JP2014228291A - Radio detection device and radio detection method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a radio detection device capable of detecting the direction of an object with high accuracy.SOLUTION: The present radio detection device comprises: a complex transfer function vector arithmetic processing unit for obtaining, on the basis of a plurality of radio signals received by a plurality of receiving antennas, a complex transfer function vector between a transmission antenna and the receiving antennas; a recording unit for recording a temporal change of the complex transfer function vector obtained by the complex transfer function vector arithmetic processing unit; a Fourier transform arithmetic processing unit for Fourier-converting the temporal response of each element of the complex transfer function vector recorded by the recording unit; a vector extraction arithmetic processing unit for extracting a frequency band specific to an object other than direct-current components from the complex transfer function vector Fourier-converted by the Fourier transform arithmetic processing unit; and a correlation matrix arithmetic processing unit for estimating the direction of the object on the basis of the Fourier-converted complex transfer function vector of the frequency band specific to the object other than direct-current components extracted by the vector extraction arithmetic processing unit.

Description

  The present invention relates to a wireless detection device and a wireless detection method for detecting an object by receiving a reflected wave of a wireless signal.

  An example of an apparatus that detects an object using a Doppler effect generated between a transmission signal and a reception signal that is a signal reflected by the object is described in Patent Document 1 and Patent Document 2. Yes. For example, the wireless detection device described in Patent Document 1 receives a plurality of antennas that transmit and receive radio waves for human body detection, and the reflected waves of transmitted radio waves, and the frequency of the received signal and the frequency of the transmitted signal A difference detection circuit that obtains a difference from the frequency and generates a difference signal according to the difference. In this configuration, the transmitted signal hits the object and is reflected, and is received as a reflected signal that is changed compared to the transmission signal due to the Doppler effect. In addition, the wireless detection device described in Patent Document 1 detects a difference between a transmission signal and a reception signal with a difference detection circuit for a plurality of sets of signals transmitted and received by a plurality of antennas, and detects a plurality of detected signals. The presence / absence and direction of the object are obtained by obtaining the sum signal and the difference signal of the difference signals.

JP 2010-249712 A Japanese Patent No. 4783130

  In the above-described wireless detection device, the azimuth of the object is detected by detecting the difference between the transmission signal and the reception signal using the Doppler effect. Therefore, when used in a multipath environment, a plurality of unnecessary signals are included in the received signal, resulting in a problem that an azimuth error occurs. Further, since the difference between the transmission signal and the reception signal is detected, there is a problem that the transmission antenna and the reception antenna cannot be arranged at arbitrary positions.

  The present invention has been made in consideration of the above circumstances, and an object of the present invention is to provide a wireless detection device and a wireless detection method that can accurately detect the direction of an object.

  In order to solve the above-described problem, a wireless detection device of the present invention includes a transmission antenna, a plurality of reception antennas that receive reflected waves of a wireless signal transmitted from the transmission antenna, and a plurality of reception antennas received by the plurality of reception antennas. A complex transfer function vector calculation processing unit for calculating a complex transfer function vector between the transmission antenna and the reception antenna based on a radio signal of the first and second, and a time of the complex transfer function vector calculated by the complex transfer function vector calculation processing unit A recording unit that records changes, a Fourier transform processing unit that Fourier-transforms a time response of each element of the complex transfer function vector recorded by the recording unit, and a complex transfer function that is Fourier-transformed by the Fourier transform processing unit A vector extraction calculation processing unit for extracting a frequency band specific to an object other than a DC component from a vector; and the vector A direction estimation calculation processing unit that estimates the direction of the object based on the Fourier-transformed complex transfer function vector of the frequency band specific to the object other than the DC component extracted by the extraction calculation processing unit. And

