JP2015040774A - Monitoring device and monitoring program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a monitoring device capable of suppressing erroneous detection, and improving detection accuracy, in a constitution for monitoring human motion in a prescribed area by using a radio wave, and to provide a monitoring program.SOLUTION: A monitoring device includes a receiving part for receiving a radio wave transmitted from a transmitter arranged in a prescribed area, a feature quantity calculation part for calculating feature quantity of the radio wave in each of a plurality of object periods received by the receiving part, and a monitoring part for monitoring human motion in the prescribed area based on the radio wave in the object period in which the feature quantity satisfies a prescribed condition.

Description

  The present invention relates to a monitoring device and a monitoring program, and more particularly, to a monitoring device and a monitoring program for monitoring human actions using radio waves.

  Intrusion detection devices that detect human movements in a predetermined area such as a room have been developed. As an example of the intrusion detection method, for example, “Study on indoor intruder detection by UWB-IR” Teiji Sakaji et al., IEICE Transactions B, Vol. J90-B, No. 1, 99.97-100, 2007 On January 1, 2000 (Non-Patent Document 1), a method using a propagation delay profile, that is, a power delay profile based on UWB-IR (Ultra Wide Band-Impulse Radio) is disclosed.

  However, in the method described in Non-Patent Document 1, interference with other wireless services becomes a problem because a wideband signal is used, and because the received signal power is used, it is affected by indoor multipath fading, The detection accuracy may deteriorate.

  As a technique for solving such a problem, for example, Japanese Patent Laid-Open No. 2008-216152 (Patent Document 1) discloses the following configuration. That is, the event detection device calculates an eigenvector, that is, an arrival angle distribution based on the received signal of each array antenna, and calculates an inner product value of the eigenvector and a normal eigenvector as a comparison reference. Then, the event detection device detects an event, that is, an intruder, based on a comparison result between the inner product value and a predetermined threshold value.

"Examination of indoor intruder detection by UWB-IR" Terasaka Junji et al., IEICE Transactions B, Vol. J90-B, No. 1, 99.97-100, January 1, 2007 IEEE Standard for Local and metropolitan area networks Part 15.4: Low-Rate Wireless Personal Area Network Networks (LR-WPANs), LAN / MAN Standard Co., Ltd. Takayuki Hirano and two others, “Examination of Transient Response Patterns in Radio Identification”, Communications Research Laboratory Quarterly, 2002, Vol. 48, No. 1, p. 149-154

JP 2008-216152 A JP 2002-26826 A

  The event detection device described in Patent Document 1 simultaneously receives an intrusion detection radio wave transmitted from a predetermined transmitter and a radio wave transmitted from another device in a state where the event detection device is installed in a predetermined area such as a room. There is a case.

  In this case, since the event detection apparatus processes radio waves including interference waves, the following problems may occur due to erroneous detection that erroneously detects the occurrence of an event. That is, for example, in an event detection device used for intrusion detection, even though no abnormality has actually occurred in a predetermined area, it may be determined to be abnormal due to erroneous detection, and erroneous reporting may occur.

  The present invention has been made to solve the above-described problems, and an object of the present invention is to suppress false detection and improve detection accuracy in a configuration in which human action in a predetermined area is monitored using radio waves. It is to provide a possible monitoring device and monitoring program.

  (1) In order to solve the above problem, a monitoring device according to an aspect of the present invention includes a receiving unit that receives a radio wave transmitted from a transmitter disposed in a predetermined area, and a plurality of receiving units that are received by the receiving unit. A feature amount calculation unit for calculating the feature amount of the radio wave in each of the target periods, and monitoring for monitoring human movements in the predetermined area based on the radio wave in the target period in which the feature amount satisfies a predetermined condition A part.

  (11) In order to solve the above problem, a monitoring program according to an aspect of the present invention is a monitoring program used in a monitoring device, and transmits a radio wave transmitted from a transmitter arranged in a predetermined area to a computer. A step of receiving, a step of calculating a feature quantity of the radio wave in each of a plurality of received target periods, and a step in the predetermined area based on the radio wave in the target period in which the calculated feature quantity satisfies a predetermined condition And a step for monitoring a human action.

  The present invention can be realized not only as a monitoring apparatus including such a characteristic processing unit, but also as a monitoring method using such characteristic processing as a step. Also, it can be realized as a semiconductor integrated circuit that realizes part or all of the monitoring device, or can be realized as a monitoring system including the monitoring device.

  ADVANTAGE OF THE INVENTION According to this invention, in the structure which monitors the human motion in a predetermined area using an electromagnetic wave, a misdetection can be suppressed and a detection precision can be improved.

FIG. 1 is a diagram showing a usage image of the intrusion detection system according to the first embodiment of the present invention. FIG. 2 is a diagram showing the configuration of the intrusion detection system according to the first embodiment of the present invention. FIG. 3 is a diagram illustrating an example of a frame transmitted by the transmitter according to the first embodiment of the present invention. FIG. 4 is a diagram showing a configuration of a receiver in the intrusion detection system according to the first embodiment of the present invention. FIG. 5 is a diagram illustrating an example of a sampling period set by the receiver according to the first embodiment of the present invention. FIG. 6 is a diagram illustrating another example of the sampling period set by the receiver according to the first embodiment of the present invention. FIG. 7 is a diagram illustrating a configuration of an interference canceller in the receiver according to the first embodiment of the present invention. FIG. 8 is a diagram illustrating an example of a time change of the target signal processed by the feature amount calculation unit according to the first embodiment of the present invention. FIG. 9 is a diagram illustrating an example of a frequency spectrum of a target signal processed by the feature amount calculation unit according to the first embodiment of the present invention. FIG. 10 is a flowchart defining an operation procedure when the receiver according to the first embodiment of the present invention monitors human motion. FIG. 11 is a diagram showing a configuration of a receiver in the intrusion detection system according to the second embodiment of the present invention. FIG. 12 is a diagram illustrating a configuration of an interference canceller in the receiver according to the second embodiment of the present invention. FIG. 13 is a diagram illustrating a configuration of an interference canceller in the receiver according to the second embodiment of the present invention. FIG. 14 is a diagram illustrating a configuration of an interference canceller in the receiver according to the second embodiment of the present invention. FIG. 15 is a diagram illustrating a configuration of a modification of the receiver illustrated in FIG. 11. FIG. 16 is a diagram illustrating a configuration of a modified example of the interference removing unit illustrated in FIG. 12. FIG. 17 is a diagram illustrating a configuration of a modified example of the interference removing unit illustrated in FIG. 13. FIG. 18 is a diagram illustrating a configuration of a modified example of the interference removing unit illustrated in FIG. 14.

  First, the contents of the embodiment of the present invention will be listed and described.

  (1) A monitoring device according to an embodiment of the present invention includes a receiving unit that receives a radio wave transmitted from a transmitter arranged in a predetermined area, and each of a plurality of target periods received by the receiving unit. A feature amount calculation unit that calculates a feature amount of the radio wave; and a monitoring unit that monitors a human operation in the predetermined area based on the radio wave in the target period in which the feature amount satisfies a predetermined condition.

  With such a configuration, for example, a radio wave that does not include an interference wave can be selected from radio waves in each of a plurality of target periods based on the calculated feature amount. It is possible to suppress false detection caused by processing the above and improve detection accuracy.

  As a result, it is possible to correctly monitor human movements in a predetermined area, so that when the monitoring device is used for intrusion detection, it is possible to avoid the occurrence of false alarms, and when the monitoring device is used for watching and monitoring purposes. In this case, it is possible to avoid the occurrence of misreporting.

  (2) Preferably, the feature amount calculation unit calculates, as the feature amount, a value representing a relationship between the radio wave and the reference radio wave in the target period.

  With such a configuration, the relationship between the radio wave and the reference radio wave in the target period can be quantified. Accordingly, for example, when a radio wave that does not include an interference wave is used as a reference radio wave, a radio wave that does not include an interference wave can be selected from radio waves in each of a plurality of target periods.

  (3) More preferably, the feature quantity calculation unit calculates, as the feature quantity, an inner product value of the eigenvector of the radio wave and an eigenvector of the reference radio wave in the target period or a value based on the inner product value.

  In this way, by calculating the inner product value that changes according to the relationship between radio waves or a value based on the inner product value as a feature amount, the relationship between the radio wave and the reference radio wave in the target period can be quantified with a simple calculation. Therefore, it is possible to easily select a radio wave that does not include an interference wave from radio waves in each of a plurality of target periods.

  (4) More preferably, the reference radio wave is a radio wave received by the receiving unit in a period later in time than the plurality of target periods.

  With such a configuration, the reference radio wave handled as a reference can be sequentially changed to a newer radio wave, so that it is possible to flexibly cope with changes in the radio wave environment of the monitoring device.

  (5) More preferably, the reference radio wave is a radio wave received by the receiving unit in a period in which the predetermined area is in a predetermined state before the plurality of target periods.

  As described above, the configuration using the reference radio wave serving as a specific standard can suppress inappropriate use of the reference radio wave, and can stably calculate the feature amount.

  As a result, radio waves that do not include interference waves can be accurately selected from radio waves in each of a plurality of target periods based on absolute evaluation, so that the reliability of the monitoring operation can be increased.

  (6) Preferably, the feature quantity calculation unit calculates a value representing a relationship between the radio waves in two target periods as the feature quantity, and the feature quantity calculation unit includes three or more target periods. A value representing the relationship is calculated for all combinations of the radio waves in the two target periods.

  As described above, the configuration in which the reference radio wave is not set can avoid an unstable operation of the monitoring device that occurs when the set reference radio wave is inappropriate. Further, for example, even when changing a predetermined target area, it is not necessary to newly set a reference radio wave, so that the processing in the monitoring device can be simplified.

  In addition, since the three or more target periods are sequentially updated, it is possible to flexibly cope with changes in the radio wave environment of the monitoring device.

  In addition, since the relationship between radio waves in two target periods can be quantified, radio waves that do not include interference waves among radio waves in each of three or more target periods can be easily obtained using the quantified relationship. Can be selected.

  (7) More preferably, the monitoring unit performs human action in the predetermined area based on the radio wave in the target period in which the degree of coincidence with the reference radio wave is the largest among the plurality of target periods. Monitor.

  With such a configuration, it is possible to select the target period that has the closest relationship with the reference radio wave from among the radio waves in each of the plurality of target periods. It is possible to correctly monitor human movements in the area.

  In addition, since the value representing the relationship between the radio wave and the reference radio wave in the target period that has the highest degree of coincidence with the reference radio wave can be temporally smoothed in the plurality of target periods including the target period. The occurrence of false detection can be suppressed.

  (8) Preferably, the feature amount calculating unit calculates a statistical value of the received power of the radio wave in the target period in the monitoring device as the feature amount.

  In this way, the above statistical value that changes according to the degree of interference waves included in the radio waves in the monitoring device is calculated as a feature amount, so that the interference waves are included in the radio waves in each of a plurality of target periods. A low radio wave can be selected.

  (9) Preferably, the feature amount calculation unit calculates the feature amount from the frequency distribution of the radio wave in the target period received by the reception unit.

  As described above, the feature amount is calculated from the frequency spectrum that changes in accordance with the degree of the interference wave included in the radio wave in the target period received by the receiving unit. It is possible to select radio waves that are low enough to contain waves.

  (10) Preferably, the monitoring unit monitors a human operation in the predetermined area based on the radio wave in the target period in which the feature amount is closest to a predetermined value among the plurality of target periods.

  With such a configuration, for example, when the value corresponding to the feature quantity of the radio wave transmitted from the transmitter is set as the predetermined value, the target whose feature quantity is closest to the feature quantity of the radio wave transmitted from the transmitter Since radio waves in the period can be selected, it is possible to correctly monitor human actions in a predetermined area based on the selected radio waves.

  For example, by setting the predetermined value flexibly, it is possible to select the most suitable radio wave from radio waves in each of a plurality of target periods. Specifically, for example, even in a configuration in which the monitoring apparatus processes each radio wave transmitted from a plurality of transmitters, by setting an appropriate predetermined value for each transmitter, a plurality of target periods are set for each transmitter. The radio wave with the least interference wave can be selected from the radio waves in each of the above.

  (11) A monitoring program according to an embodiment of the present invention is a monitoring program used in a monitoring device, the computer receiving a radio wave transmitted from a transmitter arranged in a predetermined area, Calculating a feature quantity of the radio wave in each of a plurality of target periods, and monitoring a human action in the predetermined area based on the radio waves in the target period in which the calculated feature quantity satisfies a predetermined condition Is a program for executing

  With such a configuration, for example, a radio wave that does not include an interference wave can be selected from radio waves in each of a plurality of target periods based on the calculated feature amount. It is possible to suppress false detection caused by processing the above and improve detection accuracy.

  As a result, it is possible to correctly monitor human movements in a predetermined area, so that when the monitoring device is used for intrusion detection, it is possible to avoid the occurrence of false alarms, and when the monitoring device is used for watching and monitoring purposes. In this case, it is possible to avoid the occurrence of misreporting.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. In the drawings, the same or corresponding parts are denoted by the same reference numerals and description thereof will not be repeated. Moreover, you may combine arbitrarily at least one part of embodiment described below.

<First Embodiment>
FIG. 1 is a diagram showing a usage image of the intrusion detection system according to the first embodiment of the present invention.

