JP2017203751A - Radio wave sensor and facility apparatus equipped with radio wave sensor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To suppress erroneous information caused by a multipath and the like.SOLUTION: A radio wave sensor 10 comprises a detection unit 11 and a processing unit 12. The detection unit 11 radiates a radio wave whose frequency changes with the passage of time within a prescribed measurement period T1 to a target space and receives a radio wave from the target space. The processing unit 12 measures the distance to an object Ob present in the target space on the basis of a sensor signal that includes information pertaining to a frequency difference between a signal equivalent to the radiated radio wave and a signal equivalent to the received radio wave. The processing unit 12 has a function to measure the distance to the object Ob present in the target space after information included in a background signal is removed from a sensor signal obtained in each of a plurality of measurement periods T1. When the object Ob present in the target space is assumed to be stationary continuously, the background signal is decided so as to heighten a ratio by which information pertaining to the object Ob is removed from the sensor signal along with the passage of time at which the object Ob is assumed to be stationary.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a radio wave sensor that detects an object using radio waves, and a facility device including a radio wave sensor that operates according to spatial information detected by the radio wave sensor.

  Conventionally, radar devices that detect objects using radio waves have been proposed (see, for example, Patent Document 1 and Patent Document 2). Patent Document 1 describes a millimeter wave radar installed on a column or the like provided on a road. The millimeter wave radar scans a vehicle traveling on a road with a fan-shaped beam, and detects the position of the vehicle. Obtaining speed is described. Further, it is described that the millimeter wave radar described in Patent Document 1 deletes a background such as a guardrail based on a background difference and leaves only vehicle and obstacle data.

  Patent Document 2 describes a radar device that specifies two-dimensionally the positions of a vehicle, an obstacle on a road, and a fixed structure located on a road shoulder. The radar apparatus described in Patent Literature 2 includes a first detection unit that detects a stopped vehicle, a fallen object from the vehicle, a fallen rock, a large garbage, and the like as a temporary fixed object, and a second detection that detects a moving object. Department. Furthermore, in the technique described in Patent Document 2, information about an object that has been stationary for a long period of time (first background information) and information about an object that has been stationary most recently (second background information) are: It is prepared independently. The second detection unit includes a background difference processing unit that extracts detection information about an object whose position information does not match the second background information.

JP 2002-99986 A JP 2013-257288 A

  By the way, a background such as a tree may fluctuate, and a multipath may occur in a radio wave radiated from a radar apparatus or a radio wave reflected by a vehicle or an obstacle as an object. For this reason, the radar apparatus described in Patent Document 1 cannot always reliably delete the background. In addition, when the object has a background and the object and the background overlap each other when viewed from the radar apparatus, the information regarding the background changes. Therefore, depending on the positional relationship between the object and the background, the background information cannot be removed even if background difference is performed, and incorrect information about the object may be extracted.

  Patent Document 2 describes that it is possible to detect a temporary fixed object and a moving object independently, but it is considered to remove erroneous information due to multipath or the like. Absent.

  An object of this invention is to provide the radio wave sensor which enabled suppression of the erroneous information by a multipath etc., and also aims at providing the equipment provided with this radio wave sensor.

  The radio wave sensor according to the present invention includes a detection unit and a processing unit. The detection unit radiates a radio wave whose frequency changes over time within a predetermined measurement period to the target space, and receives the radio wave from the target space. The processing unit is configured to detect an object existing in the target space based on a sensor signal including information on a frequency difference between a signal corresponding to the radiated radio wave and a signal corresponding to the received radio wave in each of the plurality of measurement periods. Measure distance. In addition, the processing unit has a function of measuring a distance to an object existing in the target space after removing information included in a background signal from sensor signals obtained in each of the plurality of measurement periods. When the object existing in the target space is considered to be still still, the background signal indicates the information of the object as the sensor signal as time elapses when the object is considered to be stationary. It is stipulated to increase the rate of removal from

  Moreover, the equipment according to the present invention includes the above-described radio wave sensor and a load device whose operation is instructed according to the state of the object monitored by the radio wave sensor.

  The present invention has an advantage that erroneous information due to multipath or the like can be suppressed by adopting the above configuration.