  The radio detection method of the present invention is based on a transmission antenna, a plurality of reception antennas that receive reflected waves of radio signals transmitted from the transmission antenna, and a plurality of radio signals received by the plurality of reception antennas. A complex transfer function vector arithmetic processing unit for obtaining a complex transfer function vector between the transmitting antenna and the receiving antenna, and a record for recording time changes of the complex transfer function vector obtained by the complex transfer function vector arithmetic processing unit From the complex transfer function vector Fourier-transformed by the Fourier transform operation processing unit, Fourier transform operation unit for Fourier transforming the time response of each element of the complex transfer function vector recorded by the recording unit The direction estimation calculation processing unit uses a vector extraction calculation processing unit that extracts a frequency band specific to an object other than And estimating the direction of the object based on the complex transfer function vector wherein is the Fourier transform of the object-specific frequency band other than the DC component the vector extracting processing unit has extracted.

  According to the present invention, a complex transfer function vector (complex channel matrix) obtained from signals received by a plurality of receiving antennas is recorded for a certain period of time, and a frequency band unique to an object other than a direct current of a complex channel matrix subjected to Fourier transform Is extracted, and the direction of the object is estimated based on the extracted vector. Here, since the complex transfer function vector can easily correspond to the multipath environment, according to the present invention, the detection accuracy can be easily increased even in the multipath environment.

It is the block diagram which showed the structural example of the radio | wireless detection apparatus as 1st Embodiment of this invention. It is a figure which shows an example (when there is no person) of the actual value using the radio | wireless detection apparatus 1 of FIG. It is a figure which shows an example of the actual value using the radio | wireless detection apparatus 1 of FIG. It is a figure which shows an example (when one person is arrange | positioned in a -10 degree direction in general) using the wireless detection apparatus 1 of FIG. It is a figure which shows an example of the actual value using the radio | wireless detection apparatus 1 of FIG. It is a figure which shows the frequency average of the measured value shown in FIG. It is a figure which shows the frequency average of the actual value shown in FIG. It is a figure which shows the frequency average of the actual value shown in FIG. It is the block diagram which showed the structural example of the radio | wireless detection apparatus as 2nd Embodiment of this invention. It is a figure which shows an example (when there is no person) of the actual measurement value using the radio | wireless detection apparatus 1a of FIG. FIG. 10 is a diagram illustrating an example of actual measurement values using the wireless detection device 1a of FIG. FIG. 10 is a diagram illustrating an example of actual measurement values using the wireless detection device 1a of FIG. 9 (in the case where one person is arranged approximately in the −10 ° direction). FIG. 10 is a diagram illustrating an example of actually measured values using the wireless detection device 1a of FIG. 9 (in the case where one person is disposed in each of approximately −20 ° direction and approximately 20 ° direction). It is a figure which shows the frequency average of the actual value shown in FIG. It is a figure which shows the frequency average of the actual value shown in FIG. It is a figure which shows the frequency average of the actual value shown in FIG.

  Embodiments of the present invention will be described below with reference to the drawings.

[First Embodiment]
FIG. 1 shows a configuration example of a wireless detection apparatus as a first embodiment of the present invention. The wireless detection device 1 illustrated in FIG. 1 includes a transmitter 10, a transmission antenna 11, a plurality of receivers 12, a plurality of reception antennas 13, and an arithmetic processing device 15. The arithmetic processing unit 15 includes a complex transfer function vector arithmetic processing unit 16, a recording unit 17, a Fourier transform arithmetic processing unit 18, a vector extraction arithmetic processing unit 19, and a steering vector multiplication arithmetic processing unit (of the direction estimation arithmetic processing unit). Example) 20. The transmitter 10 transmits a predetermined radio signal from the transmission antenna 11. The plurality of receiving antennas 13 are arranged in a straight line, for example, and constitute an array antenna 14. It is desirable that the interval between the plurality of receiving antennas 13 be, for example, about a half wavelength of the radio signal or more. The plurality of reception antennas 13 receive the reflected wave of the radio signal transmitted from the transmission antenna 11.

