JP7628024B2 - Visibility determination device and terminal holder detection system - Google Patents

Description

The present invention relates to a visibility determination device and a terminal holder detection system.

In recent years, in order to prevent information leaks, there has been a demand to detect anyone who brings a transmitting device such as a smartphone into a highly confidential room.
Patent Document 1 describes a technology for estimating a person carrying a transmitting terminal in a monitored space. The terminal holder detection system described in Patent Document 1 arranges multiple receiving devices in a highly confidential room (hereinafter also referred to as a monitored area) and detects the position of a person in the room using a position sensor. Each detected person is assumed in turn as a candidate carrying a transmitting terminal, and the theoretical value of the receiving strength of each receiving device when radio waves are transmitted from the transmitting terminal carried by the candidate person is calculated, taking into account the influence of known fixtures and the like in the monitored area. The actually measured receiving strength is compared with the theoretical value calculated for each person to find which is closest to it, and the person carrying the transmitting terminal is identified. As the receiving strength, an angular spectrum, which is the distribution of the receiving strength of radio waves for each direction of arrival of the radio waves as seen from the antenna, is used.

JP 2017-181137 A

When receiving radio waves transmitted from a transmitting terminal with an antenna, the characteristics of the received signal obtained by receiving the radio waves with the antenna are different between a line-of-sight (LOS) environment where there are no obstacles between the transmitting terminal and the antenna and a non-line-of-sight (NLOS) environment where there are obstacles between the transmitting terminal and the antenna. For this reason, it is very useful to determine whether the environment between the transmitting terminal and the antenna is a line-of-sight environment or a non-line-of-sight environment.
The present invention focuses on the above-mentioned points and has an object to provide a means for determining whether the environment between a transmitting terminal and an antenna is a line-of-sight environment or a non-line-of-sight environment.

According to one aspect of the present invention, a visibility state determination device is provided that includes an angular spectrum calculation unit that calculates an angular spectrum, which is a distribution of the reception strength of radio waves for each direction of arrival of the radio waves, based on a received signal obtained by receiving radio waves transmitted from a transmitting terminal with an antenna, and a visibility determination unit that determines that the environment between the transmitting terminal and the antenna is a line-of-sight environment if the time fluctuation of the angular spectrum is small, and determines that the environment between the transmitting terminal and the antenna is a non-line-of-sight environment if the time fluctuation is large.

According to another aspect of the present invention, there is provided a terminal holder detection system comprising: an antenna installed in a monitored space for receiving radio waves from a transmitting terminal carried by a mover within the monitored space; a position detection device for detecting the position of each mover; an angular spectrum calculation unit for calculating an actual measured value of an angular spectrum, which is the distribution of the reception strength of radio waves for each direction of arrival of the radio waves, based on a received signal obtained by receiving radio waves transmitted from the transmitting terminal by the antenna; a memory unit for storing, for each antenna, a theoretical value of the angular spectrum when radio waves transmitted from near the position of the mover are received by the antenna, assuming that the mover is carrying a transmitting terminal; a comparison and decision unit for determining and integrating the similarity between the theoretical value and the actual measured value for each antenna and determining whether or not the mover is carrying a transmitting terminal based on the integrated similarity; and a line-of-sight determination unit for determining that the environment between the transmitting terminal and the antenna is a line-of-sight environment if the time fluctuation of the actual measured value of the angular spectrum is small, and for determining that the environment between the transmitting terminal and the antenna is a non-line-of-sight environment if the time fluctuation is large.
The visibility determination unit calculates a visibility accuracy indicating the accuracy that the environment between the transmitting terminal and the antenna is within the line of sight, and increases the visibility accuracy when the time variation is small compared to when the time variation is large. The comparison determination unit weights the similarity according to the visibility accuracy.

The present invention provides a means for determining whether the environment between a transmitting terminal and an antenna is line-of-sight or non-line-of-sight.

1 is a block diagram showing a schematic configuration of an example of a visibility condition determination device according to a first embodiment of the present invention; FIG. 13 is a diagram illustrating an example of an angular spectrum. 6A and 6B are diagrams illustrating a first and a second example of a method for calculating the time variation of an angular spectrum. FIG. 2A is a diagram showing an example of an angle-time spectrum in a non-line-of-sight environment, and FIG. 2B is a diagram showing an example of an angle-time spectrum in a line-of-sight environment. FIG. 11 is a schematic diagram showing how a terminal holder detection system according to a second embodiment of the present invention is installed and operated. 1 is a block diagram showing a schematic configuration of an example of a terminal owner detection system; 13A to 13D are graphs showing theoretical values of the angular spectrum of radio waves received by the receiving devices 300 to 303, respectively. 13A to 13D are graphs showing actual measured values of the angular spectrum of radio waves received by the receiving devices 300 to 303, respectively. 10 is a flowchart showing an example of a terminal owner detection method according to a second embodiment of the present invention. 1A shows an example of a scene in which a person carrying a transmitting terminal is present in a monitored space, and FIG. 1B is an explanatory diagram of the position at which the person carries the transmitting terminal. (a) is the similarity between the theoretical value of the angular spectrum of radio waves received by a receiving device when assuming that each person is holding a transmitting terminal in their respective holding positions, and the actual measured value of the angular spectrum of radio waves received by the receiving device, and (b) is a table showing the similarity after the similarity in (a) is weighted by the line of sight accuracy. (a) is the similarity obtained by integrating the similarities for each receiving device shown in Figure 11 (a) for each person, and (b) is the similarity obtained by integrating the similarities for each receiving device shown in Figure 11 (b) for each person.

The following describes an embodiment of the present invention with reference to the drawings. Note that the embodiment of the present invention shown below is an example of an apparatus and method for embodying the technical concept of the present invention, and the technical concept of the present invention does not specify the structure, arrangement, etc. of the components as described below. The technical concept of the present invention can be modified in various ways within the technical scope defined by the claims.

First Embodiment
1 is a block diagram showing a schematic configuration of an example of a visibility state determination device according to a first embodiment of the present invention. The visibility state determination device 1 according to the first embodiment can be applied not only to the terminal holder detection system described above, but also to various applications in which radio waves transmitted from a transmitting terminal 2 are received by an antenna 3, in order to determine whether the environment between the transmitting terminal 2 and the antenna 3 is a line-of-sight environment or a non-line-of-sight environment.