FIG. 1 is a block circuit diagram showing a radio wave sensor according to an embodiment. FIG. 2 is an explanatory diagram of a transmission wave in the embodiment. FIG. 3 is an explanatory diagram of the operation of the embodiment. FIG. 4 is a graph showing an example of a frequency spectrum in the embodiment. FIG. 5A is a diagram illustrating a measurement period in the embodiment, and FIG. 5B is a diagram illustrating another example of the measurement period in the embodiment. FIG. 6 is a block circuit diagram illustrating a configuration example of the first processing unit and the second processing unit in the embodiment. FIG. 7A is a graph showing an example of a frequency spectrum in the embodiment, and FIG. 7B is a graph after removing a frequency spectrum corresponding to a background object in the embodiment by subtraction. FIG. 8A is a graph showing the movement of an object over time in the embodiment, FIG. 8B is a graph showing a change in distance over time in the embodiment, and FIG. 8C is a graph showing a speed change over time in the embodiment. It is. FIG. 9 is a graph showing an example of a change over time in the distance to an object in the embodiment. FIG. 10 is a block diagram illustrating an example of facility equipment including a radio wave sensor in the embodiment.

  The configuration example described below relates to a radio wave sensor that extracts a spatial information by radiating a radio wave to a target space and receiving the radio wave from the target space, and a facility device including the radio wave sensor. The spatial information here is information on the distance to the object existing in the target space, the speed at which the object existing in the target space moves, information on whether or not the object exists in the monitoring area defined in the target space, the monitoring area Is selected based on information on whether or not the object existing in is an object to be monitored.

  In the radio wave sensor described below, the speed at which an object existing in the target space moves, information on whether or not an object exists in the monitoring area, and information on whether or not the object existing in the monitoring area is the target space It is information based on the information of the distance to the object existing in. Therefore, the radio wave sensor basically has a function of measuring a distance to an object existing in the target space. The object may be a human body, and in this case, the radio wave sensor can be used as a human sensor. The radio wave sensor may be configured to output a control signal for the load device.

  The equipment provided with the radio wave sensor is configured to control the load device according to the spatial information monitored by the radio wave sensor. The load device is selected from a lighting device, an electric door, an air conditioner, an alarm device, and the like. However, the type of load device is not particularly limited.

  For example, when the load device is an electric door, when the radio wave sensor detects a person trying to pass through the electric door, the radio wave sensor outputs a control signal to open the electric door and closes the electric door after the person passes through. Thus, the radio wave sensor can be configured to output a control signal. In other words, an automatic door, which is a facility device, can be configured by an electric door as a load device and a radio wave sensor.

  Moreover, the illuminating device which comprises a street light may be sufficient as a load apparatus. In this case, when the radio wave sensor detects a person approaching the lighting device, the radio wave sensor outputs a control signal so as to increase the light output of the lighting device, and after the person passes, the radio wave sensor reduces the light output. Can be configured to output a control signal. That is, a street light as equipment can be configured by the illumination device as the load device and the radio wave sensor.

  The equipment described above is an example, and various equipment can be configured by appropriately combining the radio wave sensor and the load device.

  As shown in FIG. 1, the radio wave sensor 10 described below includes a detection unit 11 and a processing unit 12. The detection unit 11 radiates radio waves to the target space and receives radio waves from the target space. The processing unit 12 extracts spatial information based on the output of the detection unit 11. In the configuration example shown in FIG. 1, the detection unit 11 determines the timing at which radio waves are radiated to the target space. However, the processing unit 12 may determine the timing at which radio waves are radiated into the target space, and the configuration different from the detection unit 11 and the processing unit 12 may be determined at timings at which radio waves are radiated into the target space.

  For example, as illustrated in FIG. 2, the detection unit 11 operates such that a radiation period Ts that radiates radio waves into the target space and a pause period Tr that does not radiate radio waves into the target space alternately occur. That is, the detection unit 11 radiates radio waves intermittently into the target space. Hereinafter, a period obtained by combining one radiation period Ts and one pause period Tr is referred to as a processing cycle T0. In the operation example shown in FIG. 2, the radiation period Ts is set to ½ with respect to the processing period T0. For example, if the processing period T0 is 2 [ms], the emission period Ts is 1 [ms].

  The processing period T0 and the emission period Ts can be determined as appropriate. For example, it is possible to determine that the processing period T0 is 50 [ms] and the emission period Ts is 1 [ms]. The frequency of the radio wave is selected from a frequency band in the range from microwave to millimeter wave. The radiation period Ts, the processing period T0, and the like are determined by the processing capability of the processing unit 12, the required specifications, and the like. The required specifications are the distance accuracy (resolution), the expected moving speed of the object Ob, the range of the target space, and the like.