  The complex transfer function vector calculation processing unit 16 obtains a complex transfer function vector (that is, a complex channel matrix) between the transmission antenna 11 and the reception antenna 13 based on a plurality of radio signals received by the plurality of reception antennas 13. The recording unit 17 records the time change of the complex transfer function vector obtained by the complex transfer function vector calculation processing unit 16. The Fourier transform calculation processing unit 18 performs Fourier transform on the time response of each element of the complex transfer function vector recorded by the recording unit 17.

  The vector extraction calculation processing unit 19 extracts a frequency band specific to the object other than the DC component from the complex transfer function vector Fourier-transformed by the Fourier transform calculation processing unit 18. Here, the object is an object to be detected with respect to the presence / absence of the wireless detection device 1 and the direction in which the wireless detection device 1 exists, for example, a human body. The natural frequency of the object is, for example, a frequency specific to the human body, and corresponds to biological information of the human body such as heartbeat and respiration. Since the heart rate (or pulse rate) and respiration rate vary depending on the age and movement of the human body, when the natural frequency of the object is the human body specific frequency, the natural frequency of the object has a frequency band with a certain spread. It will be a thing. As confirmed by the experiment, a favorable detection result for the human body is obtained in the experiment described later by setting the frequency band of 0.15 Hz to 1.6 Hz to a frequency unique to the human body. The steering vector multiplication arithmetic processing unit 20 multiplies the vector extracted by the vector extraction arithmetic processing unit 19 by the steering vector representing the amplitude information and the phase information corresponding to the directions of the plurality of receiving antennas 13. The direction of the object is estimated based on the evaluation function for the direction viewed from the receiving antenna.

[Operation of First Embodiment]
Next, the operation of the wireless detection device 1 described with reference to FIG. 1 will be described. In the wireless detection device 1 illustrated in FIG. 1, for example, presence / absence of a human body and a direction in the presence of a human body are estimated by using a steering vector based on reception signals of a plurality of antennas 13 arranged on the receiver 12 side. That is, in the wireless detection device 1, the steering vector multiplication operation processing unit 20 detects the direction of an object using an arrival direction estimation method generally called a beamformer method (reference documents on arrival direction estimation method: Noriyoshi Kikuma, “Basics of Array Antenna”, Thursday, November 26, 2009, APMC National Committee, IEICE, 2009 Microwave Works & Exhibition (MWE2009), Basic Course 03, <URL: http: // apmc-mwe. org / MicrowaveExhibition2010 / program / tutorial2009 / TL03-01.pdf>).

  In the wireless detection device 1 shown in FIG. 1, the signal generated from the transmitter 10 is transmitted from the transmission antenna 11, propagates through space, and is received by the reception antenna 13. The signal received by the receiver 12 is arithmetically processed by the complex transfer function vector arithmetic processing unit 16. When the number of transmission antennas is 1 and the number of reception antennas is m, the complex transfer function vector (complex channel matrix) calculated by the complex transfer function vector calculation processing unit 16 is expressed by the following equation.

In Expression (1), H (t) represents a complex channel matrix having h ₁₁ (t),..., H _1m (t) as elements, and t represents a time at which the complex channel matrix was acquired. In addition, h ₁₁ (t),..., H _1m (t) represents a transfer function for each reception antenna 13, and when a radio signal transmitted from the transmission antenna 11 is received by each reception antenna 13, It represents the rate at which the amplitude and phase change. Since this complex channel matrix can be calculated based on an actual received signal, it can represent a transfer function of a transmission path such as a multipath environment in which reflected waves and diffracted waves are generated in addition to direct waves. The complex channel matrix calculated by the complex transfer function vector calculation processing unit 16 is recorded for a certain period of time by the recording unit 17, and each element vector of the recorded complex channel matrix is recorded by the Fourier transform calculation processing unit 18. Fourier transformed. The Fourier transformed complex channel matrix is defined by the following equation. In Equation (2), F (f) is a Fourier-transformed complex channel matrix, and f is a frequency. F ₁₁ (f),..., F _m1 (f) represents the frequency response for each receiving antenna 13.