The visibility condition determining device 1 includes an angular spectrum calculating unit 4 and a visibility determining unit 5 .
The angular spectrum calculation unit 4 calculates an angular spectrum, which is the distribution of the reception strength of radio waves for each direction of arrival of the radio waves, based on the received signal obtained by receiving the radio waves transmitted from the transmitting terminal 2 using the antenna 3, which is an array antenna.
2 is a diagram showing an example of an angular spectrum. The angular spectrum represents a distribution of reception strength obtained by arranging the strength of radio waves arriving from each angle, with a predetermined reference direction as seen from the antenna 3 being 0°.

The angular spectrum can be calculated using the well-known Beamformer method, the MUSIC (Multiple Signal Classification) method, etc. These methods are described, for example, in "IEICE Knowledge Base Knowledge Forest 4 Group-2 Chapter 8 8-4 Direction of Arrival Estimation http://www.ieice-hbkb.org/files/04/04gun_02hen_08.pdf#page=11".

When the environment between the transmitting terminal 2 and the antenna 3 is non-line-of-sight, direct waves from the transmitting terminal 2 do not reach the antenna 3, and radio waves reflected in complex ways by walls and other objects are combined and received by the antenna 3. When radio waves are combined in this way, the amplitude of the combined wave contains a component that depends on the phase difference between the radio waves contained in the combined wave. For this reason, the received strength of the combined wave varies depending on the phase of the radio waves contained in the combined wave. As a result, the angular spectrum in a non-line-of-sight environment varies more over time than in a line-of-sight environment.

For example, in the widely used LTE, the efficiency of information transmission is improved by changing the frequency in a time-division manner, but the phase also changes at the boundary of the frequency change. Furthermore, the information carried by the radio waves also changes, causing an even greater phase change.
In this way, the reception strength, which depends on the phase of the received radio wave, fluctuates greatly with time in a non-line-of-sight environment, so it can be determined that the smaller the time fluctuation, the more likely it is a line-of-sight environment, and the larger the time change, the more likely it is a non-line-of-sight condition.

Therefore, the visibility determining unit 5 determines whether the environment between the transmitting terminal 2 and the antenna 3 is a line-of-sight environment or a non-line-of-sight environment, based on the time variation of the angular spectrum calculated by the angular spectrum calculating unit 4 .
For this reason, the visibility determination unit 5 calculates the magnitude of time variation of the angular spectrum calculated by the angular spectrum calculation unit 4. The visibility determination unit 5 calculates the magnitude of time variation of the angular spectrum based on a plurality of angular spectra obtained from radio waves received at different times.

For example, angular spectra are acquired from radio waves received at multiple different sampling times t _i (i = 1, 2, ...), and the angular spectra for a predetermined time period (i.e., angular spectra acquired at a predetermined number of consecutive sampling times (t _n , t _n-1 , ... t _n-N : n and N are natural numbers)) are stored. The visibility determination unit 5 may calculate the magnitude of time variation based on any M angular spectra (M is a natural number) from these stored angular spectra. M is 2 or more. As the amount of information to be stored increases as M is increased, it is determined as an appropriate design matter. For example, M may be 5.

For example, the correlation (similarity) between the angular spectra may be calculated as an index representing the magnitude of the time fluctuation. Also, for example, the difference in the angle (direction of arrival of the radio wave) at which the maximum peak of the angular spectrum is obtained may be calculated as an index representing the magnitude of the time fluctuation. For example, the average of the correlation between M angular spectra or the average of the difference in the angle at which the maximum peak is obtained may be used as an index representing the magnitude of the time fluctuation.
For example, the visibility determination unit 5 may determine that the environment between the transmitting terminal 2 and the antenna 3 is a line-of-sight environment when the time variation is less than a threshold value, and may determine that the environment between the transmitting terminal 2 and the antenna 3 is a non-line-of-sight environment when the time variation is greater than or equal to the threshold value.

Furthermore, in a non-line-of-sight environment, the reception strength changes over time, but for a very short period of time, the time fluctuation may coincidentally become small, making it indistinguishable from a line-of-sight environment. Therefore, the line-of-sight determination unit 5 may calculate an index representing the degree of certainty that the environment between the transmitting terminal 2 and the antenna 3 is a line-of-sight environment (the degree to which the environment is line-of-sight) based on the variance of the time fluctuation in addition to the time fluctuation of the reception strength. Such an index is referred to as the "line-of-sight certainty LOSValue."

For example, the visibility determining unit 5 may calculate the visibility probability LOSValue based on the correlation (similarity) of the shapes of the angular spectra obtained from radio waves received at multiple times.
For example, a shape correlation value is calculated between angular spectra (SP _n , SP _n-1 , ..., SP _{n-M )} at any sampling time t _n and M consecutive sampling times (t _n-1 , ..., t _n-M ) from time t n, and the reception intensity similarity s(n) is calculated based on the shape correlation value. For example, a shape correlation value may be calculated for each of a plurality of combinations of angular spectra (SP _n , SP _n-1 , ..., SP _n-M ), and the average of these shape correlation values may be calculated as the reception intensity similarity s(n).

3A, for example, the visibility determination unit 5 calculates a shape correlation value corr(SP n , SP _n- m ) between an angular spectrum SP _n at an arbitrary sampling time t _n and angular spectra (SP _{n-1 ,} ..., SP _n-M ) at the past M sampling times (t _n -1 , ..., t _n- M ) (m=0, 1, ..., M-1).
The visibility determining unit 5 calculates the average value of these shape correlation values corr(SP _n , SP _nm-1 ) as the reception intensity similarity s(n) as shown in the following equation (1).

The reception intensity similarity s(n) in the above formula (1) approaches 1 as the shapes of the angular spectra become more similar, and approaches 0 as the shapes of the angular spectra become more dissimilar. Note that in the above formula (1), instead of the shape correlation value corr(SP _n , SP _n-m-1 ), it is also possible to calculate the shape correlation value corr(SP n- _m , SP n _- m-1 ) of the angular spectra SP n _{-m ,} SP _{n-m -1} at adjacent sampling times t _n-m , t n-m-1 of the angular spectra (SP _n , SP _{n-1 , ... SP n-M} ).
Furthermore, the visibility determination unit 5 calculates the variance v(n) of the reception intensity similarity s(n) from an arbitrary sampling time _tn to the (M-1)th sampling time _tn-M+1 , as shown in the following formula (2). The variance v(n) approaches 0 as the shapes of the angular spectra become more similar, and approaches 1 as they become more dissimilar.

Then, the visibility determining unit 5 calculates the visibility probability LOSValue of the following equation (3) based on the reception intensity similarity s(n) and the variance v(n).
As is clear from the following equation (3), the visibility LOSValue has a value in the range of 0 to 1, and the more similar the shapes of the angle spectra are, the closer it is to 1, and the more dissimilar it is, the closer it is to 0.