  The frequency of the radio wave, the processing period T0, and the emission period Ts are selected according to the type of spatial information extracted by the radio wave sensor 10. That is, the frequency of the radio wave, the processing period T0, and the emission period Ts are appropriately determined in consideration of the application of the radio wave sensor 10, the distance to the object Ob, the speed at which the object Ob moves, and the like. Depending on the type of spatial information extracted by the radio wave sensor 10, the radiation period Ts may be repeated without the pause period Tr.

  Here, the operation of the detection unit 11 stops during the suspension period Tr. That is, the configuration in which the operation of the detection unit 11 stops in the pause period Tr consumes less power than the configuration in which reception is performed in the pause period Tr. In addition, the radio wave sensor 10 can set the pause period Tr longer than the radiation period Ts. Therefore, suppressing power consumption during the suspension period Tr leads to reduction of power consumption.

  The detector 11 not only emits radio waves during the radiation period Ts but also receives radio waves during the radiation period Ts. Hereinafter, the radio wave radiated by the radio wave sensor 10 is referred to as a transmission wave, and the radio wave reflected by the object Ob is referred to as a reflected wave.

  In the radio wave sensor 10 of this configuration example, the detection unit 11 outputs a signal reflecting spatial information in the FMCW method (FMCW: Frequency-modulated continuous-wave). The detection unit 11 includes a transmission circuit and a reception circuit. The detector 11 generates an FMCW signal whose frequency changes with time, as indicated by reference numeral Sg1 in FIG. That is, the FMCW signal can be said to be a signal that has been subjected to frequency modulation so as to change time into frequency. The radio wave sensor 10 described here generates an FMCW signal whose frequency increases linearly with time in the radiation period Ts, as indicated by reference numeral F1 in FIG. The frequency of the FMCW signal may decrease linearly as time elapses, and may include a rising period and a falling period, and it is also essential that the frequency change is linear. Absent.

  The radio wave sensor 10 shown in FIG. 1 replaces the time from radiating a transmission wave until receiving a reflected wave reflected by the object Ob with a change in frequency. That is, the distance from the radio wave sensor 10 to the object Ob is obtained based on the frequency difference between the transmitted wave and the reflected wave at the same time. This principle will be described below.

  Now, the distance from the radio wave sensor 10 to the object Ob is represented by L [m], the time from the transmission wave to the reception of the reflected wave is represented by Δt [s], and the propagation speed of the radio wave is represented by c [m / s]. Then, the distance L [m] is expressed in the form of L = c · Δt / 2. On the other hand, if the rate of change of the frequency of the transmitted wave is expressed as k [Hz / s] and the frequency difference between the transmitted wave and the reflected wave at the same time is expressed as Δf [Hz], Δf = k · Δt, so Δt = Δf A relational expression of / k is obtained. That is, the distance L is expressed as L = (c / 2) Δf (1 / k). Here, if p = c / 2k [m · s], the relationship L = p · Δf is obtained. In this way, the distance L can be obtained simply by multiplying the frequency difference Δf by the coefficient p.

  As shown in FIG. 3, if the difference between the minimum value fmin and the maximum value fmax of the frequency of the transmission wave in the radiation period Ts [s] is Bw [Hz], k = Bw / Ts. Therefore, it can be expressed as p = (c / 2) (Ts / Bw). Here, Bw = fmax−fmin. Hereinafter, the difference Bw between the minimum value fmin and the maximum value fmax of the frequency of the transmission wave is referred to as “sweep frequency width”.

As an example, the radiation period Ts is set to 1 [ms], and the sweep frequency width Bw in the radiation period Ts is set to 150 [MHz]. The light speed c is 3 × 10 ⁸ [m / s]. When these numerical values are applied, the coefficient p is p = (c / 2) (Ts / Bw) = 1 × 10 ⁻³ [m · s]. Therefore, when the frequency difference Δf between the transmitted wave and the reflected wave at the same time is 1 [kHz], the distance L is 1 [m].

  The target space in which the radio wave sensor 10 monitors the object Ob is determined based on the distance to the object Ob. Hereinafter, a boundary closer to the radio wave sensor 10 in the target space is referred to as a “lower limit distance”, and a boundary farther from the radio wave sensor 10 in the target space is referred to as an “upper limit distance”. The lower limit distance and the upper limit distance are determined according to the usage application of the radio wave sensor 10. The upper limit distance and the lower limit distance can be determined by limiting the range of the first data value in the processing unit 12.