  Referring to FIG. 1, intrusion detection system 201 according to the first embodiment of the present invention functions as a moving object detection sensor. As an intrusion detection system, at least one set of a transmitter 151 and a receiver 101 is installed in a closed space, for example, a house, which is to be a warning area. The intrusion detection system receives radio waves transmitted from the transmitter 151 within a predetermined interval or continuously by the receiver 101 and performs signal processing based on the received radio waves, so that the person and the door that have entered the room Detects the opening and closing of

  For example, the receiver (monitoring device) 101 in the intrusion detection system 201 is an array-type radio wave sensor, includes a plurality of receiving antenna elements, and realizes a moving object detection function using changes in radio wave propagation in a closed space. In principle, the radio waves used by the intrusion detection system 201 are not limited in frequency, bandwidth, and the like.

[Configuration and basic operation]
FIG. 2 is a diagram showing the configuration of the intrusion detection system according to the first embodiment of the present invention.

  Referring to FIG. 2, in a predetermined area such as a room, receiver 101, transmitter 151, and radios 161A, 161B, and 162 are provided. The receiver 101 and the transmitter 151 constitute an intrusion detection system 201. Hereinafter, each of the wireless devices 161A and 161B is also referred to as a wireless device 161.

  The transmitter 151 transmits radio waves intermittently according to a predetermined wireless communication method, for example. Specifically, the transmitter 151 transmits a radio signal to the receiver 101 in a cycle of 100 milliseconds in accordance with, for example, a wireless communication scheme of IEEE 802.15.4 (Non-Patent Document 2) standard. This radio signal is, for example, a 2.4 GHz band radio signal.

  The wireless device 161 is, for example, a wireless LAN access point, and communicates with other wireless devices 161 using a 2.4 GHz band wireless signal.

  The wireless device 162 is, for example, a ZigBee (registered trademark) device, and transmits and receives wireless signals to and from other wireless devices 162 in accordance with the wireless communication scheme of the IEEE 802.15.4 standard. This radio signal is, for example, a 2.4 GHz band radio signal.

  For example, the receiver 101 detects a human operation in an area where a plurality of devices such as a transmitter 151, a wireless device 161, and a wireless device 162 transmit wireless signals as an intrusion detection device.

  More specifically, the receiver 101 detects a human motion in the area based on the spatial feature amount indicating the state of the predetermined area. That is, the receiver 101 monitors the state in the predetermined area based on the wave propagation properties such as reflection and diffraction. Specifically, the receiver 101 detects a human motion by monitoring the arrival angle distribution calculated based on the reception signals of a plurality of antennas.

  In the intrusion detection system 201, the receiver 101 performs a detection operation using radio waves of a communication signal according to a wireless communication scheme of, for example, the IEEE 802.15.4 standard in the 2.4 GHz band.

  FIG. 3 is a diagram illustrating an example of a frame transmitted by the transmitter according to the first embodiment of the present invention.

  Referring to FIG. 3, frame 401 includes, for example, a synchronization header (SHR), a PHY header (PHR), and a data unit (PSDU: PHY Service Data Unit) in the order of transmission. The area of the data unit includes a MAC header, a MAC payload, and a MAC footer.

  The synchronization header has, for example, an area of 5 octets. In the area from the first octet to the fourth octet of the synchronization header, for example, a value of 0x00000000 is stored as a preamble. Hereinafter, a number starting with “0x” means that the number after “0x” is expressed in hexadecimal. In the fifth octet area of the synchronization header, for example, a value of 0xA7 is stored as an SFD (Start of Frame Delimiter) indicating the start position of the frame 401.

  The PHY header has an area of 1 octet, for example. For example, a value indicating the data unit length is stored in the first octet area of the PHY header.

  In the data unit, the MAC header and the MAC payload have a variable area, for example, and the MAC footer has a 2-octet area, for example. In the MAC header, for example, ID information that is identification information of a device that transmits the frame 401 is stored. For example, data information is stored in the MAC payload. The MAC footer stores, for example, CRC (Cyclic Redundancy Check) information for detecting whether or not there is an error in the values stored in the MAC header and the MAC payload.

[Receiver configuration]
FIG. 4 is a diagram showing a configuration of a receiver in the intrusion detection system according to the first embodiment of the present invention.

  Referring to FIG. 4, receiver 101 includes a received signal monitoring unit 51, an interference removing unit 52, a monitoring unit 53, and an array receiving unit 54. The reception signal monitoring unit 51 includes an antenna 41, a signal analysis unit 42, and a sampling control unit 43. The monitoring unit 53 includes a spatial feature amount calculation unit 11 and an event detection unit 12. The array receiving unit 54 includes an antenna unit 21, a receiving circuit 22, a branch circuit 35, and an oscillator 36. The antenna unit 21 includes, for example, four antennas. The reception circuit 22 includes four sets of a band pass filter 31, a low noise amplifier 32, an orthogonal demodulator 33, and an A / D converter (ADC) 34 corresponding to the antennas in the antenna unit 21. These sets in the receiving circuit 22 are referred to as RX1, RX2, RX3, and RX4, respectively.

  The antenna unit 21 in the array receiving unit 54 receives radio waves, that is, radio signals periodically transmitted from a transmitter 151 arranged in a predetermined area, for example. In the reception circuit 22, the band pass filter 31 attenuates a component outside a predetermined frequency band, for example, a component other than the 2.4 GHz band, among the frequency components of the radio signal received by the corresponding antenna unit 21.

  The low noise amplifier 32 amplifies the radio signal that has passed through the band pass filter 31 and outputs the amplified signal to the quadrature demodulator 33.

  The oscillator 36 generates a local signal and outputs it to the branch circuit 35. The branch circuit 35 outputs a radio signal, which is a local signal received from the oscillator 36, to the quadrature demodulator 33 of the four sets RX 1, RX 2, RX 3, RX 4 of the reception circuit 22.

  The quadrature demodulator 33 multiplies the radio signal received from the low noise amplifier 32 by the local signal received from the oscillator 36 via the branch circuit 35, thereby orthogonally demodulating the signal received from the low noise amplifier 32, for example, as a baseband signal. The signals are converted into band I and Q signals and output to the A / D converter 34.

  For example, the A / D converter 34 samples the I signal and the Q signal received from the quadrature demodulator 33 while receiving the sampling instruction signal from the reception signal monitoring unit 51, and outputs the sampling result to the interference removal unit 52. .

  More specifically, the A / D converter 34 is driven while receiving a sampling instruction signal from the reception signal monitoring unit 51, for example. At this time, the A / D converter 34 converts, for example, the I signal and the Q signal received from the quadrature demodulator 33 into digital signals of n bits (n is a natural number of 2 or more) at a predetermined sampling frequency, respectively. The digital signal is output to the interference removal unit 52. Note that the value of the digital signal generated by the A / D converter 34 indicates, for example, the reception power of the radio signal received by the array reception unit 54.

  The reception signal monitoring unit 51 monitors, for example, a radio signal from the transmitter 151. More specifically, the signal analysis unit 42 in the reception signal monitoring unit 51 receives, for example, radio signals transmitted from the transmitter 151 and the radio unit 162 via the antenna 41.

  For example, the signal analysis unit 42 acquires information included in the frame 401 illustrated in FIG. 3 from the radio signal received via the antenna 41. More specifically, for example, when the SFD in the synchronization header of the frame 401 is detected from a radio signal received via the antenna 41, the signal analysis unit 42 outputs reception start information to the sampling control unit 43 at the detected timing.

  And the signal analysis part 42 acquires the value which shows the data unit length stored in the PHY header of the flame | frame 401, for example, and acquires a data unit based on the acquired value. For example, when the acquisition of the data unit is completed, the signal analysis unit 42 outputs the reception end information to the sampling control unit 43 at the completion timing.

  In addition, the signal analysis unit 42, for example, a CRC verification process for detecting whether there is an error in the values stored in the MAC header and the MAC payload based on the CRC information stored in the MAC footer of the data unit I do. For example, if the signal analysis unit 42 determines that there is no error in the value in the CRC verification process, the signal analysis unit 42 acquires ID information indicating the transmission source of the frame 401 from the MAC header, and outputs the acquired ID information to the sampling control unit 43.

  For example, the signal analysis unit 42 may output the ID information to the sampling control unit 43 at the timing when the ID information stored in the MAC header is acquired without performing the CRC collation process. As a result, a delay due to the CRC collation process can be avoided, so that the ID information can be quickly output to the sampling control unit 43. In addition, the processing load on the received signal monitoring unit 51 can be reduced.

  For example, when the ID information received from the signal analysis unit 42 indicates the transmitter 151, the sampling control unit 43 outputs an ID conformity notification to the interference removal unit 52. For example, when the ID information received from the signal analysis unit 42 indicates the wireless device 162, the sampling control unit 43 outputs an ID nonconformity notification to the interference removal unit 52.

  In addition, the sampling control unit 43 sets, for example, a sampling period that is a period during which the array receiving unit 54 should perform sampling of a radio signal.

  FIG. 5 is a diagram illustrating an example of a sampling period set by the receiver according to the first embodiment of the present invention.

  Referring to FIG. 5, specifically, sampling control unit 43 sets, for example, a sampling period from a timing at which reception start information is received from signal analysis unit 42 to a timing at which reception end information is received. For example, the sampling control unit 43 outputs a sampling instruction signal to the array reception unit 54 and the interference removal unit 52 in the set sampling period. The length of the sampling period is 3 to 5 milliseconds.

  More specifically, the transmitter 151 transmits the frame 402 from time t11 to time t12, for example. For example, the sampling control unit 43 outputs a sampling instruction signal to the array reception unit 54 and the interference removal unit 52 in the sampling period Ts11 from time t31 to time t12 when the SFD of the frame 402 is detected.

  For example, the wireless device 161 transmits a wireless signal that is included in the sampling period Ts11 and is an interference wave for the receiver 101 during the period Ti1 from time t21 to time t22. For this reason, the array receiving unit 54 samples a radio signal including an interference wave in a part of the sampling period Ts11 and outputs the sample group SG11 as a sampling result to the interference removing unit 52.

  Further, the transmitter 151 transmits the frame 403 from time t13 to time t14, for example. For example, the sampling control unit 43 outputs a sampling instruction signal to the array receiving unit 54 and the interference removing unit 52 in the sampling period Ts12 from time t32 to time t14 when the SFD of the frame 403 is detected. The array receiver 54 samples a radio signal that does not include an interference wave in the sampling period Ts12, and outputs a sample group SG12 that is a sampling result to the interference remover 52.

  Further, the transmitter 151 transmits the frame 404 from time t15 to time t16, for example. For example, the sampling control unit 43 outputs a sampling instruction signal to the array receiving unit 54 and the interference removing unit 52 in the sampling period Ts13 from time t33 to time t16 when the SFD of the frame 404 is detected.

  For example, the radio 161 transmits a radio signal that is included in the sampling period Ts13 and is an interference wave for the receiver 101 during the period Ti2 from time t23 to time t24. For this reason, the array receiving unit 54 samples a radio signal including an interference wave in a part of the sampling period Ts13, and outputs the sample group SG13 as a sampling result to the interference removing unit 52.

  FIG. 6 is a diagram illustrating another example of the sampling period set by the receiver according to the first embodiment of the present invention.

  In FIG. 6, the periods in which the transmitter 151 transmits frames 402, 403, and 404, and the periods Ti1 and Ti2 in which the radio 161 transmits radio signals are the same as the timing shown in FIG.

  For example, the sampling control unit 43 can set the sampling period flexibly when the signal analysis unit 42 does not perform CRC collation processing. Specifically, the sampling control unit 43 sets a short sampling period, for example. More specifically, for example, the sampling control unit 43 sets the sampling period Ts21 from time t31 when the SFD of the frame 402 is detected to time t42, which is a timing before time t21.

  In addition, the sampling control unit 43 sets a sampling period that spans transmission periods of a plurality of frames, for example. More specifically, the sampling control unit 43, for example, from the time t43 to the time t14 in the transmission period of the frame 403, and from the time t33 when the SFD of the frame 404 is detected to the time t44 that is a timing before the time t23. Is set to the sampling period Ts22.

  Also, the sampling control unit 43 sets, for example, the sampling period Ts23 from time t45 to time t16, which is a timing after the time t44 and before the time t23.

  For example, the sampling control unit 43 outputs a sampling instruction signal to the array receiving unit 54 and the interference removing unit 52 in the sampling periods Ts21, Ts22, and Ts23. The array receiving unit 54 samples a radio signal that does not include an interference wave in the sampling period Ts21, and outputs a sample group SG21 that is a sampling result to the interference removing unit 52.

  The array reception unit 54 samples a radio signal that does not include an interference wave in a sampling period Ts22 that extends over the transmission periods of the frames 403 and 404, and outputs a sample group SG22 that is a sampling result to the interference removal unit 52. The array receiver 54 samples a radio signal including an interference wave in a part of the sampling period Ts23 and outputs a sample group SG23 as a sampling result to the interference remover 52.

  For example, as shown in FIG. 6, the configuration in which the sampling control unit 43 sets a short sampling period may reduce the probability that the periods Ti1 and Ti2 during which interference waves are transmitted overlap the sampling periods Ts21, Ts22, and Ts23. it can. Thereby, in the array receiving part 54, possibility that the radio signal in which an interference wave is contained in a part of sampling period will be sampled can be lowered | hung.