Furthermore, for example, the visibility determination unit 5 may calculate the visibility accuracy LOSValue based on the difference between the angles (directions of arrival of radio waves) at which the maximum peaks of the angular spectrum obtained from radio waves received at multiple times are obtained. The angle at which the maximum peak is obtained is hereinafter referred to as the "maximum peak angle."
For example, the difference in maximum peak angles between angular spectra (SP _n , SP _n-1 , ..., SP _{n-M )} at any sampling time t _n and M consecutive sampling times (t _n-1 , ..., t _n-M ) from time t n is calculated, and the reception intensity similarity s(n) is calculated based on the difference. For example, the difference in maximum peak angles for each of a plurality of combinations of angular spectra (SP _n , SP _n-1 , ..., SP _n-M ) may be calculated, and the average of these differences may be calculated as the reception intensity similarity s(n).

3B, for example, the visibility determination unit 5 calculates the difference (P _nm - P _{nm-1 )} between the maximum peak angles P _nm = argmax(SP _{nm ) and P nm-1 = argmax(SP nm} -1 ) of the angular spectrum at adjacent sampling times t _nm and t _{nm- 1} among the sampling times (t _{n , t n-1 ,} ... t _n -N ) (m = 0, ₁ , ... M-1).
The visibility determining unit 5 calculates the average value of these differences (P _nm - _{P nm -1} ) as the reception intensity similarity s(n) as shown in the following formula (4). The reception intensity similarity s(n) is normalized by dividing it by the magnitude C of the angular range of the angular spectrum. For example, when the angular range of the angular spectrum is -90° to 90°, C=180.

The reception intensity similarity s(n) in the above formula (4) approaches 0 as the shapes of the angular spectra become more similar, and approaches 1 as the shapes of the angular spectra become more dissimilar. Note that in the above formula (4), instead of the difference (P _n-m -P _{n-m-1 ), the difference (P n -P n-m-1} ) between the maximum peak angle P _n at any sampling time t _n and the maximum peak angles (P _{n-1 ,} ...P _{n-M )} at the past M sampling times (t _n - _{1 , ...t n-M} ) may be calculated.
Furthermore, the visibility determination unit 5 calculates the variance v(n) of the reception intensity similarity s(n) from an arbitrary sampling time _tn to the sampling time tn _-M+1 (M-1) times before, as shown in the following formula (5). The variance v(n) approaches 0 as the shapes of the angular spectra become more similar, and approaches 1 as they become more dissimilar.

Then, the visibility determining unit 5 calculates the visibility probability LOSValue of the following equation (6) based on the reception intensity similarity s(n) and the variance v(n).
As is clear from the following equation (6), the visibility LOSValue has a value in the range of 0 to 1, and the more similar the shapes of the angular spectra are, the closer it is to 1, and the more dissimilar it is, the closer it is to 0.

For example, the propagation time of radio waves from the transmitting terminal 2 to the antenna 3 may be estimated, an angle-time spectrum, which is the distribution of the received strength of the radio waves for each direction of arrival and each propagation time, may be calculated, and based on the time variation of the angle-time spectrum, it may be determined whether the environment between the transmitting terminal 2 and the antenna 3 is a line-of-sight environment or a non-line-of-sight environment.
The angular spectrum calculation unit 4 may estimate the propagation time by calculating the radio wave signal when it is assumed that the signal arrived at each of a plurality of different propagation times Tp ₁ , Tp ₂ , ..., and searching for the propagation time that has the highest correlation (similarity) with the actually received radio wave. Note that, considering that radio wave signals generally have an indeterminacy that makes them have the same value when the phase is shifted by 2π, it is desirable for a sufficient number of frequency components to be included in order to ensure the estimation accuracy.

Fig. 4(a) is a diagram showing an example of an angle-time spectrum when there is a non-line-of-sight environment between the transmitting terminal 2 and the antenna 3, and Fig. 4(b) is a diagram showing an example of an angle-time spectrum when there is a line-of-sight environment. The x-axis indicates the propagation time, the y-axis indicates the angle (the direction of arrival of the radio wave), and the z-axis indicates the reception strength.
Even if an x-axis relating to propagation time is added as shown in Figures 4(a) and 4(b), as with the two-dimensional graph of the angular spectrum shown in Figure 2, the time variation of the angular time spectrum and the visibility probability LOSValue can be calculated by calculating the correlation between different sampling times to determine whether the shape of the three-dimensional graph itself changes over time, and by calculating how much the angle and propagation time at the maximum peak have changed.

Alternatively, as can be seen from Fig. 4(a), when displayed in a three-dimensional graph, small mountains (local peaks) can be seen in a non-line-of-sight environment, but no particularly prominent peaks exist. This indicates that in a non-line-of-sight environment, a high-intensity direct wave is not received by the receiving device, and that roughly the same level of reflected waves are combined and received.
In contrast, a prominent peak can be seen in the line-of-sight environment of Figure 4(b). This is because in a line-of-sight environment, high-intensity direct waves are received, and when the received radio waves are processed, the characteristics of the direct waves appear in the graph.

As can be seen by comparing Figures 4(a) and 4(b), by adding the propagation time, more detailed information on the radio wave incidence is visualized. That is, when observing small peaks that exist locally, if most of them have the same intensity and no particularly prominent peak exists, it is inferred that the environment is non-line-of-sight. On the other hand, if a specific local peak has a prominent difference from the other peaks, it is inferred that the environment is line-of-sight.
For this reason, the sharpness of the peak may be calculated as the line-of-sight probability LOSValue, which indicates the likelihood of the environment being within line of sight, as shown in the following formula (7): In the following formula (7), SP[θ, t] is the reception strength at the angle θ and the propagation time t, and Tp is the maximum value in the range of the propagation time.

That is, the intensity is estimated for each angle (direction of arrival) and each propagation time, and normalization is performed so that the line of sight accuracy LOSValue is a value in the range of 0 to 1 to find the maximum value max(SP[θ,t]). In addition, the received intensity SP[θ,t] for each angle θ and each propagation time t is stored in an array. The average value of the differences between each element of the array and the maximum value max(SP[θ,t]) is then calculated to find the line of sight accuracy LOSvalue.

In addition, the time variation of the angular spectrum obtained as described above and the time variation of the angular-time spectrum may be combined to determine whether the environment between the transmitting terminal 2 and the antenna 3 is a line-of-sight environment or a non-line-of-sight environment. For example, whether the environment is a line-of-sight environment or a non-line-of-sight environment may be determined based on whether the sum, product, or average of the time variation of the angular spectrum and the time variation of the angular-time spectrum is less than a threshold value.
In addition, the visibility accuracy may be calculated by combining the visibility accuracy LOSValue(n) calculated by the above formula (3) and formula (6) with the visibility accuracy LOSValue calculated by the above formula (7). For example, the sum, product, or average of the accuracy LOSValue(n) and the visibility accuracy LOSValue may be calculated as the final visibility accuracy.

