation result may be output.

The direction estimating unit 16 also selects the direction estimating unit 15 from among the estimation results of the direction of the living body 50 in a frequency range other than the frequencies determined to have low reliability by the reliability estimating unit 15. The estimation result at a frequency determined to be highly reliable may be output.

For example, to explain using FIG. 12A, the direction estimation unit 16 selects a subcarrier that is determined to be highly reliable among the estimation results of subcarriers other than the one subcarrier determined to be low reliability by the reliability estimation unit 15. The subcarrier estimation results may be output.

Furthermore, when the reliability estimation section 15 determines that the reliability is low, the direction estimation section 16 may combine the above. That is, the direction estimating unit 16 selects a past estimation result that has been determined to be highly reliable by the reliability estimating unit 15 among the estimation results for a predetermined period and a predetermined frequency range in the past, which are the estimation results of estimating the direction of the living body 50. Estimation results at time and frequency may be output.

For example, to explain using FIG. 12A, it is assumed that the reliability estimation unit 15 determines that the reliability of one subcarrier at a certain point in time (first point in time) is low. In this case, the direction estimation unit 16 may output the estimation result at a time point other than the first time point of a subcarrier other than a certain subcarrier determined to be highly reliable by the reliability estimation unit 15.

Further, if the reliability estimation unit 15 determines that the reliability is low, and the current estimation result is separated by a time equal to or more than a threshold value from the previous estimation result, the direction estimation unit 16 determines that the reliability is low. Among the estimation results for a period past the time when the estimation was made, the estimation result for the past time determined to be more reliable may be output.

For example, this will be explained using FIG. 12A. First, the estimation result at the first time point of the one subcarrier estimated by the direction estimation unit 16 is the estimation result estimated this time, and the estimation result at the second time point immediately before the first time point of the one subcarrier is the estimation result estimated this time. Assume that this is the previous estimation result. Assume that the reliability estimating unit 15 determines that the reliability of one subcarrier at a certain first point in time is low, and that the estimation result estimated this time is separated by a time equal to or more than a threshold value from the estimation result of the immediately previous estimation.

In this case, the direction estimating unit 16 selects a subcarrier that is determined to be highly reliable among the estimation results of the one subcarrier in a period past the first time point that was determined to be highly reliable by the reliability estimating unit 15. The estimation result of the one subcarrier at a past point in time may be output.

In addition, a second interval is set longer than the reference interval depending on the reliability, and the average or median value in the frequency direction of the estimation results included in the second interval among the past estimation results is calculated and output as the estimation result. You can. In other words, when the reliability estimation unit 15 determines that the reliability is low, the direction estimation unit 16 operates in the frequency range of the second interval that is set longer than the reference interval according to the determined degree of reliability. , the average or median value of past estimation results stored in the storage unit 14 may be calculated and output as the current estimation result.

Note that when the reliability estimation unit 15 determines that the reliability is low, the direction estimation unit 16 calculates the time range of the first interval that is set longer than the reference interval according to the determined degree of reliability. The average or median value of past estimation results stored in the storage unit 14 may be calculated and output as the current estimation result.

In this way, the length of an interval with low reliability is increased from the standard and the average value or median value is calculated, thereby improving the estimation accuracy.

[Effects etc.]
According to the estimation device 1 and the estimation method of the first embodiment and the modified example, it is possible to estimate the direction of the living body 50 existing in the target space and evaluate its reliability using a wireless signal. More specifically, a biological component due to the influence of the living body 50 is extracted from the complex transfer function between one transmitting antenna element and each of N receiving antenna elements, and a correlation matrix and its eigenvalue are extracted from the extracted biological component. Calculate. Then, the reliability of the measurement results is estimated using a predetermined method using the calculated eigenvalues, and the direction of the living body 50 is estimated using the wireless signal. It is possible to know the reliability of the estimation result at the same time as making the estimation.