  On the other hand, for example, as shown in FIG. 5, the length of the sampling periods Ts11, Ts12, and Ts13 can be increased by the configuration in which the sampling control unit 43 sets the sampling period based on the reception start information and the reception end information. The amount of data included in the sample groups SG11, SG12, and SG13 can be increased. Thereby, in the subsequent calculation using the sample groups SG11, SG12, and SG13, the influence of various noises such as Gaussian noise can be reduced, so that the calculation accuracy in the subsequent stage can be improved.

  Therefore, it is preferable to set the sampling period while comparing the period during which the interference wave is transmitted and the probability that the sampling period overlaps with the calculation accuracy of the subsequent stage.

  The reception signal monitoring unit 51 is configured to recognize the reception timing of the frame 401 from the transmitter 151 based on the radio signal received via the antenna 41, but is not limited thereto. The reception signal monitoring unit 51 may recognize the reception timing of the frame 401 from the transmitter 151 based on, for example, a radio signal received via the antenna in the antenna unit 21 in the array reception unit 54.

  For example, when the transmitter 151 and the receiver 101 are connected by a cable, the reception signal monitoring unit 51 receives a signal indicating the timing at which the frame 401 is transmitted from the transmitter 151 via the cable. The configuration may be such that the reception timing of the frame 401 is recognized based on the received signal.

[Task]
As described above, when there is a wireless device 161 that uses radio waves in the same frequency band as the radio waves transmitted by the transmitter 151, the array receiving unit 54 transmits the transmitter 151 during a part or all of the sampling period. The radio signal transmitted from the radio and the radio signal transmitted from the radio 161, that is, an interference wave may be sampled.

  For this reason, the receiver 101 calculates an incorrect spatial feature amount based on a sample group generated from a radio signal including an interference wave, and cannot correctly detect a human action.

  Specifically, for example, when the receiver 101 is used for intrusion detection, the receiver 101 determines that an intruder has been detected by an interference wave even when there is no actual intruder in a predetermined area. False alarms may occur. Further, when the receiver 101 is used for watching and watching, the receiver 101 determines that there is human activity due to interference waves even when there is no human activity in a predetermined area, and overlooks the abnormality of the target person. May cause misreporting.

  Therefore, the monitoring device according to the first embodiment of the present invention solves such a problem by the following configuration and operation.

[Configuration of interference canceller]
FIG. 7 is a diagram illustrating a configuration of an interference canceller in the receiver according to the first embodiment of the present invention.

  Referring to FIG. 7, interference removal unit 52 includes a feature amount calculation unit 63, a memory control unit 71, and a memory 72. The feature quantity calculation unit 63 includes a signal extraction unit 64, a received signal analysis unit 65, and a sample group selection unit 66.

  The memory 72 stores, for example, digital signals received from the A / D converters 34 of RX1 to RX4 in the array receiving unit 54 under the control of the memory control unit 71.

  The memory control unit 71 processes a digital signal stored in the memory 72, for example. More specifically, the memory control unit 71 performs an update process on the memory 72 based on, for example, an ID conformity notification, an ID nonconformity notification, and a sampling instruction signal received from the received signal monitoring unit 51.

  Specifically, the memory control unit 71 causes the memory 72 to function as a ring buffer, for example. That is, the memory control unit 71 accumulates, for example, sample groups in the memory 72 up to the upper limit number. For example, when accumulating a new sample group in the memory 72 in a state where five sample groups are accumulated as the upper limit number, the memory control unit 71 discards the oldest sample group out of the five sample groups, and then creates a new sample group. The sample group is stored in the memory 72.

  More specifically, the memory control unit 71 receives, for example, a digital signal received from the A / D converter 34 as a new sample group SGnew in the memory 72 while receiving the sampling instruction signal from the reception signal monitoring unit 51 during the sampling period. accumulate.

  And the memory control part 71 will cancel | discard sample group SGnew, if ID notification of nonconformity is received from the received signal monitoring part 51, for example. Further, for example, when receiving the ID conformity notification from the reception signal monitoring unit 51, the memory control unit 71 updates the five sample groups SG [1] to SG [5] and indicates that the update processing of the memory 72 is completed. An update completion notification is output to the feature amount calculation unit 63.

  At the time of updating the sample group, for example, the memory control unit 71 discards the oldest sample group among the sample groups SG [1] to SG [5] before the update, and then updates the remaining sample group and sample group SGnew. The samples are sequentially stored in the sample groups SG [1] to SG [5]. That is, the memory 72 stores a sample group generated based on the radio signal transmitted from the transmitter 151.

  Here, a period in which the array receiver 54 samples a radio signal transmitted from the transmitter 151 that is a measurement target of the array receiver 54 is defined as a target period. In other words, the sampling period corresponding to the sample group generated based on the radio signal from the transmitter 151 corresponds to the target period.

  Specifically, five sampling periods corresponding to each of the sample groups SG [1] to SG [5] stored in the memory 72 correspond to the target periods T [N1] to T [N5], respectively. Hereinafter, “N *” in the “target period T [N *]” represents a numerical value. Further, it is assumed that the larger the number corresponding to “*” is, the longer the corresponding target period is. Specifically, N1 to N5 are continuous numerical values such as 51 to 55, for example. The target periods T [N1] to T [N5] are the past target periods in the order of T [N1] to T [N5].

  Although the memory control unit 71 is configured to store the sample groups corresponding to the five target periods in the memory 72, the present invention is not limited to this. For example, the memory control unit 71 may be configured to store a sample group corresponding to two or more target periods in the memory 72.

  The feature amount calculation unit 63 calculates the feature amount of the radio signal received by the array reception unit 54 based on, for example, a plurality of sample groups stored in the memory 72.

  More specifically, for example, when the signal extraction unit 64 in the feature amount calculation unit 63 receives an update completion notification from the memory control unit 71, the signal extraction unit 64 performs processing from the sample groups SG [1] to SG [5] accumulated in the memory 72. The signals S [1] to S [5] are extracted. The signal extraction unit 64 outputs the extracted target signals S [1] to S [5] to the reception signal analysis unit 65, for example.

  Here, the target signal is generated from a digital signal sampled by the A / D converter 34 corresponding to one or a plurality of sets among the sets RX1 to RX4 in the reception circuit 22, for example. Since the influence of the interference wave is considered to be equal for each set, a digital signal sampled for a target period by the A / D converter 34 corresponding to one set may be used as a representative of the target signal.

[When the statistical value during the target period of received power is used as a feature value]
FIG. 8 is a diagram illustrating an example of a time change of the target signal processed by the feature amount calculation unit according to the first embodiment of the present invention.

  Referring to FIG. 8, reception signal analysis unit 65 calculates, for example, a statistical value in each target period of reception power of radio signal transmitted from transmitter 151 in receiver 101 as a feature amount.

  Specifically, the received signal analyzer 65, for example, as shown in FIG. 8, each value of the target signals S [1] to S [5] received from the signal extractor 64, that is, the average value M [1] of the levels. To M [5] are calculated as feature quantities A [1] to A [5] in the target periods T [N1] to T [N5] of the received power at the receiver 101 of the wireless signals transmitted from the transmitter 151, respectively. To do.

  The received signal analysis unit 65 is configured to calculate the average value of the level of the target signal as the feature amount in the target period of the received power. However, the present invention is not limited to this. The reception signal analysis unit 65 may be configured to calculate, for example, a statistical value such as a median value and a mode value of the level of the target signal as a feature amount in the target period of the reception power.

  Alternatively, the reception signal analysis unit 65 transmits, for example, higher-order statistics D [1] to D [5] of each level of the target signals S [1] to S [5] received from the signal extraction unit 64, respectively. The wireless signal transmitted from the device 151 is calculated as the feature amounts A [1] to A [5] in the target period T [N1] to T [N5] of the received power in the receiver 101.

  The higher-order statistic is, for example, the variance of the level of the target signal in one target period. Further, the higher-order statistic may be a sum of squares, sums of squares, sums of fourths, and the like of the level of the target signal in one target period.

  The received signal analysis unit 65 outputs the calculated feature amounts A [1] to A [5] to the sample group selection unit 66, for example.

  For example, when the sample group selection unit 66 receives the feature amounts A [1] to A [5] from the reception signal analysis unit 65, the sample group selection unit 66, based on the received feature amounts A [1] to A [5], Select the sample group to be used in the calculation of.

  Specifically, the sample group selection unit 66 selects, for example, a feature quantity closest to a predetermined value from the feature quantities A [1] to A [5], and uses an index corresponding to the selected feature quantity. Thus, the sample group to be used for the calculation of the spatial feature is selected from the sample groups SG [1] to SG [5] stored in the memory 72.

  More specifically, the sample group selection unit 66 holds, for example, a standard feature value Astd learned in advance as a predetermined value. The standard feature amount Astd is, for example, a standard average value Mstd that is an average value of sampling results by the array receiving unit 54 of wireless signals that are normally transmitted from the transmitter 151, and higher-order statistics of the sampling results. It is at least one of the standard higher-order statistics Dstd.

  For example, the sample group selection unit 66 selects a feature amount having a value closest to the standard feature amount Astd from the feature amounts A [1] to A [5], and displays index information indicating an index corresponding to the selection result. Notify the memory control unit 71.

  Specifically, for example, when the average value M [2] of the average values M [1] to M [5] has a value closest to the standard average value Mstd, the sample group selection unit 66 sets the average value M [ 2] and notifies the memory control unit 71 of “2” that is an index corresponding to the selection result as index information.

  In addition, for example, when the high-order statistic D [2] has a value closest to the standard high-order statistic Dstd among the high-order statistic D [1] to D [5], The higher-order statistic D [2] is selected, and “2” corresponding to the selection result is notified to the memory control unit 71 as index information.

  Note that the sample group selection unit 66 should be used for the calculation of the spatial feature amount based on both the average values M [1] to M [5] and the higher-order statistics D [1] to D [5], for example. A sample group may be selected.

  Specifically, the sample group selection unit 66, for example, the difference between the average values M [1] to M [5] and the standard average value Mstd, and the higher-order statistics D [1] to D [5] and the standard A difference from the higher-order statistic Dstd may be represented by a point, and a sample group to be used for calculating the spatial feature may be selected based on a value obtained by adding the points for each index.

  Further, for example, when the target signal S [3] is extracted from the sample group SG [3] generated from the radio signal including the interference wave, as in the target signal S [3] illustrated in FIG. The level of the target signal S [3] increases during the period Ti in which the interference wave 54 is sampled.

  In this case, the average value M [3] and the higher-order statistic D [3] calculated from the target signal S [3] are larger than the standard average value Mstd and the standard higher-order statistic Dstd, respectively.

  Therefore, the sample group selection unit 66 can exclude, for example, the feature amount A [3] based on the radio signal including the interference wave from the feature amounts A [1] to A [5]. Thereby, it is possible to avoid the sample group corresponding to the feature amount A [3], that is, the sample group generated from the radio signal including the interference wave, being used in the calculation in the monitoring unit 53 at the subsequent stage.

  The memory control unit 71 processes the digital signal stored in the memory 72 based on the index information received from the sample group selection unit 66, for example. Specifically, for example, when the index information received from the sample group selection unit 66 indicates “2”, the memory control unit 71 acquires a copy of the sample group SG [2], and acquires the acquired sample group SG [2]. Is output to the monitoring unit 53 as a current sample group.

[When calculating feature values from the frequency distribution of radio waves in the target period]
FIG. 9 is a diagram illustrating an example of a frequency spectrum of a target signal processed by the feature amount calculation unit according to the first embodiment of the present invention.

  Referring to FIG. 9, reception signal analysis unit 65 calculates a feature quantity from the frequency distribution of radio waves in the target period received by array reception unit 54, for example. The frequency distribution of the radio wave in the target period received by the array receiver 54 is, for example, the frequency spectrum of the received power of the radio wave in the target period.

  Specifically, for example, as shown in FIG. 9, the received signal analysis unit 65 performs Fourier transform on each of the target signals S [1] to S [5] received from the signal extraction unit 64, thereby performing a target period T [ Each frequency spectrum FS [1] to FS [5] in N1] to T [N5] is generated.

  For example, the reception signal analysis unit 65 calculates the feature amount A from the generated frequency spectrum. Specifically, the received signal analysis unit 65 holds, for example, the standard frequency spectrum FSstd shown in FIG. 9 learned in advance. The standard frequency spectrum FSstd is, for example, a spectrum obtained by performing Fourier transform on a target signal based on a radio signal that is normally transmitted from the transmitter 151.

  For example, the reception signal analysis unit 65 performs pattern matching between each of the frequency spectrums FS [1] to FS [5] and the standard frequency spectrum FSstd, and obtains a pattern matching result indicating the similarity between them as a feature amount A [ 1] to A [5], respectively.

  Note that the received signal analysis unit 65 uses, for example, the bandwidths, peak shapes, or amplitude magnitudes in a predetermined frequency band of the frequency spectra FS [1] to FS [5] as feature amounts A [1] to A [5. ] May be calculated respectively.

  The received signal analysis unit 65 outputs the calculated feature amounts A [1] to A [5] to the sample group selection unit 66, for example.

  For example, when the sample group selection unit 66 receives the feature amounts A [1] to A [5] from the received signal analysis unit 65, the sample group selection unit 66 selects a predetermined value, that is, a standard value among the received feature amounts A [1] to A [5]. A feature amount closest to the feature amount Astd is selected, and index information indicating an index corresponding to the selection result is notified to the memory control unit 71.

  Here, the standard feature amount Astd is, for example, a pattern matching result when the target frequency spectrum completely matches the standard frequency spectrum FSstd. The standard feature amount Astd may be, for example, the bandwidth of the standard frequency spectrum FSstd, the peak shape, or the magnitude of the amplitude in a predetermined frequency band.