(Effects of the First Embodiment)
(1) The visibility condition determination device 1 includes an angular spectrum calculation unit 4 that calculates an angular spectrum, which is the distribution of the reception strength of radio waves for each direction of arrival of the radio waves, based on a received signal obtained by receiving radio waves transmitted from a transmitting terminal via an antenna 3, and a visibility determination unit 5 that determines that the environment between the transmitting terminal and the antenna 3 is a line-of-sight environment if the time fluctuation of the angular spectrum is small, and determines that the environment between the transmitting terminal and the antenna 3 is a non-line-of-sight environment if the time fluctuation is large.
This provides a means for determining, using radio waves, whether the environment between the transmitting terminal and the antenna 3 is within line of sight or not.

(2) The line-of-sight determination unit 5 may determine that the environment between the transmitting terminal and the antenna 3 is a line-of-sight environment when the time variation is less than a threshold, and may determine that the environment between the transmitting terminal and the antenna 3 is a non-line-of-sight environment when the time variation is equal to or greater than the threshold. This makes it possible to determine whether the environment between the transmitting terminal and the antenna 3 is a line-of-sight environment by using the threshold.
(3) The visibility determination unit 5 may calculate a visibility accuracy LOSValue indicating the accuracy that the environment between the transmitting terminal and the antenna 3 is within the line of sight. The visibility determination unit 5 may increase the visibility accuracy LOSValue when the time variation is small compared to when the time variation is large. This makes it possible to calculate a continuous accuracy LOSValue that indicates the likelihood of the environment being within the line of sight.

(4) The visibility determining unit 5 may determine that the lower the similarity between the angular spectra obtained from radio waves received at different times, the greater the time variation. Also, the visibility determining unit 5 may determine that the greater the change in the arrival direction of the maximum peak of the angular spectra obtained from radio waves received at different times, the greater the time variation.
This allows the magnitude of time fluctuation of the angular spectrum to be evaluated.

The angular spectrum calculation unit 4 may also estimate the propagation time of the radio waves from the transmitting terminal to the antenna 3 and calculate the angular time spectrum, which is the distribution of the reception strength of the radio waves for each direction of arrival and each propagation time of the radio waves, and the visibility determination unit 5 may determine that the lower the similarity between the angular time spectra obtained from the radio waves received at different times, the greater the time fluctuation. This allows the magnitude of the time fluctuation of the angular time spectrum to be evaluated as the magnitude of the time fluctuation of the angular spectrum.

Second Embodiment
Next, a second embodiment of the present invention will be described. As a preferred embodiment of the present invention, the visibility state determination device of the first embodiment is applied to a terminal holder detection system that estimates a person holding a transmitting terminal in a monitored space, and the monitored space is installed and operated in a highly confidential room. The description will be given with reference to the drawings. In this embodiment, the transmitting terminal is assumed to be a smartphone or a tablet terminal.

The reason for this is that in recent years, smartphones and tablet devices have become smaller and more powerful, making it possible to send large amounts of information with simple operations. As a result, even if such a device is concealed in a bag or clothing, it cannot be detected from the outside, and if it is brought into a room where confidential information is handled, there is a risk of information leaks, so it is necessary to discover and identify the owner.

Here, the terminal holder detection system of the second embodiment detects the positions of people in the room by placing multiple receiving devices and position sensors in the monitored space, and successively assumes each detected person to be a candidate for carrying a transmitting device, and calculates the theoretical value of the angular spectrum at the receiving device when radio waves are transmitted from the transmitting device carried by the candidate person. The system compares the measured angular spectrum with the theoretical value calculated for each person to see which is closest, and identifies the person carrying the transmitting device.

In this case, it is difficult to assume that the environment between all receiving devices and transmitting terminals is within line of sight, and the environment between some receiving devices and transmitting terminals is non-line of sight. In particular, if a transmitting terminal is held by a human, the human himself becomes an obstacle to radio waves, and the environment between the transmitting terminal and receiving device is likely to be non-line of sight.
In a receiving device in an out-of-line-of-sight environment, direct waves cannot be received, and instead, complex reflected waves from walls, furniture, etc. are combined and received, with the received strength changing according to the phase of the reflected waves contained in the combined wave. For this reason, there was a problem that calculating theoretical values for all phases would result in an excessively large calculation load, and modeling errors such as noise and people could lead to erroneous judgments (false reports).
Therefore, the terminal holder detection system of the second embodiment applies the visibility state determination device of the first embodiment to suppress the influence of the receiving device in a non-line-of-sight environment.

FIG. 5 is a schematic diagram showing how a terminal holder detection system according to the second embodiment of the present invention is installed and operated.
5 is a schematic diagram of a highly confidential room 20. There are three people in the room 20. In addition to a person 40 who has a calling terminal 41, for example a smartphone, hidden in his/her clothing, there are also a person 50 and a person 60 who do not have a calling terminal.
The purpose of the terminal holder detection system in this example is to detect a person 40 who is holding a calling terminal.

5, the room 20 is surrounded by walls 23, 24, 25, and 28 (not shown), and is further defined by a floor 21 and a ceiling 22. A window 27 and a door 26 also exist.
The walls and other surfaces are made using common building materials and are not perfect radio wave absorbers, but rather have the property of reflecting radio waves at least to a certain extent.
Although not shown, room 20 is assumed to be appropriately equipped with desks and chairs, as well as a safe and locked cabinets for storing important items such as documents and recording media containing confidential information. The above-mentioned internal structural information of room 20 is assumed to be expressed in the world coordinate system, known, and stored in the memory unit of the system main unit.

The room 20 is equipped with a position detection device 200. The position detection device 200 is a means for detecting the positions (coordinate values) of people present in the room 20, such as people 40, 50, and 60 in Fig. 5, expressed in a world coordinate system defined in the room 20. The detection results are used as information for calculating the theoretical value of the radio waves received by the receiving device.
The position detection device 200 is installed so as to be able to detect a person 40, etc., and is preferably installed in a corner of a room at approximately the height of the person's waist to shoulders, as shown in Fig. 5. A plurality of position detection devices 200 may be installed.
The position detection device 200 can be realized by a laser radar type distance sensor. Alternatively, an image sensor (camera) equipped with an image processing means may be used, or pressure sensors may be installed all over the surface under the floor surface 21 to detect the position based on the weight of a person.