Further, according to the estimation device 1 and the estimation method of the first embodiment and the modification, in order to extract only the components related to the living body 50 from the received signal, the living body 50 to be detected is equipped with a special device such as a transmitter. The direction of the living body 50 can be estimated even without it.

Further, by sequentially performing such direction estimation processing, it is possible to track the number of living organisms 50 and 50 positions of the living organisms. Thereby, the position of the living body 50 existing in the target space can be grasped in real time using the wireless signal.

Moreover, according to the estimation device 1 and the estimation method of the modified example, not only can multiple measurements be performed simultaneously not only in the time direction but also in the subcarrier direction, that is, the frequency direction, and reliability determination can be performed. Thereby, highly reliable estimation results in the frequency direction can be searched and output, and estimation accuracy can be improved.

(Embodiment 2)
In Embodiment 1, the case where a transmitter and a receiver having a SIMO configuration are used has been described as an example, but the present invention is not limited to this. The correlation matrix of the received signal is observed using a transmitter and receiver with a MIMO (Multiple Input Multiple Output) configuration, and the position of a living body existing in the target space is estimated using a predetermined method using the eigenvalues of the correlation matrix. It's okay. This case will be described below as a second embodiment.

Below, differences from Embodiment 1 will be mainly explained.

[Configuration of estimation device 1A]
FIG. 13 is a block diagram showing an example of the configuration of estimation device 1A in the second embodiment. Estimation device 1A shown in FIG. 13 differs from estimation device 1 of Embodiment 1 in the configuration of transmitter 10A. With this configuration, it is possible to estimate the position of the living body 50 existing in the target space and to evaluate the reliability of the position estimation result.

[Transmitter 10A]
The transmitter 10A includes a transmitting antenna section 101A and a transmitting section 102.

The transmitting antenna section 101A has a plurality of transmitting antenna elements. In this embodiment, the transmitting antenna unit 101A includes MT _antenna elements #1 to _#MT ( _MT is an integer of 2 or more), as shown in FIG. 13, for example.

Then, the transmitting section 102 transmits the generated transmission signal from the M _T transmitting antenna elements included in the transmitting antenna section 101A.

[Complex transfer function calculation unit 121]
Although the complex transfer function calculation unit 121 in the second embodiment has the same configuration as the first embodiment, the transmitting antenna unit 101A has a plurality of transmitting antenna elements, so that the complex transfer function calculation unit 121 has a uniform structure compared to the first embodiment. The processing of the parts is different. Specifically, since the estimation device 1A according to the present embodiment includes M _T transmitting antenna elements and M _R receiving antenna elements, the complex transfer function H calculated by the complex transfer function calculation unit 121 is M It becomes a matrix with _R rows and M _T columns.

The complex transfer function calculation unit 121 in this embodiment first converts the complex transfer function of M _R rows and M _T columns into a vector of M _R ×M _T rows and 1 column using (Equation 1). Thereby, the complex transfer function calculation unit 121 in this embodiment can perform subsequent processing, that is, calculation of a complex transfer function, in the same manner as in the first embodiment.

[Direction estimation unit 16]
The direction estimation unit 16 in the second embodiment estimates the position of the living body 50 using a predetermined method from the correlation matrix according to the reliability estimated by the reliability estimation unit 15. More specifically, the direction estimation unit 16 in the second embodiment uses the correlation matrix calculated by the biological information calculation unit 12 to determine the reception of the living body 50 according to the reliability estimated by the reliability estimation unit 15. The position of the living body 50 is estimated by estimating a plurality of directions viewed from the antenna section 111.

Note that, similarly to the first embodiment, the direction estimation unit 16 in the second embodiment may change the output depending on whether the reliability estimated by the reliability estimation unit 15 is high or low. For example, when the reliability estimated by the reliability estimation unit 15 is low, the direction estimation unit 16 in the second embodiment estimates the position of the living body 50 from the correlation matrix that calculates the eigenvalues used for reliability estimation. The result may be discarded and the previous estimation result may be output. More specifically, when the reliability estimating unit 15 determines that the reliability is low, the direction estimating unit 16 in the second embodiment determines that the position of the living body 50 is estimated and that the reliability is low. Among the estimation results for a period past the determined time point, the estimation result for a past time point determined to be highly reliable by the reliability estimation unit 15 may be output.