  The vector extraction calculation processing unit 19 extracts a vector in a frequency band unique to the human body other than direct current from the Fourier-transformed complex channel matrix. The extracted vector is multiplied by the steering vector multiplication operation processing unit 20 by the steering vector representing the amplitude information and the phase information corresponding to the direction of the antenna 13 mounted on the receiver 12 and mounted on the receiver 12. An evaluation function for the direction viewed from the receiving antenna 13 is calculated. The steering vector is expressed by the following equation.

  In Expression (3), d represents the element spacing of the receiving antenna 13, θ represents the direction viewed from the receiving antenna 13 (ie, the arrival direction), and λ represents the wavelength at the measurement frequency.

  The evaluation function for the direction viewed from the receiving antenna 13 is expressed by the following equation. In the following equation, the superscript H represents a complex conjugate transpose. The steering vector multiplication operation processing unit 20 estimates the presence / absence, number and direction of a human body using this evaluation function.

  The intensity distribution of the directional characteristic obtained from the evaluation function of Expression (4) is shown in graphs as shown in FIGS. The horizontal axis represents frequency and the vertical axis represents the angular direction. Further, the magnitude of the intensity is represented by the gray scale density (the darker is stronger) as shown in the legend as the amplitude (dB). 2 shows a case where there is no person, FIG. 3 shows a case where a person is arranged in a direction of about 25 °, FIG. 4 shows a case where a person is arranged in a direction of about −10 °, and FIG. This is the case where people are placed in the direction. When no one is shown in FIG. 2, the intensity distribution is uniform in any direction. When people are arranged in the 25 ° direction shown in FIG. 3, a large intensity is distributed in the vicinity of the 25 ° direction. When people are arranged in the −10 ° direction shown in FIG. 4, a large intensity is distributed in the vicinity of the −10 ° direction. And when a person is arranged in approximately −20 ° direction and approximately + 20 ° direction shown in FIG. 5, large intensity is distributed in the vicinity of −20 ° direction and + 20 ° direction.

  6 to 8 show the directional characteristics when the intensity distribution of the directional characteristics shown in FIGS. 3 to 5 is averaged in the frequency range of 0.15 Hz to 1.6 Hz. The horizontal axis represents the angular direction, and the vertical axis represents the intensity. FIG. 6 shows a case where the intensity distribution of the direction characteristic in FIG. 3 is obtained by taking the frequency average of the direction characteristic in the frequency range of 0.15 Hz to 1.6 Hz. It can be confirmed that a large intensity is distributed in the direction of approximately 25 ° and the human body direction can be estimated. FIG. 7 shows a case where the intensity distribution of the direction characteristic in FIG. 4 is obtained by taking the frequency average of the direction characteristic in the frequency range of 0.15 Hz to 1.6 Hz. It can be confirmed that a large intensity is distributed approximately in the −10 ° direction, and the human body direction can be estimated. FIG. 8 shows a case where the intensity distribution of the direction characteristic in FIG. 5 is obtained by taking the frequency average of the direction characteristic in the frequency range of 0.15 Hz to 1.6 Hz. Large intensities are distributed in the approximately −20 ° direction and the approximately + 20 ° direction, and it can be confirmed that the human body direction can be estimated.

  According to the present embodiment, a complex transfer function vector (complex channel matrix) obtained from signals received by a plurality of receiving antennas is recorded for a certain time, and a vector at an arbitrary frequency other than direct current of the complex channel matrix subjected to Fourier transform. Since the direction of the human body is estimated by an evaluation function obtained by multiplying the extracted vector and the steering vector, the detection accuracy can be increased even in a multipath environment.