  Specifically, for example, when the feature quantity A [1] has a value closest to the standard feature quantity Asstd, the sample group selection unit 66 selects the feature quantity A [1] and uses an index corresponding to the selection result. A certain “1” is notified to the memory control unit 71 as index information.

  For example, when the target signal S [3] is extracted from the sample group SG [3] generated from the radio signal including the interference wave, a standard frequency is obtained as in the frequency spectrum FS [3] illustrated in FIG. A peak may appear between the frequencies f1 and f2 at which no peak appears in the spectrum FSstd.

  In this case, the pattern matching result between the frequency spectrum FS [3] and the standard frequency spectrum FSstd becomes worse.

  Therefore, the sample group selection unit 66 can exclude, for example, the feature amount A [3] based on the radio signal including the interference wave from the feature amounts A [1] to A [5]. Thereby, it is possible to avoid the sample group corresponding to the feature amount A [3], that is, the sample group generated from the radio signal including the interference wave, being used in the calculation in the monitoring unit 53 at the subsequent stage.

  For example, when the index information received from the sample group selection unit 66 indicates “1”, the memory control unit 71 acquires a copy of the sample group SG [1] from the memory 72 and copies the acquired sample group SG [1]. To the monitoring unit 53 as a current sample group.

  Referring to FIG. 4 again, the spatial feature quantity calculation unit 11 in the monitoring unit 53 calculates a spatial feature quantity in a predetermined area based on the radio signal received by the array reception unit 54, and based on the calculated spatial feature quantity. The human motion in the predetermined area is detected.

  Specifically, for example, when receiving the current sample group from the interference removing unit 52, the spatial feature quantity calculating unit 11 receives the level and arrival timing of the radio signal received by each antenna unit 21 based on the received current sample group. Is calculated. And the spatial feature-value calculation part 11 calculates the spatial feature-value in a predetermined area based on a calculation result. That is, the spatial feature amount calculation unit 11 calculates a spatial feature amount indicating a state of the predetermined area for a predetermined area where a human motion is to be detected.

More specifically, the spatial feature quantity calculation unit 11 extracts a spatial feature quantity using the arrival angle distribution, for example, similarly to the configuration described in Patent Document 1. That is, the spatial feature quantity calculation unit 11 extracts the spatial feature quantity P (t) by calculating the inner product of the eigenvectors. This inner product indicates the amount of change from the initial vector serving as a comparison reference. Assuming that the initial vector, that is, the eigenvector when there is no intruder is vno and the eigenvector at the observation time t is vob (t), the spatial feature P (t) is expressed by the following equation.
P (t) = (vno, vob (t))

  Here, (A, B) represents the inner product value of the vector A and the vector B. The event detection unit 12 monitors human movements in a predetermined area based on the spatial feature amount P (t) calculated by the spatial feature amount calculation unit 11. Specifically, the closer the spatial feature P (t) is to “1”, the closer the state of the predetermined area at the observation time t is to the normal state where no intruder exists in the predetermined area. Therefore, the event detection unit 12 determines that a human has entered a predetermined area when the spatial feature P (t) is less than a predetermined threshold, for example, “0.999”. The event detection unit 12 transmits an alarm signal to a security company, for example, in order to notify that a human has entered a predetermined area.

[Modification]
In the interference removal unit 52 according to the first embodiment of the present invention, the memory control unit 71 discards a new sample group accumulated in the memory 72 when receiving an ID nonconformity notification from the reception signal monitoring unit 51. However, the present invention is not limited to this.

  For example, the memory control unit 71 may not discard the new sample group accumulated in the memory 72 even when receiving the ID nonconformity notification from the reception signal monitoring unit 51. That is, the memory 72 stores a sample group SGu generated based on the radio signal transmitted from the radio 162 in addition to the sample group SGn generated based on the radio signal transmitted from the transmitter 151. There is a case.

  The feature amount calculation unit 63 calculates the feature amount based on each sample group SGn, SGu stored in the memory 72, for example. For example, since the propagation path of the wireless signal transmitted from the transmitter 151 is different from the propagation path of the wireless signal transmitted from the wireless apparatus 162, the feature amount Au calculated from the sample group SGn is calculated from the sample group SGu. Different from the feature quantity An.

  That is, for example, the feature amount An calculated from the sample group SGn has a value closer to the standard feature amount Astd than the feature amount Au calculated from the sample group SGu.

  Therefore, even when the sample group SGu is accumulated in addition to the sample group SGn in the memory 72, the feature quantity calculating unit 63 can correctly select the sample group SGn to be used for calculating the spatial feature quantity from the memory 72. it can.

[Operation]
FIG. 10 is a flowchart defining an operation procedure when the receiver according to the first embodiment of the present invention monitors human motion. The transmitter 151 and the receiver 101 in the intrusion detection system 201 include a computer, and an arithmetic processing unit such as a CPU in the computer reads a program including a part or all of each step of the following flowchart from a memory (not shown). Run. Each of the programs of the plurality of apparatuses can be installed from the outside. The programs of the plurality of apparatuses are distributed while being stored in a recording medium.

  Referring to FIG. 10, in the intrusion detection mode, first, transmitter 151 periodically transmits radio waves to receiver 101. The receiver 101 receives a radio wave transmitted from the transmitter 151, that is, a radio signal (step S102).

  Next, the reception signal monitoring unit 51 in the receiver 101 continues to receive the radio signal until, for example, SFD included in the radio signal transmitted from the transmitter 151 is detected (NO in step S104 and step S102).

  Next, when the received signal monitoring unit 51 detects, for example, an SFD, it sets a sampling period based on the timing at which the SFD is detected (YES in step S104).

  Next, for example, the array receiver 54 samples the radio signal for the sampling period set by the received signal monitor 51, and outputs the sampling result to the interference remover 52 (step S106).

  Next, the interference removal unit 52 starts the update process of the memory 72 using, for example, the digital signal sampled by the array reception unit 54 as a new sample group (step S108).

  Next, for example, when the interference cancellation unit 52 receives an ID nonconformity notification indicating that the wireless signal is transmitted from the wireless device 162 from the received signal monitoring unit 51 (NO in step S110), the new sample group Is discarded (step S112).

  Next, the receiver 101 receives a radio signal newly transmitted from the transmitter 151 (step S102).

  On the other hand, for example, when the ID cancellation notification indicating that the radio signal is transmitted from the transmitter 151 is received from the received signal monitoring unit 51 (YES in step S110), the interference removing unit 52 selects the new sample group. The update process of the memory 72 is completed (step S114).

  Next, the interference removal unit 52 calculates a feature amount for each sample group stored in the memory 72, for example (step S116).

  Next, the interference removal unit 52 selects, for example, a feature amount that satisfies a predetermined condition from the calculated feature amounts, and outputs a sample group corresponding to the selected feature amount to the monitoring unit 53 (step S118).

  Next, the monitoring unit 53 calculates, for example, a spatial feature amount based on the sample group received from the interference removal unit 52 (step S120).

  Next, the monitoring unit 53 monitors a human action in a predetermined area based on, for example, the magnitude relationship between the calculated spatial feature amount and a predetermined threshold (step S122).

  Next, the receiver 101 receives a radio signal newly transmitted from the transmitter 151 (step S102).

  In the intrusion detection system according to the first embodiment of the present invention, the receiver 101 is configured to process radio waves transmitted from one transmitter 151. However, the present invention is not limited to this. Absent. For example, the receiver 101 may be configured to process each radio wave transmitted from the plurality of transmitters 151.

  By the way, the event detection apparatus described in Patent Document 1 is configured to detect an intrusion detection radio wave transmitted from a predetermined transmitter and a radio wave transmitted from another apparatus in a state where the event detection apparatus is installed in a predetermined area such as a room. There is a case of receiving simultaneously.

  In this case, since the event detection apparatus processes radio waves including interference waves, the following problems may occur due to erroneous detection that erroneously detects the occurrence of an event. That is, for example, in an event detection device used for intrusion detection, even though no abnormality has actually occurred in a predetermined area, it may be determined to be abnormal due to erroneous detection, and erroneous reporting may occur.

  On the other hand, in the monitoring apparatus according to the first embodiment of the present invention, the array receiving unit 54 receives radio waves transmitted from the transmitter 151 arranged in a predetermined area. The feature amount calculation unit 63 calculates the feature amount of the radio wave in each of a plurality of target periods received by the array reception unit 54. The monitoring unit 53 monitors human actions in a predetermined area based on radio waves in a target period in which the feature amount satisfies a predetermined condition.

  With such a configuration, for example, a radio wave that does not include an interference wave can be selected from radio waves in each of a plurality of target periods based on the calculated feature amount. Therefore, the receiver 101 includes an interference wave. False detection due to processing of radio waves can be suppressed, and detection accuracy can be improved.

  As a result, it is possible to correctly monitor human movements in a predetermined area, so that it is possible to avoid the occurrence of false alarms when the receiver 101 is used for intrusion detection.

  In the monitoring apparatus according to the first embodiment of the present invention, the feature amount calculation unit 63 calculates a statistical value in the target period of radio wave reception power in the receiver 101 as a feature amount.

  As described above, the above-described statistical value that changes according to the degree to which the interference wave is included in the radio wave in the receiver 101 is calculated as the feature amount, so that the interference wave is included in the radio waves in each of the plurality of target periods. A low radio wave can be selected.

  In the monitoring apparatus according to the first embodiment of the present invention, the feature amount calculation unit 63 calculates the feature amount from the frequency spectrum of the radio wave in the target period received by the array reception unit 54.

  As described above, the feature amount is calculated from the frequency spectrum that changes according to the degree to which the interference wave is included in the radio wave in the target period received by the array receiver 54. Therefore, it is possible to select a radio wave that is low enough to contain interference waves.

  Further, in the monitoring apparatus according to the first embodiment of the present invention, the monitoring unit 53 is a person in a predetermined area based on radio waves in a target period whose feature amount is closest to a predetermined value among a plurality of target periods. Monitor the operation of

  With such a configuration, for example, when a value corresponding to the feature quantity of the radio wave transmitted from the transmitter 151 is set as the predetermined value, the feature quantity is closest to the feature quantity of the radio wave transmitted from the transmitter 151. Since a radio wave in a certain target period can be selected, it is possible to correctly monitor a human operation in a predetermined area based on the selected radio wave.

  For example, by setting the predetermined value flexibly, it is possible to select the most suitable radio wave from radio waves in each of a plurality of target periods. Specifically, for example, even in a configuration in which the receiver 101 processes each radio wave transmitted from the plurality of transmitters 151, by setting an appropriate predetermined value for each transmitter 151, for each transmitter 151, The radio wave with the least interference wave can be selected from the radio waves in each of the plurality of target periods.

  In the monitoring apparatus according to the first embodiment of the present invention, the monitoring unit 53 is configured to detect a human intrusion into a predetermined area as a human operation in the predetermined area. Not what you want. The monitoring unit 53 may be configured to detect the start of suspicious behavior of a human being present in a predetermined area as a human action in the predetermined area. Also in this case, the monitoring unit 53 can detect the start of suspicious behavior of a human being existing in the predetermined area based on the variation of the spatial feature P (t).

  Moreover, although the monitoring apparatus according to the first embodiment of the present invention is configured to perform a binary determination of the presence or absence of an intruder, the present invention is not limited to this. The structure which outputs the parameter | index which shows the level of intrusion possibility may be sufficient.

  In the monitoring apparatus according to the first embodiment of the present invention, the antenna unit 21 in the array receiving unit 54 is configured to include four antennas. However, the present invention is not limited to this. For example, the antenna unit 21 may be configured to include one or a plurality of antennas.

  For example, when the antenna unit 21 is configured to include one antenna instead of the array antenna, the receiver 101 may detect human actions by the following operations. That is, for example, the receiver 101 monitors the reception power of a radio signal, and detects human action when the monitored reception power is disturbed. In addition, the receiver 101 calculates a spatial feature amount by a method using a power delay profile by UWB-IR described in Non-Patent Document 1, for example, and detects a human motion based on the calculated spatial feature amount.

  Next, another embodiment of the present invention will be described with reference to the drawings. In the drawings, the same or corresponding parts are denoted by the same reference numerals and description thereof will not be repeated.

<Second Embodiment>
The present embodiment relates to a monitoring apparatus that calculates a feature amount based on the relationship between eigenvectors generated from each accumulated sample group, as compared with the monitoring apparatus according to the first embodiment. The contents other than those described below are the same as those of the intrusion detection system according to the first embodiment.

[Overview]
The receiver detects the “total variation” from the received radio signal. The “total variation amount” is represented, for example, as the sum of “the variation amount due to human movement” and “the variation amount due to the interference wave”.

  For example, when the radio wave law or the like sets an upper limit of radio wave transmission time per fixed time, the radio device 161 serving as the interference wave transmission source for the monitoring device transmits a burst signal. For this reason, the period in which the interference wave is transmitted is often instantaneous with respect to the scale of the sampling period. On the other hand, since the human motion continues for 2 to 3 seconds, for example, it can be regarded as continuous with respect to the scale of the sampling period.

  Therefore, as shown in FIGS. 5 and 6, the period in which the interference wave is transmitted is often shorter than the period in which the frame is transmitted. For this reason, interference waves are included in radio waves in a small number of target periods among the plurality of target periods.