Further, instead of discarding the estimation result, the direction estimation unit 16 in the second embodiment may output the estimation result together with information such as a mark indicating that reliability is low.

Further, the direction estimation unit 16 in the second embodiment is configured to perform the following operations when the reliability estimated by the reliability estimation unit 15 is low and the direction estimation unit 16 is far from the previous estimation result performed by the direction estimation unit 16 by a predetermined threshold value or more. , the estimation result may be discarded and the previous estimation result may be output. In other words, if the reliability estimation unit 15 determines that the direction estimation unit 16 has low reliability, and the current estimation result is separated from the previous estimation result by a time equal to or more than the threshold, the reliability of the direction estimation unit 16 is low. Among the estimation results for a period past the time determined as , the estimation results for a past time determined to be highly reliable may be output.

Further, the direction estimation unit 16 in the second embodiment sets an interval that is set according to the reliability estimated by the reliability estimation unit 15, and calculates the time average or median value of the past estimation results for the interval. It may be calculated and output as an estimation result. In other words, when the reliability estimation unit 15 determines that the reliability is low, the direction estimation unit 16 determines the time range of the first interval that is set longer than the reference interval according to the determined degree of reliability. Then, the average or median value of the past estimation results stored in the storage unit 14 may be calculated and output as the current estimation result.

Note that the direction estimation unit 16 in the second embodiment obtains a spectrum of an evaluation function when performing position estimation using a method such as the MUSIC method, the beamformer method, or the Capon method. In such a case, the direction estimation unit 16 may use the maximum value of each spectrum to evaluate the reliability of the estimation result, and may output the maximum value according to the evaluated reliability.

Further, the transmitter 10A may use a multicarrier signal such as an OFDM signal as a transmission wave (radio wave), as in the modification of the first embodiment. The configuration and effects in this case are the same as those of the modified example of Embodiment 1, except that not only the direction of the living body but also the position of the living body can be estimated as the estimation result, so the explanation will be omitted.

[Effects etc.]
According to the estimation device 1A and the estimation method of the second embodiment, the position of the living body 50 existing in the target space can be estimated using a wireless signal. More specifically, a biological component due to the influence of the living body 50 is extracted from a complex transfer function between a plurality of transmitting antenna elements and each of N receiving antenna elements, and a correlation matrix and its eigenvalue are extracted from the extracted biological component. Calculate. Then, using the calculated eigenvalues, the position of the living body 50 is estimated and the reliability of the estimation result is evaluated using a predetermined method. In this way, the position of a living body existing within the target space can be estimated using radio signals.

Further, in the estimation device 1A and the estimation method of the second embodiment, in order to extract only the components related to the living body 50 from the received signal, similarly to the estimation device 1 and the like of the first embodiment, the transmitter is attached to the living body 50 to be detected. It is possible to estimate the location of a living body and the reliability of its measurement results without having to carry special equipment such as .

Furthermore, in the second embodiment, since the transmitter 10A has two or more transmitting antenna elements, it is possible to estimate not only the direction in which the living body 50 exists but also the position thereof.

By continuously performing such position estimation for a predetermined period of time, the positions of one or more living organisms can be continuously tracked for a predetermined period of time.

(Embodiment 3)
In Embodiments 1 and 2, the reliability of the estimation result of the direction or position of the living body is estimated using the eigenvalue of the observed received signal, but the present invention is not limited thereto. The reliability of the estimation result of the direction or position of the living body may be estimated using the evaluation function spectrum of a direction of arrival estimation method such as the MUSIC method or the Capon method. This case will be described below as a third embodiment. Below, differences from Embodiment 1 will be mainly explained.