Receiving devices 300 to 303 are antenna devices installed on ceiling surface 22 of room 20 for receiving radio waves. Each of receiving devices 300 to 303 includes an antenna array and angular spectrum calculation unit 4 of the first embodiment, and detects the reception strength of radio waves transmitted from transmitting terminal 41 as well as the direction of arrival, and calculates the actual measured value of the angular spectrum described with reference to FIG.
Furthermore, under the condition that the radio waves transmitted from the transmitting terminal 41 have a certain bandwidth, the receiving devices 300 to 303 may estimate the propagation time from the transmitting terminal 41 to the receiving devices 300 to 303, and calculate the actual measured values of the angle-time spectrum described with reference to Figures 4(a) and 4(b).
The direction of arrival of radio waves for each receiving device is defined as 0° based on a specific direction, looking down from the ceiling to the floor, and 360° counterclockwise. However, when the receiving device is installed in a corner of the ceiling as in Fig. 5, the front direction may be defined as 0° based on the direction of arrival, and the range may be defined as -45° to +45°.

6 is a block diagram showing a schematic configuration of an example of a terminal holder detection system according to the second embodiment. The terminal holder detection system 10 is made up of the receiving devices 300 to 303 and the position detection device 200 already described, as well as a system main device 100.
The communication means connecting these devices can be realized by a wired method conforming to standards such as Ethernet (registered trademark), RS232C, RS485, or a wireless method using Wi-Fi (registered trademark), Bluetooth (registered trademark), etc., so explanation is omitted.

The system main unit 100 includes a storage unit 110, a model generation unit 115, a propagation information calculation unit 120, a theoretical value buffer 140, an actual measurement value buffer 150, a comparison and determination unit 160, an output unit 170, the visibility determination unit 5 of the first embodiment, and a weight calculation unit 190. Each unit will be described below.
The storage unit 110 is a storage means realized by known means such as a magnetic medium such as a HDD or a semiconductor memory. In addition to a program that controls the operation of the entire system main unit 100, it stores program modules that configure each part of the system main unit 100, parameters such as thresholds, an area for temporarily storing data during processing, structural information 111, and a human model 112. The information stored in the storage unit 110 is exchanged with each part of the system main unit 100 as appropriate.

The structural information 111 is information on elements that define the room 20, and is used to obtain theoretical values of reception strength in each receiving device. That is, a coordinate system (world coordinate system) is defined with the corner of the floor as the origin, and the information is expressed by a modeling method such as BIM (Building Information Model), including geometric information such as the size and position of the walls 23 to 28, the ceiling surface 22, the floor surface 21, the window 27, the door 26, and fixtures (not shown), mathematical formula information defining the shape of the surfaces of the walls, surface material information, and radio wave reflectance.
The human model 112 is information that expresses the three-dimensional shape of a person who may exist in the room 20 using a polygon model, a surface model, or the like, and is a model that expresses at least the height (170 cm) and width (60 cm) of a standard human build. The human model 112 has a reflectivity of radio waves on its body surface.

The model generation unit 115 refers to the structural information 111, virtually places a person model 112 at the position detected by the position detection device 200 for each person present in the room 20, generates a detection model that models the positional relationships of the moving persons inside the room 20, and outputs the detection model to the propagation information calculation unit 120.
The propagation information calculation unit 120 obtains theoretical values of the path along which the radio wave transmitted from the transmitting terminal 41 can propagate to each receiving device and the strength at which the radio wave is received. To this end, the propagation information calculation unit 120 has a path calculation unit 125 and a theoretical reception strength value calculation unit 130.

The path calculation unit 125 appropriately refers to the detection model generated by the model generation unit 115, the structural information 111 stored in the storage unit 110, and the person model 112, and sets each person present in the room 20 as a candidate, assuming that they are carrying the transmitting terminal 41, and calculates the propagation path of the radio waves from their nearby positions, taking into account reflections on walls and obstructions by other people. The calculation of the propagation path is based on a method similar to the ray tracing method well known in the field of computer graphics.

The theoretical reception strength calculation unit 130 sequentially selects the people detected by the position detection device 200 as candidates assumed to be carrying the transmitting terminal 41, and calculates the theoretical value of the angular spectrum of the radio waves at each receiving device from the information on the propagation path calculated by the path calculation unit 125. This will be explained using Figures 7(a) to 7(d).

7(a) to 7(d) are schematic diagrams showing the case where person 40 has transmitting terminal 41 and person 50 and person 60 do not, and show the theoretical values of the angular spectrum calculated by theoretical receiving strength calculation unit 130 for radio waves received by receiving devices 300, 301, 302, and 303, respectively.
In addition, the theoretical value when person 50 has transmitting terminal 41 and person 40 and person 60 do not, and the theoretical value when person 60 has transmitting terminal 41 and person 40 and person 50 do not are also calculated in the same manner, but are not shown in the figures.

As shown in Figures 7(a) to 7(d), the theoretical reception strength calculation unit 130 calculates, for each receiving device, the theoretical value of the strength distribution of the radio waves arriving from the direction for each angle defined for the receiving device, as the theoretical value of the angular spectrum.
For this purpose, the theoretical reception strength calculation unit 130 considers the degree of attenuation of radio waves in the air, and also considers that attenuation occurs with each reflection according to the reflectance determined from the material and small irregularities of the surfaces of the walls, ceiling 22, and floor 21 stored as the structural information 111 in the storage unit 110, and the reflectance on the body surface of each person, and calculates a theoretical reception strength corresponding to the propagation distance calculated along the propagation path calculated by the path calculation unit 125 for each arrival angle of the radio waves. That is, the theoretical reception strength calculation unit 130 reads out the reflectance stored in the structural information 111 and the person model 112, and determines that the smaller the reflectance, the more difficult it is for radio waves to reach long distances and that they are received with weaker strength.

Returning to the block diagram of Fig. 6, the theoretical value buffer 140 is a buffer that temporarily stores the theoretical value of the angular spectrum for each receiving device together with time information calculated by the theoretical reception intensity value calculation unit 130. It can be realized as appropriate using a semiconductor memory or a magnetic disk, but it may be integrated with the storage unit 110 by reserving a certain area in the storage unit 110 rather than being an independent element.
The actual measurement value buffer 150 is a buffer that temporarily stores the actual measurement value of the angular spectrum transmitted from each receiving device together with time information. It can be realized by a semiconductor memory or a magnetic disk as appropriate, but it may be realized by allocating a certain area in the storage unit 110 and integrating it with the storage unit 110 instead of being an independent element.
8(a) to 8(d) are schematic diagrams showing actual measured values of the angular spectrum stored in the actual measured value buffer 150 at a certain time.