  In addition, the numerical value shown here is an example, Comprising: It is not the meaning which limits operation | movement. That is, the radiation period Ts and the sweep frequency width Bw are appropriately determined according to the use of the radio wave sensor 10, the distance to the object Ob, the speed at which the object Ob moves, and the like. Further, the range of the target space is determined by the distance.

  In order to extract the frequency difference Δf between the transmitted wave and the reflected wave at the same time, the detection unit 11 has a transmitter 111, a mixer 112, two amplifiers 113 and 114, and a transmitter for transmission in the configuration example shown in FIG. A configuration including an antenna 115 and a receiving antenna 116 is employed. The amplifier 113 amplifies the high frequency signal output from the transmitter 111 and supplies it to the transmitting antenna 115. The transmitting antenna 115 converts the high-frequency signal received from the amplifier 113 into a transmission wave and radiates it to the target space.

  On the other hand, the receiving antenna 116 converts the radio wave received from the target space into a high frequency signal and delivers it to the amplifier 114. The amplifier 114 amplifies the high frequency signal from the receiving antenna 116 and passes the amplified high frequency signal to the mixer 112. The mixer 112 mixes the high-frequency signal received from the amplifier 114 with the signal output from the transmitter 111. Therefore, the mixed signal output from the mixer 112 includes the sum and difference components of the frequency of the radio wave radiated to the target space by the transmitting antenna 115 and the frequency of the radio wave received from the target space by the receiving antenna 116. That is, when the radio wave received by the receiving antenna 116 includes a reflected wave reflected by the object Ob, the mixed signal output from the mixer 112 includes information on the frequency difference between the transmitted wave and the reflected wave.

  In the configuration example shown in FIG. 1, the transmitting antenna 115 and the receiving antenna 116 are individually provided, but one antenna may be used as both the transmitting antenna 115 and the receiving antenna 116. The amplifiers 113 and 114 may be provided as necessary and may be omitted.

  Furthermore, in the configuration example illustrated in FIG. 1, the detection unit 11 includes one mixer 112, but a configuration in which the detection unit 11 includes two mixers may be employed. In the configuration in which the detection unit 11 includes two mixers, a configuration in which the two mixers receive high-frequency signals output from the amplifier 114 may be employed. In this configuration, the two signals mixed with the high-frequency signal output from the amplifier 114 in the two mixers have the same frequency but are given a phase difference.

  For example, the high frequency signal output from the transmitter 111 is given to one of the two mixers, and the high frequency signal output from the transmitter 111 is passed through the phase shifter to the other of the two mixers to give a phase difference of 90 degrees. Give a high frequency signal. That is, the detection unit 11 outputs two mixed signals including components corresponding to the frequency difference between the transmission wave and the reflected wave and having different phases by performing quadrature detection of the high-frequency signal output from the amplifier 114. Configured. In this configuration, it is desirable to perform frequency analysis in the complex signal domain in order to improve distance accuracy.

  Further, the detection unit 11 may employ a configuration in which the high-frequency signal output from the amplifier 114 is down-converted in two stages. That is, the detection unit 11 may have a double conversion type configuration in which down-conversion is further performed after down-conversion of a high-frequency signal. The down conversion may employ either a configuration performed on the analog signal output from the mixer 112 in the detection unit 11 or a configuration performed on the digital signal in the processing unit 12.

  The mixed signal output from the mixer 112 is converted into a digital signal by an analog-digital converter. The analog-to-digital converter converts the mixed signal into a serial digital signal. Here, either the detection unit 11 or the processing unit 12 may include the analog-digital converter. In recent years, an integrated circuit corresponding to the detection unit 11 is manufactured as a device constituting the FMCW millimeter-wave radar. Therefore, this type of device may be used for the detection unit 11. This type of device has a configuration that includes an analog-digital converter and a configuration that does not.

  The antenna constituting the transmitting antenna 115 and the receiving antenna 116 is selected from a patch antenna, a slot antenna, a horn antenna, and the like. The transmitting antenna 115 and the receiving antenna 116 are arranged close to each other so that the difference between the distance from the transmitting antenna 115 to the object Ob and the distance from the receiving antenna 116 to the object Ob is relatively small. The

  The transmitting antenna 115 and the receiving antenna 116 are designed to correspond to a frequency band of more than 24.05 GHz and 24.25 GHz or less, for example. In such a frequency band, the size and interval between the transmitting antenna 115 and the receiving antenna 116 may be about several mm. Here, the distance between the transmitting antenna 115 and the receiving antenna 116 means the size of a gap formed between the transmitting antenna 115 and the receiving antenna 116.