[Configuration of estimation device 1B]
FIG. 14 is a block diagram showing an example of the configuration of estimation device 1B in the third embodiment. Estimating device 1B shown in FIG. 14 differs from estimating device 1 of Embodiment 1 in the configurations of direction estimating section 16B and reliability estimating section 15B.

[Direction estimation unit 16B]
The direction estimation unit 16B calculates an evaluation function using the correlation matrix calculated by the correlation matrix calculation unit 123, and estimates the direction or position of the living body 50 using the calculated evaluation function. Here, the evaluation function calculated by the direction estimation unit 16B may be a MUSIC spectrum, a Capon method spectrum, or a beamformer method spectrum.

In this embodiment, the direction estimation unit 16B uses the eigenvalues and eigenvectors calculated by the calculation unit 13 and stored in the storage unit 14 to perform the MUSIC method, Capon method, or beamformer method as in the first embodiment. Direction estimation is performed using a direction of arrival estimation method such as Further, the direction estimation unit 16B transmits the spectrum of the evaluation function used in the direction of arrival estimation method to the reliability estimation unit 15B.

Since the direction or position of the living body 50 can be similarly estimated using any of the MUSIC method, Capon method, and beamformer method, the case where the MUSIC method is used will be described below as an example.

と、推定装置１Ｂの対象空間内（センシング範囲内）に存在する生体の数Ｌとから、（式２）で表される評価関数Ｐ_{ｍｕｓｉｃ}（θ）の角度θに対するスペクトルを算出する。このスペクトルは、ＭＵＳＩＣスペクトルと称される。 That is, the direction estimation unit 16B calculates the eigenvalue Λ and the eigenvector calculated by the calculation unit 13.

and the number L of living organisms existing within the target space (within the sensing range) of the estimation device 1B, the spectrum of the evaluation function P _music (θ) expressed by (Equation 2) with respect to the angle θ is calculated. This spectrum is called the MUSIC spectrum.

The direction estimation unit 16B searches for the top L maximum values of the MUSIC spectrum expressed by (Equation 2), and estimates the corresponding θ as the direction of the living body. Further, the direction estimation unit 16B outputs the value of P _music (θ) corresponding to the top L local maximum values.

[Reliability estimation unit 15B]
The reliability estimator 15B uses the evaluation function calculated by the direction estimator 16B to estimate the reliability of the estimation result when estimating the position or direction of the living body 50. The reliability estimating unit 15B determines that the reliability of the estimation result is low when the estimated reliability is lower than the threshold, and determines that the reliability of the estimation result is high when the estimated reliability is greater than or equal to the threshold.

In this embodiment, the reliability estimation section 15B estimates the reliability of the estimation result of the direction estimation section 16B based on the maximum value of the evaluation function P _music (θ) calculated by the direction estimation section 16B. Here, the reliability estimating unit 15B may use the maximum value of the evaluation function P _music (θ) as is, or calculate the ratio between the maximum value and the minimum value of the evaluation function P _music (θ) and calculate the ratio. may also be used. In the following, an example using the ratio between the local maximum value and the local minimum value will be described.

That is, the reliability estimation unit 15B determines whether each of the L maximum values of the evaluation function P _music (θ) is equal to or greater than the threshold value. When the reliability estimating unit 15B determines that some of the L maximum values are greater than or equal to the threshold, there is a possibility that θ, which takes some of the L maximum values, is the correct direction of the living body 50. is estimated to be high. In other words, the reliability estimating unit 15B determines that the estimation results corresponding to the partial maximum values of θ are highly reliable. Here, the threshold value may be a fixed value such as 2 dB, or a value determined by the value of L such as 2 dB when L is 2 and 3 dB when L is 1.

FIG. 15 is a diagram showing an example of a graph of an evaluation function in the third embodiment. FIG. 15 shows a graph of a MUSIC spectrum 501 when L is 3 as the evaluation function. As shown in FIG. 15, the MUSIC spectrum 501 has three maximum values 502A, 502B, and 502C.