[Second Embodiment]
FIG. 9 shows a configuration example of a wireless detection device as a second embodiment of the present invention. The wireless detection device 1a illustrated in FIG. 4 includes a transmitter 10, a transmission antenna 11, a plurality of receivers 12, a plurality of reception antennas 13, and an arithmetic processing device 15a. The arithmetic processing unit 15a includes a complex transfer function vector arithmetic processing unit 16, a recording unit 17, a Fourier transform arithmetic processing unit 18, a vector extraction arithmetic processing unit 19, and a correlation matrix arithmetic processing unit (an example of a direction estimation arithmetic processing unit). ) 20a. In FIG. 9, the same components as those shown in FIG. The wireless detection device 1a illustrated in FIG. 9 includes a correlation matrix operation processing unit 20a instead of the steering vector multiplication operation processing unit 20 included in the wireless detection device 1 illustrated in FIG. The correlation matrix calculation processing unit 20a multiplies the vector extracted by the vector extraction calculation processing unit 19 and the complex conjugate transposed vector of the extracted vector to obtain a correlation matrix, and determines the direction of the object based on the correlation matrix. presume. More specifically, in this embodiment, the correlation matrix calculation processing unit 20a multiplies the eigenvector of the correlation matrix by the steering vector representing the amplitude information and the phase information corresponding to the directions of the plurality of reception antennas described above. The direction of the object is estimated based on the evaluation function for the direction seen from the plurality of receiving antennas.

[Operation of Second Embodiment]
Next, the operation of the wireless detection device 1a described with reference to FIG. 9 will be described. In the wireless detection device 1a shown in FIG. 9, for example, the presence / absence of a human body and the direction in the presence of the human body are estimated based on a correlation matrix obtained from reception signals of a plurality of antennas 13 arranged on the receiver 12 side. That is, in the wireless detection device 1a, the correlation matrix calculation processing unit 20a detects the direction of an object using an arrival direction estimation method generally referred to as MUSIC (Multiple Signal Classification) (reference document on the arrival direction estimation method). .

  For example, for the wireless detection device 1 a shown in FIG. 9, a signal generated from the transmitter 10 is transmitted from the transmission antenna 11, propagates through space, and is received by the reception antenna 13. The signal received by the receiver 12 is arithmetically processed by the complex transfer function vector arithmetic processing unit 16. When the number of transmission antennas is 1 and the number of reception antennas is m, the complex transfer function vector (complex channel matrix) that is processed by the complex transfer function vector calculation processing unit 16 is expressed by Expression (1).

  The complex channel matrix calculated by the complex transfer function vector calculation processing unit 16 is recorded for a certain period of time by the recording unit 17, and each element vector of the recorded complex channel matrix is recorded by the Fourier transform calculation processing unit 18. Fourier transformed. The Fourier transformed complex channel matrix is defined by Equation (2).

  From the Fourier-transformed complex channel matrix, a vector extraction calculation processing unit 19 extracts a vector in a frequency band unique to the human body other than DC. The correlation matrix calculation processing unit 20a multiplies the vector extracted by the vector extraction calculation processing unit 19 and the complex conjugate transposed vector of the extracted vector to obtain a correlation matrix represented by the following equation.

  The correlation matrix calculation processing unit 20a estimates the human body direction using the correlation matrix shown in Expression (5). As described above, in this embodiment, the correlation matrix calculation processing unit 20a uses the MUSIC method as an estimation method by multiple signal separation based on eigenvalue expansion of the correlation matrix. When the correlation matrix shown in Expression (5) is decomposed into eigenvalues, the following expression is obtained.

In equation (6), V (f) = [v ₁ (f),. . . , V _m (f)] is the eigenvector of the correlation matrix, S (f) = diag [λ ₁ ,. . . , Λ _m ] represents the eigenvalue of the correlation matrix (where diag represents a diagonal matrix). From the eigenvector of the correlation matrix and the steering vector of Equation (3), the evaluation function for the direction seen from the receiving antenna can be defined by the following equation. The correlation matrix calculation processing unit 20a estimates the presence / absence of the human body, the number of people, and the direction using this evaluation function.