The visibility determination unit 5 uses the actual measured values of the angular spectrum at each time stored in the actual measured value buffer 150 for the receiving devices 300 to 303 to calculate a visibility probability LOSValue indicating that the transmitting terminal 41 is likely to be in a visibility state from the receiving device.
The method of calculating the visibility accuracy LOSValue is the same as that in the first embodiment. For example, any M (M is a natural number) actual measurement values of the angle spectra among the angle spectra for a predetermined time (i.e., the angle spectra acquired at a predetermined number of consecutive sampling times (t _n , t _n-1 , ... t _n-N : n and N are natural numbers)) may be read from the actual measurement value buffer 150, and the visibility accuracy LOSValue may be calculated based on the read actual measurement values.

The actual measured values of the angular spectrum to be read from the actual measured value buffer 150 are selected and read from a time width in which the change in visibility caused by the movement of the people 40 to 60, who are moving bodies, can be ignored.
The time span over which the actual measured values of the angular spectrum are read out from the actual measured value buffer 150 (that is, the time span from time t _n−N to time t _n ) may be dynamically changed according to the moving speed of the persons 40 to 60 .

The weight calculation unit 190 determines a weight to be used in the comparison and determination unit 160 (described later) to determine which person is carrying the calling terminal, based on the visibility probability LOSValue calculated by the visibility determination unit 5 .
Since the visibility accuracy LOSValue is defined to have a value closer to 1 as the visibility state becomes more likely, the visibility accuracy LOSValue may be used as is for simplicity. Alternatively, the weight may be obtained by defining a monotonically increasing function with the visibility accuracy LOSValue as an independent variable and the weight as a dependent variable.

The comparison and determination section 160 compares the theoretical values of the angular spectrum stored in the theoretical value buffer 140 with the actual measured values of the angular spectrum stored in the actual measured value buffer 150 for the same time for each receiving device.
Explaining this with reference to the schematic diagrams of theoretical values in Figures 7(a) to 7(d), the comparison and determination unit 160 first reads out the actual measurement values of the receiving device 300 from the actual measurement value buffer 150, and obtains a similarity indicating the degree of similarity with each of Figures 7(a) to 7(d). This similarity can be obtained by normalized correlation.
The comparison and determination unit 160 then similarly calculates the similarities for the other receiving devices 301 to 303, calculates an evaluation value indicating which person has the transmitting terminal, and determines the person as the owner of the transmitting terminal. For example, the comparison and determination unit 160 determines the person with the largest evaluation value calculated as the owner of the transmitting terminal.
At this time, the comparison and determination section 160 multiplies the similarity by the weight calculated by the weight calculation section 190 to obtain an evaluation value.

Next, a terminal holder detection method according to a second embodiment of the present invention will be described with reference to the flowchart shown in FIG.
5 shows an example in which three people 40-60 are present in the room 20, but the number of people is arbitrary, and the number of people carrying the transmitting terminal 41 is also arbitrary. Similarly, the number of receiving devices 300-303 is arbitrary, but it is desirable to place two or more receiving devices in order to ensure the accuracy of the determination. In the following embodiment, the number of people carrying the transmitting terminal is described as one.

Before executing the following process, the internal structure of the room 20 (the installation positions of the position detection sensors, the installation positions of the receiving devices, the monitoring area, and geometric information of objects already present in the monitoring area) and their reflectance (information on the material (reflection characteristics) of stationary objects such as walls, floors, and furniture) are set in the structural information 111 of the memory unit 110 as initial settings, and a human model 112 is also set. And it is assumed that each component of the main device is in a state where it can operate appropriately.

In step S100, the position detection device 200 detects the positions of objects, particularly people, present in the room 20 in the world coordinate system, and transmits the detected positions to the propagation information calculation unit 120 of the system main unit 100. When multiple position detection devices 200 are used, the detection results for the same person are compiled from the output results of each device, so as to prevent the same person from being erroneously detected as multiple people.
Note that even if an object is detected, it may not be transmitted if the detection result indicates that the vehicle is unattended, such as when the object merely matches an object stored in the structural information 111. Alternatively, the propagation information calculation unit 120 may determine that the vehicle is unattended and not perform the process described below.
Here, as shown in FIG. 10A, a scene is assumed in which three people 40 to 60 are present in a room 20 and the position detection device 200 detects the positions of each of the people 40 to 60.

In step S110, each of the receiving devices 300 to 303 measures the reception strength for each direction of arrival defined in the world coordinate system to obtain an actual measurement value. The result is transmitted to the system main unit 100, which stores it in the actual measurement value buffer 150.
Each of the receiving devices 300 to 303 is equipped with an array antenna, and receives signals simultaneously with each antenna, and measures the angle of arrival using the beamformer method, the MUSIC (Multiple Signal Classification) method, or the like.

The measured values of the reception strength are stored for each receiving device in the form of an angular spectrum. That is, the values are stored as values expressed in a predetermined unit, for example, decibels (dB), for each arrival angle. In addition to the most recent measurement, a certain amount of past measurement is also stored. When the most recent measurement is obtained, the oldest measurement is sequentially erased.
If all of the measured reception strength values are low and are at a level where noise is being received, no further processing is carried out (step S120).

In step S125, the model generation unit 115 refers to the structural information 111 and the person model 112 stored in the memory unit 110, creates a detection model from the positions of each person in the room 20 measured by the position detection device 200 in step S100, and outputs the detection model to the propagation information calculation unit 120.
Steps S130 to S138 described below are performed by sequentially treating each person detected by the position detection device 200 in step S100, that is, persons 40 to 60 as shown in FIG.

In step S130, the route calculation unit 125 and the theoretical reception strength calculation unit 130 of the propagation information calculation unit 120 calculate, for each of the receiving devices 300 to 303, the theoretical value of the reception strength that would be measured if the transmitting terminal 41 were carried by a candidate, and obtain the theoretical value of the angular spectrum and store it in the theoretical value buffer 140.
First, the route calculation unit 125 sequentially selects people detected by the position detection device 200 as candidates who possess a transmitting terminal, and by referring to the detection model generated in step S125, determines the propagation routes along which radio waves can propagate from the candidates to each receiving terminal using a ray tracing method.
The path calculation unit 125 outputs each piece of propagation path information thus obtained to the theoretical reception strength value calculation unit 130 .