When estimating the reliability of the estimation result of the direction estimation section 16B using the evaluation function graph shown in FIG. do. In the example shown in FIG. 15, the reliability estimation unit 15B determines that the local maximum value 502A and the local maximum value 502C are greater than or equal to the threshold value, and estimates that the estimation results corresponding to the local maximum value 502A and the local maximum value 502C are highly reliable. . On the other hand, the reliability estimation unit 15B determines that the local maximum value 502B is smaller than the threshold value, and estimates that the reliability of the estimation result corresponding to the local maximum value 502B is low.

Note that, similarly to the modification of the first embodiment, the transmitter 10 may use a multicarrier signal such as an OFDM signal as a transmission wave (radio wave). The configuration and effects in this case are the same as those in the modified example of the first embodiment, so description thereof will be omitted.

In addition, regarding the operation of the estimation device 1B configured as described above, it is possible to estimate not only the direction of the living body 50 but also the position of the living body 50 as an estimation result, and the evaluation function used when estimating the position is used to evaluate the estimation result. Since the process is the same as described in Embodiment 1 except for the point of evaluating reliability, the description will be omitted.

[Effects etc.]
According to the estimation device 1B and the estimation method of the third embodiment, the position of the living body 50 existing in the target space can be estimated using a radio signal. More specifically, a biological component due to the influence of the living body 50 is extracted from the complex transfer function between one transmitting antenna element and each of N receiving antenna elements, and a correlation matrix and its eigenvalue are extracted from the extracted biological component. Calculate. Then, by estimating the reliability of the measurement results using a predetermined method using the calculated eigenvalues and estimating the direction of the living body 50, the position of the living body 50 in the target space is estimated using the wireless signal. It is possible to know the reliability of the estimation result at the same time as making the estimation.

Further, according to the estimation device 1B and the estimation method of the third embodiment, since only the components related to the living body 50 are extracted from the received signal, the living body 50 to be detected does not need to have a special device such as a transmitter. , the position of the living body 50 can be estimated.

Further, in the estimation device 1B and the estimation method of the third embodiment, in order to extract only the components related to the living body 50 from the received signal, similarly to the estimation device 1 and the like of the first embodiment, the transmitter is attached to the living body 50 to be detected. It is possible to estimate the location of a living body and the reliability of its measurement results without having to carry special equipment such as .

Furthermore, by sequentially performing such position estimation processing, it is possible to track the number of living organisms 50 as well as the positions of the living organisms 50. Thereby, the position of the living body 50 existing in the target space can be grasped in real time using the wireless signal.

Further, according to the estimation device 1B and the estimation method of the third embodiment, not only can multiple measurements be performed simultaneously not only in the time direction but also in the subcarrier direction, that is, the frequency direction, and reliability determination can be performed. Thereby, highly reliable estimation results in the frequency direction can be searched and output, and estimation accuracy can be improved.

Although the estimation device and estimation method according to one aspect of the present disclosure have been described based on the embodiments, the present disclosure is not limited to these embodiments. Unless departing from the spirit of the present disclosure, various modifications that can be thought of by those skilled in the art to the present embodiment, or forms constructed by combining components of different embodiments are also included within the scope of the present disclosure. .

Further, the present disclosure can be realized not only as an estimation device including such characteristic components, but also as an estimation method using the characteristic components included in the estimation device as steps. can. Moreover, each characteristic step included in such a method can also be realized as a computer program that causes a computer to execute it. It goes without saying that such a computer program can be distributed via a non-transitory computer readable recording medium such as a CD-ROM or a communication network such as the Internet.

INDUSTRIAL APPLICABILITY The present disclosure can be used for an estimation method and an estimation device for estimating the direction or position of a living body using a wireless signal, and in particular can be used for an estimation method and an estimation device used in a monitoring device that detects the intrusion of a living body. .