  In Expression (7), the number L of direction estimation targets is 1 ≦ L ≦ m−1, and the number of receiving antennas is equal to or greater than the number of human bodies + 1. The intensity distribution of the directional characteristic obtained from the evaluation function of Expression (7) is shown in graphs as shown in FIGS. 10 to 13 show intensity distributions of direction characteristics obtained from the correlation matrix. 10 to 13, as in FIGS. 2 to 5, the horizontal axis represents the frequency, and the vertical axis represents the angular direction. 10 shows a case where there is no person, FIG. 11 shows a case where a person is arranged in a direction of about 25 °, FIG. 12 shows a case where a person is arranged in a direction of about −10 °, and FIG. This is when a person is placed in the direction. When there is no person shown in FIG. 10, the intensity distribution is uniform in any direction. When people are arranged in the 25 ° direction shown in FIG. 11, a large intensity is distributed in the vicinity of the 25 ° direction. When a person is arranged in the direction of approximately −10 ° shown in FIG. 12, a large intensity is distributed in the vicinity of the −10 ° direction (there is a region having a direction characteristic having a higher intensity than the other portions). When a person is arranged in approximately the −20 ° direction and approximately + 20 ° direction shown in FIG. 13, a large intensity is distributed in the vicinity of the −20 ° direction and the + 20 ° direction.

  14 to 16 show the directional characteristics when the intensity distribution of the directional characteristics shown in FIGS. 11 to 13 is averaged in the frequency range of 0.15 Hz to 1.6 Hz. The horizontal axis represents the angular direction, and the vertical axis represents the intensity. FIG. 14 shows a case where the intensity distribution of the direction characteristic in FIG. 11 is obtained by taking the frequency average of the direction characteristic in the frequency range of 0.15 Hz to 1.6 Hz. It can be confirmed that a large intensity is distributed in the direction of approximately 25 ° and the human body direction can be estimated. FIG. 15 shows a case where the intensity distribution of the direction characteristic in FIG. 12 is obtained by taking the frequency average of the direction characteristic in the frequency range of 0.15 Hz to 1.6 Hz. It can be confirmed that a large intensity is distributed approximately in the −10 ° direction, and the human body direction can be estimated. FIG. 16 shows a case where the intensity distribution of the direction characteristic in FIG. 13 is obtained by taking the frequency average of the direction characteristic in the frequency range of 0.15 Hz to 1.6 Hz. Large intensities are distributed in the approximately −20 ° direction and the approximately + 20 ° direction, and it can be confirmed that the human body direction can be estimated.

  As described above, according to the present embodiment, a complex transfer function vector (complex channel matrix) obtained from signals received by a plurality of receiving antennas is recorded for a certain period of time, and the Fourier transform of the complex channel matrix other than direct current is arbitrary. The direction of the human body is estimated from the correlation matrix obtained by extracting the vector at the frequency of, and multiplying the extracted vector by the complex conjugate transposed vector of the extracted vector, so that the detection accuracy can be increased even in a multipath environment. Can do. Moreover, as shown in FIGS. 14-16, 2nd Embodiment may be able to raise detection accuracy rather than 1st Embodiment (FIGS. 6-8).

  In the second embodiment, ESPRIT (Estimation of Signal Parameters via Rotational Innovation Techniques) method, Root-MUSIC method, etc., may be used for multiple signal separation based on eigenvalue expansion of the correlation matrix. In addition, as a direction estimation method based on a linear prediction method using a correlation matrix, a linear prediction method, a Capon method, or the like may be used (see the above-mentioned reference).

  Moreover, it is also possible to obtain an evaluation function by using a value averaged within the frequency band specific to the human body for the correlation matrix shown in Expression (6). In other words, each element of the correlation matrix can be a value obtained by averaging within the frequency band specific to the object.

  Further, with respect to the equations (4) and (7), it is possible to obtain an evaluation function for a plurality of frequencies specific to the human body, and average all the evaluation functions to estimate the direction of the human body. That is, the direction of the object can also be estimated based on the evaluation function of Expression (4) and the evaluation function of Expression (7).

  The embodiment of the present invention is not limited to the above. For example, the number of transmission antennas may be plural instead of one. The detection target is not limited to a human body, and an object having a certain natural frequency or an animal other than a human may be a detection target. In that case, the frequency band specific to the object is changed in accordance with the natural frequency of the object. Further, the frequency band specific to the object may be divided into a plurality of regions. In the wireless detection device of the present invention, a part or all of the arithmetic processing device 15 or 15a can be configured using a computer. In that case, the program executed on the computer can be distributed via a computer-readable recording medium or a communication line.