The theoretical values are obtained for each of the receiving devices 300-303 for each of the persons 40-60 and the way in which the transmitting terminal 41 is held. Fig. 10(b) shows examples of holding positions P1-P4 assumed when the persons 40-60 hold the transmitting terminal 41. Holding positions P1, P2, P3, and P4 are the front, left side, right side, and back of the persons 40-60, respectively.
For example, if there are three people 40 to 60, there are four ways of holding the device as shown in FIG. 10(b), and four receiving devices 300 to 303 are arranged as shown in FIG. 5, 3×4×4=48 theoretical values are calculated.

In step S135, the visibility determining unit 5 uses the actual measurement values stored in the actual measurement value buffer 150 to calculate a visibility probability LOSValue that indicates the possibility that the environment between the receiving devices 300 to 303 and the transmitting terminal 41 is within the line of sight.
In step S138, the weight calculation unit 190 calculates the weight using the visibility probability LOSvalue calculated by the visibility determination unit 5 in step S135.
In step S140, the comparison and determination unit 160 compares the actual measured value of the angular spectrum stored in the actual measured value buffer 150 with the theoretical value of the angular spectrum stored in the theoretical value buffer 140, and further uses the weight calculated by the weight calculation unit 190 to determine which of the persons detected by the position detection device 200 is holding the transmitting terminal, and outputs the result to the output unit 170.

The comparison and determination unit 160 calculates the similarity for each receiving device and each position where the transmitting terminal is held (Figure 10(b)) for the actual measured values (Figures 8(a) to 8(d)), the theoretical values when person 40 is a candidate (Figures 7(a) to 7(d)), the theoretical values when person 50 is a candidate (not shown), and the theoretical values when person 60 is a candidate (not shown), and stores the similarity in the memory unit 110.

11A shows the calculation results of the similarity stored in the temporary area of the storage unit 110 for the receiving device 300 and the receiving device 302. The same is true for the receiving device 301 and the receiving device 303, so they are not shown.
The similarity shown in Fig. 11(a) is a value before being weighted by the weight calculated by the weight calculation unit 190. Referring to this value, the similarity is maximum when the person 50 holds the transmitting terminal 41 at the holding position P4, so if the holder of the transmitting terminal 41 is estimated based on the similarity before weighting, the result that the person 50 holds it is obtained. However, as shown in Fig. 10(a) as a schematic, the result is incorrect because the person 40 holds it. This is because the line-of-sight accuracy LOSValue of the receiving device 300 is low at "0.56", so there is a high possibility that the environment between the receiving device 300 and the transmitting terminal 41 is non-line-of-sight.

Therefore, the comparison and determination unit 160 uses the visibility LOSValue as the weighting factor as it is, and multiplies each similarity by the weighting factor obtained for each of the receiving devices 300 to 303. In other words, each similarity is weighted by the weighting factor. The similarity after weighting is shown in FIG. 11(b).
11B, the weighted similarity is maximized for the combination of the receiving device 302, the person 40, and the carrying position P1. Since the visibility LOSValue of the receiving device 302 is high at "0.86" and the receiving device 302 is in a line-of-sight environment with respect to the transmitting terminal 41, it can be correctly determined that the person 40 is carrying the transmitting terminal 41.

In step S150, the output unit 170 outputs the information on the owner of the calling terminal input from the comparison and determination unit 160 to the outside. As already mentioned, an external device (not shown) performs a notification process to alert the owner of the calling terminal, etc.

In the above explanation, in order to succinctly explain the effect of weighting, it has been explained how weighting can correctly identify person 40, the terminal owner, when the LOSValue value is used as the weight for each similarity and the similarity is judged based on its level.
Alternatively, the similarity calculated for each of the persons 40 to 60 and each of the carrying positions P1 to P4 may be integrated for the receiving devices 300 to 303. That is, the similarity calculated for each of the receiving devices 300 to 303 may be integrated. For example, the sum of the similarity calculated for each of the receiving devices 300 to 303 may be calculated to determine whether or not the person is the owner of the transmitting terminal 41.

Fig. 12(a) shows the similarity obtained by integrating the similarity for each receiving device shown in Fig. 11(a) (i.e., the similarity before weighting) for each person. In this case, the sum of the similarities of the receiving devices 301 and 302 when the person 60 holds the transmitting terminal 41 at the holding position P1 (0.67 + 0.65 = 1.32) is the maximum, and the person 60 is erroneously determined to be the holder.
In contrast, Fig. 12(b) shows a similarity obtained by integrating the similarities (i.e., weighted similarities) for each receiving device shown in Fig. 11(b) for each person. In this case, the sum of the similarities (0.20 + 0.86 = 1.06) of receiving devices 301 and 302 when person 40 holds transmitting terminal 41 at holding position P1 is the maximum, and person 40 is correctly determined to be the holder.

As another method for integrating the similarities, the sum of the similarities for the receiving devices may be calculated, and then the sum may be divided by the number of receivers to calculate the average value.
In addition, the angular spectrum of a receiving device for which a line-of-sight accuracy LOSValue that can be considered to be in a clearly non-line-of-sight environment is calculated may be excluded from the processing of the above comparison and determination unit 160, and the terminal owner may be determined based on the actual measurement value of the angular spectrum of a receiving device considered to be in a line-of-sight environment (for example, a receiving device for which a line-of-sight accuracy LOSValue equal to or greater than a threshold value is calculated). This makes it possible to suppress the influence of a receiving device in a non-line-of-sight environment.

(Effects of the Second Embodiment)
(1) The terminal holder detection system includes an antenna installed in a monitored space for receiving radio waves from a transmitting terminal carried by a moving person within the monitored space, a position detection device 200 for detecting the position of each moving person, an angular spectrum calculation unit 4 for calculating an actual measured value of an angular spectrum, which is the distribution of the reception strength of radio waves for each direction of arrival of the radio waves, based on a received signal obtained by receiving radio waves transmitted from the transmitting terminal by the antenna, a theoretical value buffer 140 for storing, for each antenna, a theoretical value of the angular spectrum when radio waves transmitted from near the position of the moving person are received by the antenna, assuming that the moving person is carrying a transmitting terminal, a comparison decision unit 160 for calculating and integrating the similarity between the theoretical value and the actual measured value for each antenna and deciding whether or not the moving person is carrying a transmitting terminal based on the integrated similarity, and a line-of-sight determination unit 5 for determining that the environment between the transmitting terminal and the antenna is within line-of-sight if the actual measured value of the angular spectrum has a small time fluctuation, and for determining that the environment between the transmitting terminal and the antenna is beyond line-of-sight if the time fluctuation is large.
The visibility determination unit 5 calculates a visibility accuracy LOSValue indicating the accuracy that the environment between the transmitting terminal and the antenna is within the line of sight, and increases the visibility accuracy LOSValue when the time fluctuation is small compared to when the time fluctuation is large, and the comparison determination unit 160 weights the similarity according to the visibility accuracy.