1, 1A, 1B Estimation device 10, 10A Transmitter 11 Receiver 12 Biological information calculation section 13 Calculation section 14 Storage section 15, 15B Reliability estimation section 16, 16B Direction estimation section 50 Living body 101, 101A Transmission antenna section 102 Transmission section 111 Receiving antenna unit 112 Receiving unit 121 Complex transfer function calculation unit 122 Extraction unit 123 Correlation matrix calculation unit 151 Eigenvalue distribution calculation unit 152 Feature determination unit 153 Reliability determination unit

Claims (15)

An estimation device that estimates the position or direction of a living body existing in a target space using a transmitter having at least one transmitting antenna element and a receiver having N receiving antenna elements (N is a natural number of 2 or more). An estimation method,
A transfer function that calculates a plurality of complex transfer functions representing propagation characteristics between the transmitting antenna element and each of the N receiving antenna elements from a received signal received for a predetermined period by each of the N receiving antenna elements. a calculation step;
an extraction step of extracting a biological component in each of the N reception antenna elements, which is a biological component due to the influence of a living body, from the plurality of complex transfer functions calculated in the transfer function calculation step;
a correlation matrix calculation step of calculating a correlation matrix from the biological components in each of the N receiving antenna elements extracted in the extraction step;
a calculation step of calculating one or more eigenvalues of the correlation matrix calculated in the correlation matrix calculation step;
An estimation result when the position or direction of the living body is estimated using living body count information indicating the value of the number of living creatures existing in the target space and the one or more eigenvalues calculated in the calculation step. a reliability estimation step of estimating reliability;
an estimation step of estimating the position or direction of the living body using a predetermined method from the correlation matrix depending on the reliability of the estimation result;
Estimation method. In the reliability estimation step,
(1) calculating ratio information indicating the ratio or difference between adjacent eigenvalues when the one or more eigenvalues calculated in the calculation step are sorted in descending order of values;
(2) When the value indicated by the biological number information is L (L is a natural number of 1 or more), and the ratio information corresponding to the L-th eigenvalue when sorted is a predetermined value or more, the estimation result It is determined that the reliability of
The estimation method according to claim 1. An estimation device for estimating the direction or position of a living body existing in a target space, using a transmitter having at least one transmitting antenna element and a receiver having N receiving antenna elements (N is a natural number of 2 or more). An estimation method,
A transfer function that calculates a plurality of complex transfer functions representing propagation characteristics between the transmitting antenna element and each of the N receiving antenna elements from a received signal received for a predetermined period by each of the N receiving antenna elements. a calculation step;
an extraction step of extracting a biological component in each of the N reception antenna elements, which is a biological component due to the influence of a living body, from the plurality of complex transfer functions calculated in the transfer function calculation step;
a correlation matrix calculation step of calculating a correlation matrix from the biological components in each of the N receiving antenna elements extracted in the extraction step;
an estimation step of calculating an evaluation function using the correlation matrix and estimating the direction or position of the living body using the calculated evaluation function;
and a reliability estimation step of estimating the reliability of the estimation result when estimating the position or direction of the living body using the evaluation function calculated in the estimation step.
Estimation method. the evaluation function calculated in the estimation step is a MUSIC spectrum;
The estimation method according to claim 3. The evaluation function calculated in the estimation step is a spectrum of the Capon method.
The estimation method according to claim 3. The evaluation function calculated in the estimation step is a spectrum of a beamformer method,
The estimation method according to claim 3. In the reliability estimation step,
If the estimated reliability is lower than a threshold, the reliability is determined to be low; if the estimated reliability is greater than or equal to the threshold, the reliability is determined to be high;
The estimation method according to any one of claims 3 to 6. In the estimation step,
If the reliability is determined to be low in the reliability estimation step, the estimation result of estimating the direction or position of the living body, which is an estimation result from a period in the past than the time when the reliability was determined to be low. outputting an estimation result at a past point in time where the reliability was determined to be high in the reliability estimation step;
The estimation method according to any one of claims 1 to 7. In the estimation step,
If the reliability is determined to be low in the reliability estimation step, among the estimation results of estimating the direction or position of the living body in a frequency range other than the frequency for which the reliability is determined to be low. , outputting the estimation result at the frequency determined to have high reliability in the reliability estimation step;