  When the interference wave is included in the radio wave, the “variation amount due to the interference wave” increases, so that, for example, the “total variation amount” increases. Therefore, the “total variation amount” having a large value among the “total variation amounts” calculated from the radio signals in a plurality of target periods is highly likely to be calculated from the radio signal including the interference wave.

  That is, among the “total fluctuation amounts” calculated from radio signals in a plurality of target periods, the “total fluctuation amount” having the smallest value is the “total fluctuation amount” calculated from a radio signal that does not include an interference wave. Can be expected.

  Therefore, the monitoring apparatus according to the present embodiment calculates the “total variation amount”, that is, the feature amount based on the relationship between the eigenvectors generated from each sample group with the following configuration.

  The “total variation amount” is specifically a spatial feature amount P (t) indicating an inner product value of the eigenvector vno when there is no intruder and the eigenvector vob (t) at the observation time t. Therefore, the “total variation amount” having the smallest value corresponds to the spatial feature amount P (t) having a value closest to 1.

  For this reason, among the spatial feature amounts P (t) calculated from the radio signals in a plurality of target periods, the spatial feature amount P (t) having a value closest to 1 is calculated from a radio signal that does not include an interference wave. It can be expected that the spatial feature amount P (t).

[Receiver configuration]
FIG. 11 is a diagram showing a configuration of a receiver in the intrusion detection system according to the second embodiment of the present invention.

  Referring to FIG. 11, the receiver 102 is different from the receiver 101 according to the first embodiment in that an interference removing unit 252, 253, or 254 is used instead of the interference removing unit 52 and the monitoring unit 53. And a monitoring unit 55. The monitoring unit 55 includes a spatial feature quantity calculation unit 211 instead of the spatial feature quantity calculation unit 11 as compared with the monitoring unit 53 according to the first embodiment.

[When the radio wave is older than multiple target periods]
The interference removal unit 252 may, for example, based on the reference radio waves received by the array reception unit 54 during a period in which the predetermined area is in a predetermined state before the plurality of target periods. Select a radio wave from each radio wave corresponding to the target period.

[Configuration of interference removing unit 252]
FIG. 12 is a diagram illustrating a configuration of an interference canceller in the receiver according to the second embodiment of the present invention.

  Referring to FIG. 12, the interference removal unit 252 includes a feature amount calculation unit 73 instead of the feature amount calculation unit 63 as compared to the interference removal unit 52 according to the first embodiment. The feature amount calculation unit 73 includes an eigenvector generation unit 75, an inner product calculation unit 76, and a sample group selection unit 77.

  Similarly to the memory control unit 71 shown in FIG. 7, the memory control unit 71 causes the memory 72 to function as a ring buffer having an upper limit of 5, for example. In the memory 72, for example, sample groups SG [1] to SG [5] are stored in order from the newest. Five sampling periods corresponding to each of the sample groups SG [1] to SG [5] correspond to the target periods T [N1] to T [N5], respectively.

  For example, in the memory 72, reference radio waves that are radio waves received by the array receiving unit 54 in a period in which a predetermined area is in a predetermined state before the target periods T [N1] to T [N5]. The generated sample group RSG is accumulated.

  Specifically, the sample group RSG is a digital signal obtained by sampling the reference radio wave from the transmitter 151 in the sampling period in which the predetermined area is unmanned.

  For example, the feature amount calculation unit 73 calculates a value representing a relationship between each radio wave and the reference radio wave in the target period T [N1] to T [N5] as a feature amount.

  Specifically, the eigenvector generation unit 75 in the feature amount calculation unit 73 generates a normalized eigenvector from the sample group stored in the memory 72 when receiving an update completion notification from the memory control unit 71, for example.

  More specifically, the eigenvector generation unit 75, for example, according to the calculation procedure shown in Patent Document 1, from each of the sample groups SG [1] to SG [5], the radio waves in the target periods T [N1] to T [N5] Eigenvectors vsg [1] to vsg [5] are generated.

  The eigenvector generation unit 75 generates the eigenvector vsg [0] of the reference radio wave from the sample group RSG stored in the memory 72, for example. The sign of the eigenvector is arbitrary. For this reason, the eigenvector may have either a positive sign or a negative sign.

  For example, the eigenvector generation unit 75 outputs the generated eigenvectors vsg [1] to vsg [5] and the eigenvector vsg [0] to the inner product calculation unit 76.

  For example, when the eigenvector generation unit 75 specifies how to determine the sign of the eigenvector, the inner product calculation unit 76 specifies the eigenvector vsg [1] to the radio waves in the target periods T [N1] to T [N5], respectively. Each inner product value of vsg [5] and eigenvector vsg [0] of the reference radio wave is calculated as a feature amount.

  Specifically, the inner product calculation unit 76, for example, (vsg [0], vsg [1]), (vsg [0], vsg [2]), (vsg [0], vsg [3]), (Vsg [0], vsg [4]) and (vsg [0], vsg [5]) are calculated as feature values, and the calculated feature values are output to the sample group selection unit 77.

  Further, for example, when the method of determining the sign of the eigenvector in the eigenvector generation unit 75 is arbitrary, the inner product calculation unit 76 performs eigenvector vsg [1] respectively corresponding to the radio waves in the target periods T [N1] to T [N5]. A value based on each inner product value of ~ vsg [5] and the eigenvector vsg [0] of the reference radio wave, that is, an absolute value of each inner product value is calculated as a feature amount.

  Specifically, the inner product calculation unit 76, for example, | (vsg [0], vsg [1]) |, | (vsg [0], vsg [2]) |, | (vsg [0], vsg [3]) |, | (vsg [0], vsg [4]) | and | (vsg [0], vsg [5]) | are calculated as feature quantities, and the calculated feature quantities are sample group selection unit 77. Output to. Here, | (A, B) | represents the absolute value of the inner product value of the vector A and the vector B.

  Each feature amount corresponds to, for example, a spatial feature amount P (t) in each target period corresponding to the eigenvectors vsg [1] to vsg [5] used for the calculation.

  For example, the sample group selection unit 77 selects a feature quantity corresponding to a radio wave in the target period that has the highest degree of coincidence with the reference radio wave among a plurality of target periods, and uses an index corresponding to the selected feature quantity. Thus, the sample group to be used for the calculation of the spatial feature is selected from the sample groups SG [1] to SG [5] stored in the memory 72.

  Specifically, for example, when the sample group selection unit 77 receives an inner product value or an absolute value of the inner product value as a feature amount from the inner product calculation unit 76, the feature amount closest to a predetermined value among the received feature amounts. Select. More specifically, since the eigenvector is standardized, the sample group selection unit 77 selects, for example, a feature quantity having a value closest to 1 from the received feature quantities.

  Here, the feature amount selected by the sample group selection unit 77 is the same as the spatial feature amount P (t) calculated by the monitoring unit 55 in the subsequent stage, for example.

  For example, the sample group selection unit 77 notifies the memory control unit 71 of index information indicating an index corresponding to the eigenvector used for calculating the selected feature amount.

  Specifically, the sample group selection unit 77 uses, for example, (vsg [0], vsg [2]) or | (vsg [0], vsg [2]) | as a feature amount having a value closest to 1. If selected, the indexes “0” and “2” corresponding to the eigenvectors vsg [0] and vsg [2] are notified to the memory control unit 71 as index information.

  For example, based on the index information received from the sample group selection unit 77, the memory control unit 71 acquires a copy of the sample group SG [2] and the sample group RSG from the memory 72. Then, for example, the memory control unit 71 outputs the acquired copy of the sample group SG [2] as the current sample group, and outputs the copy of the sample group RSG to the monitoring unit 55 as the reference sample group.

[When the radio wave is later in time than multiple target periods]
The reference radio wave need not be a radio wave received by the array receiving unit 54 during a period in which the predetermined area is in a predetermined state. As a sequential method, a configuration in which a radio wave corresponding to a sample group later in time than a target sample group is used as a reference radio wave is also possible.

  Referring to FIG. 11 again, the interference removing unit 253, for example, based on the reference radio wave received by the array receiving unit 54 in a period later in time than the plurality of target periods, Select the radio wave from each radio wave corresponding to the period.

[Configuration of interference removing unit 253]
FIG. 13 is a diagram illustrating a configuration of an interference canceller in the receiver according to the second embodiment of the present invention.

  With reference to FIG. 13, the interference removal unit 253 includes a feature amount calculation unit 83 instead of the feature amount calculation unit 63 as compared with the interference removal unit 52 according to the first embodiment. The feature amount calculation unit 83 includes an eigenvector generation unit 85, an inner product calculation unit 86, and a sample group selection unit 77.

  Memory control unit 71 is similar to memory control unit 71 shown in FIG. 12, and thus detailed description will not be repeated. The memory control unit 71 causes the memory 72 to function as a ring buffer having an upper limit of 6, for example. In the memory 72, for example, sample groups SG [1] to SG [6] are stored in order from the newest. Six sampling periods corresponding to each of the sample groups SG [1] to SG [6] correspond to target periods T [N1] to T [N6], respectively.

  For example, the feature amount calculation unit 83 treats radio waves in a target period T [N1] that is temporally later than a plurality of target periods, that is, the target periods T [N2] to T [N6], as reference radio waves. For example, the feature amount calculation unit 83 calculates a value and a feature amount that represent the relationship between each radio wave in the target period T [N2] to T [N6] and the reference radio wave in the target period T [N1].

  Specifically, the eigenvector generation unit 85 in the feature quantity calculation unit 83 generates a normalized eigenvector from the sample group stored in the memory 72 when receiving an update completion notification from the memory control unit 71, for example.

  More specifically, the eigenvector generation unit 85, for example, according to the calculation procedure shown in Patent Document 1, from each of the sample groups SG [1] to SG [6], radio waves in the target periods T [N1] to T [N6]. Eigenvectors vsg [1] to vsg [6] are generated. The sign of the eigenvector is arbitrary. For this reason, the eigenvector may have either a positive sign or a negative sign.

  The eigenvector generation unit 85 outputs the generated eigenvectors vsg [1] to vsg [6] to the inner product calculation unit 86, for example.

  For example, when the eigenvector generation unit 85 specifies how to determine the sign of the eigenvector, the inner product calculation unit 86 has eigenvectors vs2 [2] to corresponding to radio waves in the target periods T [N2] to T [N6], respectively. Each inner product value of vsg [6] and the eigenvector vsg [1] of the radio wave in the target period T [N1] treated as the reference radio wave is calculated as a feature amount.

  Specifically, the inner product calculation unit 86, for example, (vsg [1], vsg [2]), (vsg [1], vsg [3]), (vsg [1], vsg [4]), (Vsg [1], vsg [5]) and (vsg [1], vsg [6]) are calculated as feature amounts, and the calculated feature amounts are output to the sample group selection unit 77.

  For example, when the method of determining the sign of the eigenvector in the eigenvector generation unit 85 is arbitrary, the inner product calculation unit 86 has eigenvectors vs [2] respectively corresponding to radio waves in the target periods T [N2] to T [N6]. A value based on each inner product value between the vsg [6] and the eigenvector vsv [1] of the radio wave in the target period T [N1] treated as the reference radio wave, that is, an absolute value of each inner product value is calculated as a feature amount.

  Specifically, the inner product calculation unit 86, for example, | (vsg [1], vsg [2]) |, | (vsg [1], vsg [3]) |, | (vsg [1], vsg [4]) |, | (vsg [1], vsg [5]) | and | (vsg [1], vsg [6]) | are calculated as feature quantities, and the calculated feature quantities are sample group selection unit 77. Output to.

  Each feature amount corresponds to, for example, a spatial feature amount P (t) in each target period corresponding to the eigenvectors vsg [2] to vsg [6] used for the calculation.

  For example, the sample group selection unit 77 selects a feature quantity corresponding to a radio wave in the target period that has the highest degree of coincidence with a radio wave treated as a reference radio wave among a plurality of target periods, and corresponds to the selected feature quantity The sample group to be used for calculation of the spatial feature amount is selected from the sample groups SG [N2] to SG [N6] stored in the memory 72 using the index.

  Specifically, for example, when the sample group selection unit 77 receives an inner product value or an absolute value of the inner product value as a feature amount from the inner product calculation unit 86, the feature amount closest to a predetermined value among the received feature amounts. Select. More specifically, since the eigenvector is standardized, the sample group selection unit 77 selects, for example, a feature quantity having a value closest to 1 from the received feature quantities.

  Here, the feature amount selected by the sample group selection unit 77 is the same as the spatial feature amount P (t) calculated by the monitoring unit 55 in the subsequent stage, for example.

  For example, the sample group selection unit 77 notifies the memory control unit 71 of index information indicating an index corresponding to the eigenvector used for calculating the selected feature amount.

  Specifically, the sample group selection unit 77 uses, for example, (vsg [1], vsg [2]) or | (vsg [1], vsg [2]) | as a feature amount having a value closest to 1. When selected, the indexes “1” and “2” corresponding to the eigenvectors vsg [1] and vsg [2] are notified to the memory control unit 71 as index information.

  For example, based on the index information received from the sample group selection unit 77, the memory control unit 71 acquires a copy of the sample groups SG [1] and SG [2] from the memory 72. Then, for example, the memory control unit 71 outputs the acquired copy of the sample group SG [2] to the monitoring unit 55 as the current sample group and the copy of the sample group SG [1] as the reference sample group.