  In Japan, this frequency band is used to obtain information selected from the presence, position, movement, size, etc. of a target such as a person or an object, and is used without obtaining a radio station license. Is possible. This type of radio station is a radio station that is used for purposes other than navigation of ships or aircraft, and is called “a specific low-power radio station for mobile object detection radio wave sensors” in Japan. The frequency band shown here is an example and can be changed as necessary.

  The processing unit 12 includes a frequency analysis unit 120 that obtains a frequency spectrum for a component corresponding to the frequency difference between the transmitted wave and the reflected wave in the mixed signal output from the mixer 112. The frequency analysis unit 120 is configured to perform frequency analysis on a digital signal, and performs processing selected from DFT (Discrete Fourier Transform), FFT (Fast Fourier Transform), DCT (Discrete Cosine Transform), and the like. That is, the frequency analysis unit 120 performs orthogonal transform that converts a time-domain signal (a time-series signal obtained by converting a mixed signal into a digital signal) into a frequency-domain signal. In addition, it is also possible to employ | adopt the structure which performs wavelet transformation in order to obtain | require a frequency spectrum. When the mixed signal output from the mixer 112 is converted into a digital signal, the mixed signal is passed through an anti-aliasing filter. Therefore, the frequency range for which the frequency analysis unit 120 obtains the frequency spectrum is limited by the anti-aliasing filter. The upper limit value of the frequency that passes through the anti-aliasing filter is determined according to the above-described upper limit distance of the target space.

  When the mixed signal output from the mixer 112 includes a signal having a component corresponding to the frequency difference between the transmitted wave and the reflected wave, the frequency analysis unit 120 calculates the frequency spectrum as shown in FIG. 4 from the time domain signal. Ask for. That is, the frequency analysis unit 120 converts time domain information into frequency domain information. The frequency spectrum here is expressed in a format in which power (energy spectrum density, power spectrum density) or amplitude value is associated with each of a plurality of frequency bins determined by the frequency analysis unit 120. The frequency analysis unit 120 converts a digital signal corresponding to the input signal into a frequency spectrum in the frequency domain for each predetermined measurement period T1 as shown in FIG. 5A. The measurement period T1 is set to an integral multiple of the processing period T0, with one processing period T0 as a minimum period. The measurement period T1 may generally be a period corresponding to one processing cycle T0 (see FIG. 5A). However, as shown in FIG. 5B, the measurement period T1 may be a period corresponding to a plurality of processing cycles T0.

  Since the frequency spectrum shown in FIG. 4 includes three peaks P11, P12, and P13, it is estimated that the object Ob exists at distances corresponding to the frequencies corresponding to the three peaks P11, P12, and P13. However, even if the three peaks P11, P12, and P13 are generated, the three objects Ob are not always present, and a plurality of peaks may be generated for one object Ob.

  By the way, the processing unit 12 distinguishes between an object Ob stationary in the target space (hereinafter referred to as “stationary object”) and an object Ob moving in the target space (hereinafter referred to as “moving object”). In addition, a first processing unit 121 and a second processing unit 122 are provided. The 1st process part 121 removes the component of a stationary object between the sensor signal obtained for every measurement period T1, and the sensor signal obtained in advance at a predetermined time. On the other hand, the second processing unit 122 removes a stationary object component between sensor signals obtained every measurement period T1.

  The sensor signal is a signal including information on the frequency difference between the transmitted wave and the reflected wave. The time domain mixed signal before being input to the frequency analysis unit 120 and the frequency spectrum obtained by the frequency analysis unit 120 (FIG. 4). For example). That is, the first processing unit 121 and the second processing unit 122 use either a time domain sensor signal or a frequency domain sensor signal.

  For both the first processing unit 121 and the second processing unit 122, it is necessary to obtain a difference between sensor signals obtained at different times, and one of the two sensor signals precedes the other. Therefore, it is necessary to store the sensor signal obtained at the previous time. The first processing unit 121 and the second processing unit 122 have the same configuration although stored information is different. As shown in FIG. 6, the first processing unit 121 and the second processing unit 122 are both subtracting units. 12A and a storage unit 12B. The storage unit 12B included in the first processing unit 121 stores sensor signals obtained in advance at a predetermined point in time, and the storage unit 12B included in the second processing unit 122 updates the sensor signals stored for each measurement period T1. Is done.