DESCRIPTION OF SYMBOLS 1, 1a Radio | wireless detection apparatus 10 Transmitter 11 Transmitting antenna 12 Receiver 13 Receiving antenna 14 Array antenna 15, 15a Arithmetic processor 16 Complex transfer function vector arithmetic processing part 17 Recording part 18 Fourier transform arithmetic processing part 19 Vector extraction arithmetic processing part 20 Steering vector multiplication operation processing unit 20a Correlation matrix operation processing unit

Claims (8)

A transmitting antenna;
A plurality of receiving antennas for receiving reflected waves of radio signals transmitted from the transmitting antenna;
A complex transfer function vector arithmetic processing unit for obtaining a complex transfer function vector between the transmitting antenna and the receiving antenna based on a plurality of radio signals received by the plurality of receiving antennas;
A recording unit for recording a time change of the complex transfer function vector obtained by the complex transfer function vector arithmetic processing unit;
A Fourier transform arithmetic processing unit that Fourier transforms a time response of each element of the complex transfer function vector recorded by the recording unit;
A vector extraction operation processing unit for extracting a frequency band specific to an object other than a direct current component from a complex transfer function vector Fourier-transformed by the Fourier transform operation processing unit;
A direction estimation calculation processing unit that estimates the direction of the object based on the Fourier-transformed complex transfer function vector in the frequency band unique to the object other than the DC component extracted by the vector extraction calculation processing unit. A wireless detection device. The object is a human body;
The wireless detection device according to claim 1, wherein the frequency specific to the object corresponds to biological information of a human body such as heartbeat and respiration. The direction estimation calculation processing unit obtains a correlation matrix by multiplying the vector extracted by the vector extraction calculation processing unit and a complex conjugate transposed vector of the extracted vector, and the direction of the object based on the correlation matrix The wireless detection device according to claim 1, wherein the wireless detection device is estimated. The direction estimation calculation processing unit is viewed from the plurality of receiving antennas obtained by multiplying the eigenvector of the correlation matrix by a steering vector representing amplitude information and phase information corresponding to directions of the plurality of receiving antennas. The wireless detection device according to claim 3, wherein the direction of the object is estimated based on an evaluation function with respect to the determined direction. The wireless detection device according to claim 3 or 4, wherein each element of the correlation matrix is an average value within a frequency band specific to the object. The direction estimation calculation processing unit is obtained by multiplying the vector extracted by the vector extraction calculation processing unit by a steering vector representing amplitude information and phase information corresponding to directions of the plurality of receiving antennas. The wireless detection device according to claim 1, wherein the direction of the object is estimated based on an evaluation function with respect to a direction viewed from a plurality of receiving antennas. A wireless detection device, wherein the direction of the object is estimated based on the evaluation function according to claim 4 and the evaluation function according to claim 6. A transmitting antenna;
A plurality of receiving antennas for receiving reflected waves of radio signals transmitted from the transmitting antenna;
A complex transfer function vector arithmetic processing unit for obtaining a complex transfer function vector between the transmitting antenna and the receiving antenna based on a plurality of radio signals received by the plurality of receiving antennas;
A recording unit for recording a time change of the complex transfer function vector obtained by the complex transfer function vector arithmetic processing unit;
A Fourier transform arithmetic processing unit that Fourier transforms a time response of each element of the complex transfer function vector recorded by the recording unit;
From the complex transfer function vector Fourier-transformed by the Fourier transform arithmetic processing unit, using a vector extraction arithmetic processing unit that extracts a frequency band specific to the object other than the DC component,
A direction estimation calculation processing unit estimates the direction of the object based on the Fourier-transformed complex transfer function vector in the frequency band unique to the object other than the DC component extracted by the vector extraction calculation processing unit. A wireless detection method.