As a result, even if multiple people are present in the monitored area, it is possible to determine which person is the holder of the transmitting terminal by taking into consideration the visibility as seen from the antenna.
In particular, this is highly effective in situations where the owner conceals the transmitting terminal, as well as when someone is operating the terminal secretly from a person with authority to prevent confidential information leaks, in an attempt to leak information. When operating a transmitting terminal, it emits radio waves of higher intensity and for a longer period of time than when in normal communication standby mode, but when operating the terminal secretly, the user is likely to be in a position where they are holding the terminal to their body, resulting in an environment where the line of sight is not visible.

(2) The comparison and determination unit 160 may use an antenna whose measured angular spectrum varies over time less than a predetermined variation to determine whether or not the mobile person is carrying the transmitting terminal 41. This can reduce the influence of an antenna in a non-line-of-sight environment.

(3) Note that, when the propagation time from the transmitting terminal 41 to the antenna (i.e., the propagation time from the receiving devices 300 to 303) is estimated and the angle-time spectrum is calculated as in the first embodiment, it is possible to determine the visibility state for each measurement time. In addition, if the determination is made based on time fluctuations without estimating the propagation time, the effect of noise can be suppressed.

(others)
The application of the present invention is not limited to rooms with high confidentiality. For example, by referring to the authority information of each person in the monitored space, if a person who is not authorized to have a transmitting terminal is carrying a transmitting terminal, the present invention may be applied to a purpose of issuing a warning, temporarily holding the transmitting terminal, or requesting that the terminal be turned off.
For example, it has been pointed out that radio waves from devices other than those carried by doctors and nurses in hospitals can affect medical equipment. For this reason, hospitals can ask patients who are carrying or operating devices to turn them off.
In this case, such authority settings can be varied for each room (monitored space). For example, in the same hospital example, it is possible to set it so that only patients are not allowed to possess or operate the device in the waiting room, but even doctors and nurses are not allowed to do so in the examination room.

If we apply this to a highly confidential data center, we can implement an operation that allows employees working at the data center to possess and operate the devices, but not visitors.
This makes it possible to prevent confidential information from leaking by simply installing a receiving device in the monitored space and a system main unit in the security room, without having to take strict measures such as installing gate-type metal detectors to prevent confidential information from leaking.

1... visibility state determination device, 2, 41... transmitting terminal, 3... antenna, 4... angular spectrum calculation unit, 5... visibility determination unit, 10... terminal holder detection system, 20... room, 21... floor surface, 22... ceiling surface, 23, 24, 25, 28... wall surface, 26... door, 27... window, 40, 50, 60... person, 41... transmitting terminal, 100... system main unit, 110... storage unit, 111... structural information, 112... person model, 115... model generation unit, 120... propagation information calculation unit, 125... path calculation unit, 130... reception strength theoretical value calculation unit, 140... theoretical value buffer, 150... actual measurement value buffer, 160... comparison and determination unit, 170... output unit, 190... weight calculation unit, 200... position detection device, 300-303... receiving device

Claims (5)

an angular spectrum calculation unit that calculates an angular spectrum, which is a distribution of the reception strength of a radio wave for each direction of arrival of the radio wave, based on a reception signal obtained by receiving the radio wave transmitted from a transmitting terminal by an antenna;
a visibility determination unit that determines that an environment between the transmitting terminal and the antenna is a line-of-sight environment when the time variation of the angular spectrum is small, and determines that an environment between the transmitting terminal and the antenna is a non-line-of-sight environment when the time variation of the angular spectrum is large;
Equipped with
The visibility determination unit determines that the lower the similarity between shapes of the angular spectra obtained from the radio waves received at different times, the greater the time variation . an angular spectrum calculation unit that calculates an angular spectrum, which is a distribution of the reception strength of a radio wave for each direction of arrival of the radio wave, based on a reception signal obtained by receiving the radio wave transmitted from a transmitting terminal by an antenna;
a visibility determination unit that determines that an environment between the transmitting terminal and the antenna is a line-of-sight environment when the time variation of the angular spectrum is small, and determines that an environment between the transmitting terminal and the antenna is a non-line-of-sight environment when the time variation of the angular spectrum is large;
Equipped with
the angular spectrum calculation unit estimates a propagation time of the radio wave from the transmitting terminal to the antenna, and calculates an angular time spectrum which is a distribution of the reception strength of the radio wave for each direction of arrival and each propagation time of the radio wave;
The visibility determination unit determines that the lower the similarity between shapes of the angle-time spectra obtained from the radio waves received at different times, the greater the time variation. 3. The visibility state determination device according to claim 1, wherein the visibility determination unit determines that the time variation is greater as the change in the arrival direction of the maximum peak of the angular spectrum obtained from the radio waves received at different times becomes greater. An antenna installed in the monitored space for receiving radio waves from a transmitting terminal carried by a person moving within the monitored space;
a position detection device for detecting the position of each of the moving persons;
an angular spectrum calculation unit that calculates an actual measurement value of an angular spectrum, which is a distribution of the reception strength of the radio wave for each direction of arrival of the radio wave, based on a reception signal obtained by receiving the radio wave transmitted from the transmitting terminal by the antenna;
a storage unit that stores, for each antenna, a theoretical value of the angular spectrum when a radio wave transmitted from a vicinity of a location of the moving person is received by the antenna, assuming that the moving person carries the transmitting terminal;
a comparison and decision unit that calculates and integrates a similarity between the theoretical value and the actual measurement value for each of the antennas, and decides whether or not the mobile person carries the transmitting terminal based on the integrated similarity;
a line-of-sight determination unit that determines that an environment between the transmitting terminal and the antenna is a line-of-sight environment when a time variation of the actual measured value of the angular spectrum is small, and that an environment between the transmitting terminal and the antenna is a non-line-of-sight environment when the time variation is large;
Equipped with
the visibility determination unit calculates a visibility accuracy indicating an accuracy that an environment between the transmitting terminal and the antenna is within the line of sight, and increases the visibility accuracy when the time variation is small compared to when the time variation is large;
The comparison determination unit weights the similarity according to the visibility accuracy.
A terminal holder detection system comprising: The terminal holder detection system according to claim 4 , characterized in that the comparison and decision unit determines whether the moving person is carrying the transmitting terminal by using the antenna for which the time variation of the actual measured value of the angular spectrum is less than a threshold value.

Images