  Note that the feature quantity calculation unit 83 is configured to handle radio waves in the target period T [N1] immediately after the target periods T [N2] to T [N6] corresponding to the radio wave whose inner product value is to be calculated as reference radio waves. However, the present invention is not limited to this. For example, the feature amount calculation unit 83 may be configured to further delay and handle radio waves in the target period after the target period T [N1] as reference radio waves.

  However, the smaller the difference between the time at which the signal for calculating the inner product value is sampled and the time at which the reference signal is sampled, the more the influence of various noises such as Gaussian noise can be reduced. Delay can be suppressed. Therefore, it is preferable that the radio wave in the target period with a smaller delay than the target period corresponding to the radio wave for which the inner product value is calculated be handled as the reference radio wave.

  The feature amount calculation unit 83 is configured to handle the radio wave in the target period T [N7] immediately before the target period T [N2] to T [N6] corresponding to the radio wave whose inner product value is to be calculated as the reference radio wave. Alternatively, the radio wave in the target period before the target period T [N8] may be handled as the reference radio wave.

[When the reference radio wave is not prepared]
Referring to FIG. 11 again, for example, in three or more target periods, interference removal unit 254 is characterized by a value representing the relationship between radio waves corresponding to each combination for all combinations of radio waves in two target periods. Calculate as a quantity. For example, the interference removal unit 254 selects a radio wave from each radio wave corresponding to the three or more target periods based on the calculated feature amount.

[Configuration of interference removing unit 254]
FIG. 14 is a diagram illustrating a configuration of an interference canceller in the receiver according to the second embodiment of the present invention.

  Referring to FIG. 14, interference removal unit 254 includes a feature amount calculation unit 93 instead of feature amount calculation unit 63, as compared with interference removal unit 52 according to the first embodiment. The feature amount calculation unit 93 includes an eigenvector generation unit 95, an inner product calculation unit 96, a sample group selection unit 97, and a combination setting unit 98.

  Memory control unit 71 is similar to memory control unit 71 shown in FIG. 12, and thus detailed description will not be repeated. The memory control unit 71 causes the memory 72 to function as a ring buffer having an upper limit of 5, for example. In the memory 72, for example, sample groups SG [1] to SG [5] are stored in order from the newest. Five sampling periods corresponding to each of the sample groups SG [1] to SG [5] correspond to the target periods T [N1] to T [N5], respectively.

  Although the memory control unit 71 is configured to store the sample groups corresponding to the five target periods in the memory 72, the present invention is not limited to this. For example, the memory control unit 71 may be configured to store the sample group corresponding to three or more target periods in the memory 72.

  For example, the feature quantity calculation unit 93 does not use a reference radio wave. Specifically, for example, the feature amount calculation unit 93 calculates, as the feature amount, a value representing the relationship between the radio waves corresponding to each combination for all combinations of radio waves in the target period T [N1] to T [N5]. To do.

  Specifically, the eigenvector generation unit 95 in the feature quantity calculation unit 93 generates a normalized eigenvector from the sample group stored in the memory 72 when receiving an update completion notification from the memory control unit 71, for example.

  More specifically, the eigenvector generation unit 95, for example, according to the calculation procedure shown in Patent Document 1, from each of the sample groups SG [1] to SG [5], radio waves in the target periods T [N1] to T [N5]. Eigenvectors vsg [1] to vsg [5] are generated. The sign of the eigenvector is arbitrary. For this reason, the eigenvector may have either a positive sign or a negative sign.

  The eigenvector generation unit 95 outputs the generated eigenvectors vsg [1] to vsg [5] to the combination setting unit 98, for example.

  For example, the combination setting unit 98 sets all combinations of eigenvectors corresponding to two target periods in eigenvectors corresponding to three or more target periods.

  Specifically, when the combination setting unit 98 receives eigenvectors vsg [1] to vsg [5] corresponding to the target periods T [N1] to T [N5] from the eigenvector generation unit 95, for example, the received eigenvector vsg For [1] to vsg [5], C (5,2) = 10 combinations are set. The combination setting unit 98 outputs the set combination and eigenvectors vsg [1] to vsg [5] to the inner product calculation unit 96, for example.

  For example, when the eigenvector generation unit 95 specifies how to determine the sign of the eigenvector, the inner product calculation unit 96 calculates each inner product value of eigenvectors corresponding to each combination received from the combination setting unit 98 as a feature amount. .

  Specifically, the inner product calculation unit 96, for example, (vsg [1], vsg [2]), (vsg [1], vsg [3]), (vsg [1], vsg [4]), (Vsg [1], vsg [5]), (vsg [2], vsg [3]), (vsg [2], vsg [4]), (vsg [2], vsg [5]), (vsg [3], vsg [4]), (vsg [3], vsg [5]) and (vsg [4], vsg [5]) are calculated as feature quantities, and the calculated feature quantities are sample group selection unit 97. Output to.

  Further, the inner product calculation unit 96, for example, when the method of determining the sign of the eigenvector in the eigenvector generation unit 95 is arbitrary, a value based on each inner product value of eigenvectors corresponding to each combination received from the combination setting unit 98, that is, The absolute value of each inner product value is calculated as a feature value.

  Specifically, the inner product calculation unit 96, for example, | (vsg [1], vsg [2]) |, | (vsg [1], vsg [3]) |, | (vsg [1], vsg [4]) |, | (vsg [1], vsg [5]) |, | (vsg [2], vsg [3]) |, | (vsg [2], vsg [4]) |, | ( vsg [2], vsg [5]) |, | (vsg [3], vsg [4]) |, | (vsg [3], vsg [5]) | and | (vsg [4], vsg [5] ]) Is calculated as a feature value, and the calculated feature value is output to the sample group selection unit 97.

  Each feature amount corresponds to, for example, the spatial feature amount P (t) in any one of the two target periods corresponding to the eigenvectors vsg [1] to vsg [5] used for the calculation.

  For example, when the sample group selection unit 97 receives the inner product value or the absolute value of the inner product value as the feature value from the inner product calculation unit 96, the sample group selection unit 97 selects the feature value closest to the predetermined value from the received feature values. Specifically, since the eigenvector is standardized, the sample group selection unit 97 selects, for example, a feature quantity having a value closest to 1 from the received feature quantities.

  Here, the feature amount selected by the sample group selection unit 97 is, for example, the same as the spatial feature amount P (t) calculated by the monitoring unit 55 in the subsequent stage.

  For example, the sample group selection unit 97 notifies the memory control unit 71 of index information indicating an index corresponding to the eigenvector used for calculating the selected feature amount.

  Specifically, for example, the sample group selection unit 97 uses (vsg [2], vsg [3]) or | (vsg [2], vsg [3]) | as a feature quantity having a value closest to 1. When selected, the indexes “2” and “3” corresponding to the eigenvectors vsg [2] and vsg [3] are notified to the memory control unit 71 as index information.

  For example, based on the index information received from the sample group selection unit 97, the memory control unit 71 acquires a copy of the sample groups SG [2] and SG [3] from the memory 72. Then, for example, the memory control unit 71 outputs the acquired copy of the sample group SG [2] to the monitoring unit 55 as the current sample group and the copy of the sample group SG [3] as the reference sample group. Note that the memory control unit 71 may output, for example, a copy of the acquired sample group SG [3] as the current sample group and a copy of the sample group SG [2] to the monitoring unit 55 as the reference sample group. .

  Referring to FIG. 11 again, the spatial feature quantity calculation unit 211 in the monitoring unit 55 is based on, for example, the current sample group and the reference sample group received from the interference removal unit 252, 253, or 254, and the eigenvector vob at the observation time t. (T) and an initial vector vno are generated respectively.

  The spatial feature quantity calculation unit 211 extracts the spatial feature quantity P (t) by, for example, calculating the inner product of the calculated eigenvector vob (t) and the initial vector vno.

[Modification]
As described above, in the receiver 102, a part of the inner product calculation performed in the inner product calculation units 76, 86, and 96 is also performed in the spatial feature amount calculation unit 211, and each time the memory 72 is updated. Since eigenvectors are generated from the sample group, the processing load is heavy. Accordingly, a modification in which the processing load in the receiver 102 is reduced will be described below.

FIG. 15 is a diagram illustrating a configuration of a modification of the receiver illustrated in FIG. 11.
Referring to FIG. 15, compared with receiver 102, receiver 103 replaces interference removing unit 252, 253 or 254, and monitoring unit 55 with interference removing unit 255, 256 or 257, and monitoring unit 56. With. Compared with the monitoring unit 55, the monitoring unit 56 includes an event detection unit 212 instead of the event detection unit 12 and the spatial feature amount calculation unit 211.

[Modification example based on past radio waves from multiple target periods]
FIG. 16 is a diagram illustrating a configuration of a modified example of the interference removing unit illustrated in FIG. 12.

  Referring to FIG. 16, interference removal unit 255 has memory control unit 171 and feature amount calculation unit 173 instead of memory control unit 71 and feature amount calculation unit 73 compared to interference removal unit 252 shown in FIG. 12. Including. The feature amount calculation unit 173 includes an eigenvector generation unit 175, an inner product calculation unit 176, and an inner product value selection unit 177.

  For example, the memory control unit 171 performs the update process of the memory 72 based on the ID conformity notification, the ID nonconformity notification, and the sampling instruction signal received from the received signal monitoring unit 51.

  Specifically, the memory control unit 171 causes the memory 72 to function as a ring buffer, for example. That is, the memory control unit 171 accumulates eigenvectors in the memory 72 up to the upper limit number, for example. For example, when accumulating new eigenvectors in the memory 72 in a state where five eigenvectors are accumulated as the upper limit number, the memory control unit 171 discards the oldest eigenvector among the five eigenvectors and then stores the new eigenvectors in the memory 72. To accumulate.

  More specifically, the memory control unit 171 receives, for example, a digital signal received from the A / D converter 34 as a new sample group SGnew in the memory 72 while receiving a sampling instruction signal from the reception signal monitoring unit 51 during the sampling period. accumulate.

  For example, when the accumulation of the sample group SGnew is completed, the memory control unit 171 causes the eigenvector generation unit 175 to generate a normalized eigenvector.

  For example, the eigenvector generation unit 175 generates an eigenvector vsgnew from the sample group SGnew. The sign of the eigenvector is arbitrary. For this reason, the eigenvector may have either a positive sign or a negative sign.

  And the memory control part 171 will cancel | discard eigenvector vsgnnew, for example, when ID nonconformity notification is received from the received signal monitoring part 51. FIG. For example, when receiving the ID conformity notification from the reception signal monitoring unit 51, the memory control unit 171 updates the five eigenvectors vsg [1] to vsg [5] to indicate that the update processing of the memory 72 is completed. An update completion notification is output to the feature amount calculation unit 173.

  At this time, for example, the memory control unit 171 discards the oldest eigenvectors among the eigenvectors vsg [1] to vsg [5] before the update, and then sets the remaining eigenvectors and eigenvectors vsgnew in order of eigenvectors vsg [1] to vsg. Store in [5]. That is, the memory 72 stores eigenvectors generated based on the radio signal transmitted from the transmitter 151.

  Therefore, five sampling periods corresponding to each of the eigenvectors vsg [1] to vsg [5] stored in the memory 72 correspond to the target periods T [N1] to T [N5], respectively.

  Further, for example, the memory 72 stores the reference radio wave received by the array receiving unit 54 during a period in which the predetermined area is in a predetermined state before the target period T [N1] to T [N5]. Eigenvectors vr generated from radio waves are accumulated.

  Although the memory control unit 171 is configured to store the eigenvectors corresponding to the five target periods in the memory 72, the present invention is not limited to this. For example, the memory control unit 171 may be configured to store eigenvectors corresponding to two or more target periods in the memory 72.

  Further, for example, after receiving the ID conformity notification from the reception signal monitoring unit 51, the memory control unit 171 may cause the eigenvector generation unit 175 to generate a normalized eigenvector. In this case, the memory control unit 171 can complete the update process of the memory 72 without discarding the eigenvector generated by the eigenvector generation unit 175.

  The feature amount calculation unit 173 calculates the feature amount of the radio signal received by the array reception unit 54 based on, for example, a plurality of eigenvectors stored in the memory 72.

  Specifically, when the inner product calculation unit 176 in the feature quantity calculation unit 173 receives an update completion notification from the memory control unit 171, for example, the eigenvectors vsg [1] to vsg [5] stored in the memory 72 and the eigenvectors Extract vr.

  For example, when the eigenvector generation unit 175 specifies how to determine the sign of the eigenvector, the inner product calculation unit 176 has eigenvectors vs [1] to corresponding to radio waves in the target periods T [N1] to T [N5], respectively. Each inner product value of vsg [5] and the eigenvector vr of the reference radio wave is calculated as a feature amount.

  For example, when the eigenvector generation unit 175 determines how to determine the sign of the eigenvector, the inner product calculation unit 176 has eigenvectors vs [1] corresponding to radio waves in the target periods T [N1] to T [N5], for example. A value based on each inner product value of ~ vsg [5] and the eigenvector vr of the reference radio wave, that is, an absolute value of each inner product value is calculated as a feature amount. For example, the inner product calculation unit 176 outputs the calculated feature value to the inner product value selection unit 177.

  For example, when the inner product value selection unit 177 receives the inner product value or the absolute value of the inner product value as the feature value from the inner product calculation unit 176, the inner product value selection unit 177 selects the feature value closest to the predetermined value from the received feature values. Specifically, since the eigenvector is standardized, the inner product value selection unit 177 selects, for example, a feature amount having a value closest to 1 from the received feature amounts, and monitors the selected feature amount. To 56.