The estimation method according to any one of claims 1 to 7. In the estimation step,
If it is determined that the reliability is low in the reliability estimation step, the reliability estimation is the estimation result of estimating the direction or position of the living body, and is the estimation result of the past predetermined period and predetermined frequency range. outputting the estimation result at the past time and frequency determined to have high reliability in the step;
The estimation method according to any one of claims 1 to 7. In the estimation step,
If the reliability is determined to be low in the reliability estimation step, and the estimation result estimated in the estimation step is separated by a time or frequency equal to or more than a threshold value from the immediately preceding estimation result, the reliability is determined to be low. Outputting the estimation results of the past time point determined to have high reliability among the estimation results of the past period than the time point at which the reliability was determined to be high;
The estimation method according to any one of claims 1 to 7. Further, the method further includes a storage step of storing estimation results estimated in the past in the estimation step in a storage unit,
In the estimation step,
If the reliability is determined to be low in the reliability estimation step, the past time stored in the storage unit is determined in the time range of the first interval, which is set longer than the reference interval according to the degree of reliability. Calculate the average or median of the estimation results and output it as the current estimation result,
The estimation method according to any one of claims 1 to 7. Further, the method further includes a storage step of storing estimation results estimated in the past in the estimation step in a storage unit,
If the reliability is determined to be low in the reliability estimation step, the frequency range of the second interval is set longer than the reference interval according to the degree of reliability, and the frequency range of the past stored in the storage unit is set to be longer than the reference interval. Calculate the average or median of the estimation results and output it as the current estimation result,
The estimation method according to any one of claims 1 to 7. An estimation device that estimates the position or direction of a living body existing in a target space using a transmitter having at least one transmitting antenna element and a receiver having N receiving antenna elements (N is a natural number of 2 or more). There it is,
A transfer function that calculates a plurality of complex transfer functions representing propagation characteristics between the transmitting antenna element and each of the N receiving antenna elements from a received signal received for a predetermined period by each of the N receiving antenna elements. A calculation section,
an extraction unit that extracts a biological component in each of the N receiving antenna elements, which is a biological component due to the influence of a living body, from the plurality of complex transfer functions calculated in the transfer function calculation unit;
a correlation matrix calculation unit that calculates a correlation matrix from the biological components in each of the N reception antenna elements extracted in the extraction unit;
a calculation unit that calculates one or more eigenvalues of the correlation matrix calculated in the correlation matrix calculation unit;
an estimation result when the position or direction of the living body is estimated using living body count information indicating the value of the number of living creatures existing in the target space and the one or more eigenvalues calculated in the calculation unit; a reliability estimation unit that estimates reliability;
an estimation unit that estimates the position or direction of the living body from the correlation matrix using a predetermined method depending on the reliability of the estimation result;
Estimation device. An estimation device that estimates the position or direction of a living body existing in a target space using a transmitter having at least one transmitting antenna element and a receiver having N receiving antenna elements (N is a natural number of 2 or more). A program that causes a computer to execute an estimation method,
A transfer function that calculates a plurality of complex transfer functions representing propagation characteristics between the transmitting antenna element and each of the N receiving antenna elements from a received signal received for a predetermined period by each of the N receiving antenna elements. a calculation step;
an extraction step of extracting a biological component in each of the N reception antenna elements, which is a biological component due to the influence of a living body, from the plurality of complex transfer functions calculated in the transfer function calculation step;
a correlation matrix calculation step of calculating a correlation matrix from the biological components in each of the N receiving antenna elements extracted in the extraction step;
a calculation step of calculating one or more eigenvalues of the correlation matrix calculated in the correlation matrix calculation step;
An estimation result when the position or direction of the living body is estimated using living body count information indicating the value of the number of living creatures existing in the target space and the one or more eigenvalues calculated in the calculation step. a reliability estimation step of estimating reliability;
causing a computer to perform an estimating step of estimating the position or direction of the living body using a predetermined method from the correlation matrix depending on the reliability of the estimation result;
program.

Images