  Here, the feature quantity selected by the inner product value selection unit 177 is, for example, a spatial feature quantity P (calculated based on radio waves in the target period in which the degree of coincidence with the reference radio wave is the largest among a plurality of target periods. t).

  In addition, in the interference removal unit 255 according to the second embodiment of the present invention, every time the update process of the memory 72 is completed, five (vr, vsg [1]) to (vr, vsg [5]) Although the inner product value or the absolute value of the five inner product values of | (vr, vsg [1]) | to | (vr, vsg [5]) | is calculated, the present invention is not limited to this. Absent.

  For example, the interference removal unit 255 stores the calculated inner product value or a part of the absolute value of the inner product value in the memory 72, and uses the accumulated inner product value or the absolute value of the inner product value as a feature amount at the next update timing. A configuration using values may also be used.

  Specifically, as an example of the case where the interference removal unit 255 uses the inner product value, the interference removal unit 255 is characterized by the inner product values (vr, vsg [1]) to (vr, vsg [5]), for example. When the inner product value selected as the quantity is output to the monitoring unit 56, (vr, vsg [1]) to (vr, vsg [4]) other than the oldest inner product value are stored in the memory 72.

  Then, for example, when the update process of the next memory 72 is completed, the interference removal unit 255 calculates the inner product value (vr, vsgnew), and calculates the calculated inner area (vr, vsgnew) and the inner product value accumulated in the memory 72. The inner product value selected as the feature amount is output to the monitoring unit 56.

  Thereby, in the interference removal part 255, since the frequency | count of calculating an inner product value or the absolute value of an inner product value can be reduced, the processing load in the receiver 103 can be reduced.

[Modification example based on radio waves later in time than multiple target periods]
FIG. 17 is a diagram illustrating a configuration of a modified example of the interference removing unit illustrated in FIG. 13.

  Referring to FIG. 17, interference removal unit 256 has memory control unit 171 and feature amount calculation unit 183 instead of memory control unit 71 and feature amount calculation unit 83 compared to interference removal unit 253 shown in FIG. 13. Including. The feature amount calculation unit 183 includes an eigenvector generation unit 175, an inner product calculation unit 186, and an inner product value selection unit 177.

  Memory control unit 171 and eigenvector generation unit 175 are the same as memory control unit 171 and eigenvector generation unit 175 shown in FIG. 16, respectively, and therefore detailed description will not be repeated.

  The memory control unit 171 causes the memory 72 to function as a ring buffer having an upper limit of 6, for example. In the memory 72, eigenvectors vsg [1] to vsg [6] are stored, for example, in order from the newest. Six sampling periods corresponding to each of the eigenvectors vs [1] to vsg [6] correspond to the target periods T [N1] to T [N6], respectively.

  For example, the feature quantity calculation unit 183 calculates the feature quantity of the radio signal received by the array reception unit 54 based on a plurality of eigenvectors stored in the memory 72.

  Specifically, the inner product calculation unit 186 in the feature quantity calculation unit 183 extracts eigenvectors vsg [1] to vsg [6] stored in the memory 72 when receiving an update completion notification from the memory control unit 171, for example. To do.

  For example, when the eigenvector generation unit 175 specifies how to determine the sign of the eigenvector, the inner product calculation unit 186 has eigenvectors vs [2] to corresponding to radio waves in the target periods T [N2] to T [N6], respectively. Each inner product value of vsg [6] and the eigenvector vsg [1] of the radio wave in the target period T [N1] treated as the reference radio wave is calculated as a feature amount.

  Further, for example, when the eigenvector generation unit 175 determines how to determine the sign of the eigenvector, the inner product calculation unit 186 has eigenvectors vs [2] respectively corresponding to radio waves in the target periods T [N2] to T [N6]. A value based on each inner product value between the vsg [6] and the eigenvector vsv [1] of the radio wave in the target period T [N1] treated as the reference radio wave, that is, an absolute value of each inner product value is calculated as a feature amount. For example, the inner product calculation unit 186 outputs the calculated feature value to the inner product value selection unit 177.

  For example, when the inner product value selection unit 177 receives the inner product value or the absolute value of the inner product value as the feature value from the inner product calculation unit 186, the inner product value selection unit 177 selects the feature value closest to the predetermined value from the received feature values. Specifically, since the eigenvector is standardized, the inner product value selection unit 177 selects, for example, a feature amount having a value closest to 1 from the received feature amounts, and monitors the selected feature amount. To 56.

  Here, the feature quantity selected by the inner product value selection unit 177 is, for example, a space calculated based on radio waves in the target period in which the degree of coincidence with the radio waves treated as reference radio waves is the largest among the plurality of target periods. This is a feature amount P (t).

[Modification example when the reference radio wave is not prepared]
FIG. 18 is a diagram illustrating a configuration of a modified example of the interference removing unit illustrated in FIG. 14.

  Referring to FIG. 18, interference removal unit 257 has memory control unit 171 and feature amount calculation unit 193 instead of memory control unit 71 and feature amount calculation unit 93 compared to interference removal unit 254 shown in FIG. 14. Including. The feature quantity calculation unit 193 includes an eigenvector generation unit 175, an inner product calculation unit 196, an inner product value selection unit 197, and a combination setting unit 198.

  Memory control unit 171 and eigenvector generation unit 175 are the same as memory control unit 171 and eigenvector generation unit 175 shown in FIG. 16, respectively, and therefore detailed description will not be repeated.

  The memory control unit 171 causes the memory 72 to function as a ring buffer having an upper limit of 5, for example. In the memory 72, eigenvectors vsg [1] to vsg [5] are stored, for example, in order from the newest. Five sampling periods corresponding to each of the eigenvectors vs [1] to vsg [5] correspond to the target periods T [N1] to T [N5], respectively.

  Although the memory control unit 171 is configured to store the eigenvectors corresponding to the five target periods in the memory 72, the present invention is not limited to this. For example, the memory control unit 171 may be configured to store eigenvectors corresponding to three or more target periods in the memory 72.

  For example, the feature amount calculation unit 193 calculates the feature amount of the radio signal received by the array reception unit 54 based on a plurality of eigenvectors stored in the memory 72.

  Specifically, the combination setting unit 198 in the feature quantity calculation unit 193 sets, for example, all combinations of eigenvectors corresponding to two target periods in eigenvectors corresponding to three or more target periods.

  Specifically, for example, when the combination setting unit 198 receives an update completion notification from the memory control unit 171, the eigenvectors vsg [1] to vsg [5] corresponding to the target periods T [N1] to T [N5], respectively. Are extracted from the memory 72, and C (5,2) = 10 combinations are set for the extracted eigenvectors vsg [1] to vsg [5]. The combination setting unit 198 outputs the set combination and eigenvectors vsg [1] to vsg [5] to the inner product calculation unit 196, for example.

  For example, when the eigenvector generation unit 175 specifies how to determine the sign of the eigenvector, the inner product calculation unit 196 calculates each inner product value of eigenvectors corresponding to each combination received from the combination setting unit 198 as a feature amount. .

  Further, for example, when the eigenvector generation unit 175 determines how to determine the sign of the eigenvector, the inner product calculation unit 196 has a value based on each inner product value of eigenvectors corresponding to each combination received from the combination setting unit 198, that is, The absolute value of each inner product value is calculated as a feature value. For example, the inner product calculation unit 196 outputs the calculated feature value to the inner product value selection unit 197.

  For example, when the inner product value selection unit 197 receives the inner product value or the absolute value of the inner product value as the feature value from the inner product calculation unit 196, the inner product value selection unit 197 selects the feature value closest to the predetermined value from the received feature values. Specifically, since the eigenvector is standardized, the inner product value selection unit 197 selects, for example, a feature amount having a value closest to 1 from the received feature amounts, and monitors the selected feature amount. To 56.

  Here, the feature value selected by the inner product value selection unit 197 is, for example, a spatial feature value P (t calculated based on a radio wave having the highest degree of coincidence among radio waves corresponding to a plurality of target periods. ).

  In addition, in the interference removal unit 257 according to the second embodiment of the present invention, the inner product value or the absolute value of the inner product value corresponding to C (5,2) = 10 combinations each time the update process of the memory 72 is completed. Although the configuration is such that the value is calculated, the present invention is not limited to this.

  The interference removal unit 257 stores, for example, the calculated inner product value or a part of the absolute value of the inner product value in the memory 72, and stores the accumulated inner product value or the absolute value of the inner product value as a feature amount at the next update timing. A configuration using values may also be used.

  Specifically, as an example of the case where the interference removal unit 257 uses the inner product value, the interference removal unit 257, for example, selects the inner product value selected as the feature value from the inner product values corresponding to the ten combinations. Is output, the inner product value other than the inner product value using the oldest eigenvector vsg [5] is stored in the memory 72.

  Then, for example, when the update processing of the next memory 72 is completed, the interference removal unit 257 updates the updated eigenvectors vsg [2], vsg [3], vsg [4], and vsg [5], and the updated eigenvectors. The inner product value with respect to vsg [1], that is, vsgnew is calculated. Then, the interference removal unit 257 outputs, for example, the inner product value selected as the feature amount from the calculated inner product value and the inner product value accumulated in the memory 72 to the monitoring unit 56.

  Thereby, in the interference removal part 257, since the frequency | count of calculating the inner product value or the absolute value of an inner product value can be reduced, the processing load in the receiver 103 can be reduced.

  Referring to FIG. 15 again, the event detection unit 212 in the monitoring unit 56 monitors a human action in a predetermined area based on the feature amount received from the interference removal unit 255, 256, or 257. Specifically, for example, when the feature amount is less than a predetermined threshold, the event detection unit 212 determines that a human has entered a predetermined area.

  Since other configurations and operations are the same as those of the intrusion detection system according to the first embodiment, detailed description thereof will not be repeated here.

  As described above, in the monitoring apparatus according to the second embodiment of the present invention, the array receiver 54 receives radio waves transmitted from the transmitter 151 arranged in a predetermined area. The feature amount calculation unit 73, 83, 93, 173, 183, or 193 calculates the feature amount of the radio wave in each of the plurality of target periods received by the array reception unit 54. The monitoring unit 55 or 56 monitors a human operation in a predetermined area based on radio waves in a target period in which the feature amount satisfies a predetermined condition.

  With such a configuration, for example, a radio wave that does not include an interference wave can be selected from radio waves in each of a plurality of target periods based on the calculated feature amount. It is possible to suppress erroneous detection due to processing of included radio waves and improve detection accuracy.

  This makes it possible to correctly monitor human actions in a predetermined area, so that it is possible to avoid the occurrence of false alarms when the receivers 102 and 103 are used for intrusion detection.

  In the monitoring apparatus according to the second embodiment of the present invention, the feature amount calculation unit 73, 83, 173, or 183 calculates a value representing the relationship between the radio wave and the reference radio wave in the target period as the feature amount. .

  With such a configuration, the relationship between the radio wave and the reference radio wave in the target period can be quantified. Accordingly, for example, when a radio wave that does not include an interference wave is used as a reference radio wave, a radio wave that does not include an interference wave can be selected from radio waves in each of a plurality of target periods.

  In the monitoring apparatus according to the second embodiment of the present invention, the feature amount calculation unit 73, 83, 173, or 183 calculates the inner product value or inner product value of the eigenvector of the radio wave and the eigenvector of the reference radio wave in the target period. The base value is calculated as a feature amount.

  In this way, by calculating the inner product value that changes according to the relationship between radio waves or a value based on the inner product value as a feature amount, the relationship between the radio wave and the reference radio wave in the target period can be quantified with a simple calculation. Therefore, it is possible to easily select a radio wave that does not include an interference wave from radio waves in each of a plurality of target periods.

  In the monitoring apparatus according to the second embodiment of the present invention, the reference radio wave is a radio wave received by the array receiver 54 in a period later in time than the plurality of target periods.

  With such a configuration, the reference radio wave handled as a reference can be sequentially changed to a newer radio wave, so that it is possible to flexibly cope with changes in the radio wave environment of the receivers 102 and 103.

  Specifically, based on reference radio waves according to changes in the surrounding environment of the receivers 102 and 103, such as temperature or noise level, radio waves that do not include interference waves from among radio waves in each of a plurality of target periods. Since the selection can be made, it is possible to stably monitor the human motion in the predetermined area.

  For example, when dynamically changing the reference radio wave to be treated as a standard, when the changed reference radio wave is inappropriate, specifically when the changed reference radio wave contains interference waves, multiple The radio waves that do not include interference waves cannot be accurately selected from the radio waves in each period, and the operations of the receivers 102 and 103 may become unstable.

  On the other hand, in the monitoring apparatus according to the second embodiment of the present invention, the reference radio wave is transmitted to the array receiving unit during a period in which the predetermined area is in a predetermined state before the plurality of target periods. 54 is a received radio wave.

  As described above, the configuration using the reference radio wave serving as a specific standard can suppress inappropriate use of the reference radio wave, and can stably calculate the feature amount.

  As a result, radio waves that do not include interference waves can be accurately selected from radio waves in each of a plurality of target periods based on absolute evaluation, so that the reliability of the monitoring operation can be increased.

  For example, when setting a reference radio wave to be handled as a standard, if the set reference radio wave is inappropriate, specifically, if the changed reference radio wave contains an interference wave, each of multiple target periods In other words, it becomes impossible to accurately select a radio wave that does not include an interference wave from among the radio waves in, and the operations of the receivers 102 and 103 may become unstable.

  On the other hand, in the monitoring apparatus according to the second embodiment of the present invention, the feature amount calculation unit 93 or 193 calculates a value representing a relationship between radio waves in two target periods as a feature amount. And the feature-value calculation part 93 or 193 calculates the value showing the said relationship about all the combinations of the electromagnetic wave in two target periods in three or more target periods.

  As described above, the configuration in which the reference radio wave is not set can avoid unstable operations of the receivers 102 and 103 that occur when the set reference radio wave is inappropriate. Further, for example, even when the target predetermined area is changed, it is not necessary to newly set a reference radio wave, so that the processing in the receivers 102 and 103 can be simplified.

  In addition, since the three or more target periods are sequentially updated, it is possible to flexibly cope with changes in the radio wave environment of the receivers 102 and 103.

  Specifically, interference waves are included from among the radio waves in each of a plurality of target periods based on values representing the relationship between radio waves according to changes in the surrounding environment of the receivers 102 and 103, such as temperature or noise level. Since it is possible to select a radio wave that cannot be transmitted, it is possible to stably monitor human movements in a predetermined area.

  In addition, since the relationship between radio waves in two target periods can be quantified, radio waves that do not include interference waves among radio waves in each of three or more target periods can be easily obtained using the quantified relationship. Can be selected.

  Moreover, in the monitoring apparatus according to the second embodiment of the present invention, the monitoring unit 55 or 56 is based on the radio wave in the target period that has the highest degree of coincidence with the reference radio wave among the plurality of target periods. Monitor human movements in a given area.

  With such a configuration, it is possible to select the target period that has the closest relationship with the reference radio wave from among the radio waves in each of the plurality of target periods. It is possible to correctly monitor human movements in the area.

  In addition, since the value representing the relationship between the radio wave and the reference radio wave in the target period that has the highest degree of coincidence with the reference radio wave can be temporally smoothed in the plurality of target periods including the target period. The occurrence of false detection can be suppressed.

  Moreover, in the monitoring apparatus according to the second embodiment of the present invention, the monitoring unit 55 or 56 has a predetermined area based on radio waves in a target period in which a feature amount is closest to a predetermined value among a plurality of target periods. To monitor human movements.

  With such a configuration, it is possible to quantitatively and easily select the target period that has the closest relationship with the reference radio wave from among a plurality of target periods, so that the processing efficiency in the receivers 102 and 103 is improved. Can do.

  Next, another embodiment of the present invention will be described with reference to the drawings. In the drawings, the same or corresponding parts are denoted by the same reference numerals and description thereof will not be repeated.

<Third Embodiment>
The present embodiment relates to a monitoring device in which the purpose of use is changed as compared with the monitoring devices according to the first embodiment and the second embodiment. The contents other than those described below are the same as those of the intrusion detection system according to the first embodiment.

  In the intrusion detection device, that is, the receiver (monitoring device) 101 according to the first embodiment of the present invention, the event detection unit 12 is based on the spatial feature amount P (t) calculated by the spatial feature amount calculation unit 11. Detecting human intrusion or suspicious behavior start in the predetermined area as human action in the predetermined area.

  On the other hand, in the watching device, that is, the receiver (monitoring device) 101 in the watching system 202 according to the third embodiment of the present invention, as a human action in the predetermined area, a person in the predetermined area, specifically watching. Detects inactivity or minor activity of the subject.

  More specifically, the event detection unit 12 in the monitoring unit 53 of the receiver 101 determines whether there is no human action in a predetermined area based on the spatial feature amount P (t) calculated by the spatial feature amount calculation unit 11 or Detect low. Specifically, for example, the event detection unit 12 monitors whether or not there is a person who has not moved for a predetermined time due to occurrence of an abnormality such as a heart attack in a predetermined area. This predetermined area is, for example, an area where one or a plurality of humans are performing an action such as walking during normal times.

  Similarly to the intrusion detection device according to the first embodiment of the present invention, the spatial feature quantity calculation unit 11 in the monitoring unit 53 of the receiver 101 obtains the initial vector vno and the eigenvector vob (t) at the observation time t. The spatial feature amount P (t) at the observation time t is calculated.

  Here, the initial vector vno used as a comparison reference in the receiver 101 is, for example, an eigenvector when no human is present in a predetermined area.

  Therefore, as the spatial feature P (t) is smaller than “1”, the state of the predetermined area at the observation time t is closer to the normal state where one or more people are moving.

  For this reason, the event detection unit 12 determines that there is no human activity in the predetermined area when the state in which the spatial feature amount P (t) is equal to or greater than the predetermined threshold continues for a predetermined time or longer.

  Specifically, assuming that the predetermined threshold value is “0.999”, the event detection unit 12 determines that the human being in the predetermined area when the spatial feature amount P (t) is smaller than “0.999”. It is determined that this is a normal state with the operation. On the other hand, when the state where the spatial feature P (t) is “0.999” or more continues for a predetermined time or more, the event detection unit 12 determines that there is no human activity or a small abnormal state in the predetermined area. To do.

  Since other configurations and operations are the same as those of the intrusion detection system according to the first embodiment, detailed description thereof will not be repeated here.

  For example, in an event detection apparatus used for watching purposes, even though an abnormality has actually occurred in a predetermined area, it may be determined to be normal due to erroneous detection, and a false alarm may occur.

  On the other hand, in the monitoring device according to the third embodiment of the present invention, the array receiver 54 receives radio waves transmitted from the transmitter 151 arranged in a predetermined area. The feature amount calculation unit 63 calculates the feature amount of the radio wave in each of a plurality of target periods received by the array reception unit 54. The monitoring unit 53 monitors human actions in a predetermined area based on radio waves in a target period in which the feature amount satisfies a predetermined condition.

  With such a configuration, for example, a radio wave that does not include an interference wave can be selected from radio waves in each of a plurality of target periods based on the calculated feature amount. Therefore, the receiver 101 further includes an interference wave. False detection due to processing of radio waves can be suppressed, and detection accuracy can be improved.

  As a result, it is possible to correctly monitor human movements in a predetermined area, so that it is possible to avoid the occurrence of misreporting when the receiver 101 is used for watching purposes.

  In the watching system according to the third embodiment of the present invention, the receiver 101 detects that there is no or little human action in a predetermined area.

  With such a configuration, it is possible to satisfactorily watch over a person in a predetermined area.

  In the monitoring system according to the third embodiment of the present invention, it has been described that the contents other than those described above are the same as the intrusion detection system 201 according to the first embodiment of the present invention. It may be the same as the intrusion detection system according to the second embodiment of the invention.

  More specifically, the event detection unit 12 in the monitoring unit 55 of the receiver 102 determines whether there is no human action in a predetermined area based on the spatial feature amount P (t) calculated by the spatial feature amount calculation unit 211 or Detect low. In addition, the event detection unit 212 in the monitoring unit 56 of the receiver 103 performs a human action in a predetermined area based on the feature amount calculated by the feature amount calculation unit 173, 183, or 193 in the interference removal unit 255, 256, or 257. Detect that there are no or few.

  As described above, in the monitoring apparatus according to the third embodiment of the present invention, the array receiver 54 receives radio waves transmitted from the transmitter 151 arranged in a predetermined area. The feature amount calculation unit 73, 83, 93, 173, 183, or 193 calculates the feature amount of the radio wave in each of the plurality of target periods received by the array reception unit 54. The monitoring unit 55 or 56 monitors a human operation in a predetermined area based on radio waves in a target period in which the feature amount satisfies a predetermined condition.

  With such a configuration, for example, a radio wave that does not include an interference wave can be selected from radio waves in each of a plurality of target periods based on the calculated feature amount. It is possible to suppress erroneous detection due to processing of included radio waves and improve detection accuracy.

  As a result, it is possible to correctly monitor human actions in a predetermined area, so that it is possible to avoid the occurrence of false alarms when the receivers 102 and 103 are used for monitoring purposes.

  Note that the watching device according to the third embodiment of the present invention is configured to perform a binary determination of the presence or absence of human movement or not, but is not limited thereto. It may be configured to output an index indicating the level of possibility of human inactivity or low activity.

  In addition, the intrusion detection apparatus according to the first and second embodiments of the present invention and the monitoring apparatus according to the third embodiment of the present invention are configured to use eigenvectors as spatial feature quantities. However, the present invention is not limited to this. Not only the eigenvector but also a configuration using other spatial features such as a delay profile described in Non-Patent Document 1 may be used.

  In addition, in the intrusion detection device according to the first and second embodiments of the present invention and the monitoring device according to the third embodiment of the present invention, a one-dimensional feature value is used as the spatial feature value. Although it is the configuration, it is not limited to this. A configuration using a multidimensional feature value as the spatial feature value may be used. For example, the eigenvector itself can be used as the feature vector, or a delay profile as described in Non-Patent Document 1 can be used.

  Moreover, although the intrusion detection device according to the first embodiment and the second embodiment of the present invention and the monitoring device according to the third embodiment of the present invention are array type radio wave sensors, It may be a type of radio wave sensor.

  The above embodiment should be considered as illustrative in all points and not restrictive. The scope of the present invention is defined by the terms of the claims, rather than the description above, and is intended to include any modifications within the scope and meaning equivalent to the terms of the claims.

  The above description includes the following features.

[Appendix 1]
A receiver for receiving radio waves transmitted from a transmitter arranged in a predetermined area;
A feature amount calculating unit that calculates a feature amount of the radio wave in each of a plurality of target periods received by the receiving unit;
A monitoring unit that monitors a human operation in the predetermined area based on the radio wave in the target period in which the feature amount satisfies a predetermined condition;
The feature amount is an inner product value or an absolute value of an inner product value of the eigenvector of the radio wave and an eigenvector of the reference radio wave in the target period, or an absolute value of an inner product value or an inner product value of the eigenvectors of the radio wave in the target period , The average value or higher-order statistic of the received power of the radio wave in the target period, or the similarity between the frequency spectrum of the radio wave and the standard frequency spectrum in the target period,
The monitoring device is a monitoring device in which the predetermined condition is a condition in which the degree of coincidence between the eigenvectors is greatest, or a condition in which the feature amount is closest to a predetermined value.

11, 211 Spatial feature quantity calculation unit 12, 212 Event detection unit 21 Antenna unit 22 Reception circuit 31 Band pass filter 32 Low noise amplifier 33 Quadrature demodulator 34 A / D converter 35 Branch circuit 36 Oscillator 41 Antenna 42 Signal analysis unit 43 Sampling control Unit 51 received signal monitoring unit 52,252,253,254,255,256,257 interference canceling unit 53,55,56 monitoring unit 54 array receiving unit 63 feature amount calculating unit 64 signal extracting unit 65 received signal analyzing unit 66 sample group Selection unit 71,171 Memory control unit 72 Memory 73, 83, 93, 173, 183, 193 Feature quantity calculation unit 75, 85, 95, 175 Eigenvector generation unit 76, 86, 96, 176, 186, 196 Inner product calculation unit 77,97 Sample group selector 98,198 pairs Matching setting unit 177, 197 Inner product value selection unit 101, 102, 103 Receiver 151 Transmitter 161, 162 Radio 201 Intrusion detection system 202 Watching system

Claims (11)

A receiver for receiving radio waves transmitted from a transmitter arranged in a predetermined area;
A feature amount calculating unit that calculates a feature amount of the radio wave in each of a plurality of target periods received by the receiving unit;
A monitoring device comprising: a monitoring unit that monitors a human operation in the predetermined area based on the radio wave in the target period in which the feature amount satisfies a predetermined condition.   The monitoring device according to claim 1, wherein the feature amount calculation unit calculates a value representing a relationship between the radio wave and a reference radio wave in the target period as the feature amount.   The monitoring apparatus according to claim 2, wherein the feature amount calculation unit calculates, as the feature amount, an inner product value of an eigenvector of the radio wave and an eigenvector of a reference radio wave in the target period or a value based on the inner product value.   The monitoring device according to claim 2, wherein the reference radio wave is a radio wave received by the receiving unit in a period later in time than the plurality of target periods.   4. The reference radio wave according to claim 2, wherein the reference radio wave is a radio wave received by the receiving unit during a period in which the predetermined area is in a predetermined state before the plurality of target periods. 5. Monitoring device. The feature amount calculation unit calculates a value representing a relationship between the radio waves in two target periods as the feature amount,
The monitoring apparatus according to claim 1, wherein the feature amount calculation unit calculates a value representing the relationship for all combinations of the radio waves in two target periods in three or more target periods.   The monitoring unit monitors a human operation in the predetermined area based on the radio wave in the target period in which the degree of coincidence with the reference radio wave is the largest among the plurality of target periods. The monitoring device according to claim 5.   The monitoring device according to claim 1, wherein the feature amount calculation unit calculates a statistical value of the received power of the radio wave in the target period in the monitoring device as the feature amount.   The monitoring apparatus according to claim 1, wherein the feature amount calculation unit calculates the feature amount from a frequency distribution of the radio wave in the target period received by the reception unit.   The monitoring unit monitors human movements in the predetermined area based on the radio wave in the target period in which the feature amount is closest to a predetermined value among the plurality of target periods. 10. The monitoring device according to any one of 9 above. A monitoring program used in a monitoring device,
On the computer,
Receiving radio waves transmitted from a transmitter disposed in a predetermined area;
Calculating the feature quantity of the radio wave in each of a plurality of received target periods;
A monitoring program for executing a step of monitoring a human operation in the predetermined area based on the radio wave in the target period in which the calculated feature amount satisfies a predetermined condition.

Images