  In FIG. 6, the first processing unit 121 and the second processing unit 122 are not distinguished, but the subtraction unit 12A and the storage unit 12B are shared by the first processing unit 121 and the second processing unit 122. The first processing unit 121 and the second processing unit 122 may be provided separately. When the first processing unit 121 and the second processing unit 122 share the subtraction unit 12A and the storage unit 12B, the first processing unit 121 and the second processing unit 122 are operated at different times. In this case, each of the first processing unit 121 and the second processing unit 122 temporarily stores a processing result.

  The processing unit 12 can select a configuration in which the first processing unit 121 and the second processing unit 122 are both placed in front of the frequency analysis unit 120 and a configuration in which both are placed after the frequency analysis unit 120. is there. In addition, the processing unit 12 may employ a configuration in which the first processing unit 121 is placed before the frequency analysis unit 120 and the second processing unit 122 is placed after the frequency analysis unit 120.

  When either the first processing unit 121 or the second processing unit 122 is placed in front of the frequency analysis unit 120, options of a configuration for handling an analog signal and a configuration for handling a digital signal are conceivable. When handling an analog signal, the signal stored in the storage unit 12B is converted into a digital signal. When the digital signal stored in the storage unit 12B is supplied to the subtracting unit 12A, it is reconverted into an analog signal, and the subtracting unit 12A outputs a difference between the analog signals. On the other hand, when a digital signal is handled, the sensor signal (mixed signal) is converted into a serial digital signal, and the subtractor 12A is configured to output a difference between the serial digital signals.

  In either case of the first processing unit 121 and the second processing unit 122, when the frequency analysis unit 120 is placed later, a difference between signals in the frequency domain representing the frequency spectrum (see FIG. 4) is obtained. That is, the difference for each frequency bin of the frequency spectrum is obtained. That is, the storage unit 12B stores the power or amplitude value for each frequency bin, and the subtraction unit 12A obtains a difference for each frequency bin. The frequency spectrum may be either a power spectrum in which the vertical axis is power or an amplitude spectrum in which the vertical axis is an amplitude value, but in the following description, the power spectrum is assumed as an example. When obtaining a difference using an amplitude spectrum, the difference value is obtained by calculating the difference value directly by calculating the difference between the amplitude values and obtaining the difference between the real part and the imaginary part between the amplitude spectra. Either the calculation or the calculation may be adopted. Moreover, although the memory | storage part 12B may memorize | store either power and an amplitude value, below, the case where power is memorize | stored as an example is demonstrated. That is, in the following description, power can be read as an amplitude value, and a power spectrum can be read as an amplitude spectrum.

  As a configuration of the radio wave sensor 10, a configuration in which the frequency analysis unit 120 is associated with the first processing unit 121 and the second processing unit 122 on a one-to-one basis can be considered. And a configuration in which one frequency analysis unit 120 is shared. When both the first processing unit 121 and the second processing unit 122 are placed in front of one frequency analysis unit 120, the frequency analysis unit 120 outputs the output of the first processing unit 121 and the output of the second processing unit 122. Receive in different period. That is, the output of the first processing unit 121 and the second processing unit 122 are set so that the frequency analysis unit 120 does not overlap the period for receiving the output of the first processing unit 121 and the period for receiving the output of the second processing unit 122. The timing for passing the output to the frequency analysis unit 120 is determined.

  The same applies when the first processing unit 121 is placed in front of the frequency analysis unit 120 and the second processing unit 122 is placed after the frequency analysis unit 120. That is, the first processing unit 121 and the second processing unit 121 and the second processing unit 120 are configured so that the period in which the frequency analysis unit 120 receives the output of the first processing unit 121 and the period in which the frequency analysis unit 120 delivers the sensor signal to the second processing unit 122 do not overlap. The timing at which the processing unit 122 operates is determined.

  When adopting a configuration in which both the first processing unit 121 and the second processing unit 122 are post-installed in one frequency analysis unit 120, both the first processing unit 121 and the second processing unit 122 are It is possible to simultaneously receive sensor signals output from the frequency analysis unit 120. Even when this configuration is adopted, the period in which the first processing unit 121 receives the sensor signal and the period in which the second processing unit 122 receives the sensor signal can be configured not to overlap. However, it is desirable to employ a configuration in which the first processing unit 121 and the second processing unit 122 receive the sensor signal output from the frequency analysis unit 120 at the same time. When this configuration is adopted, there is no time difference between the output of the first processing unit 121 and the output of the second processing unit 122, and the sensor signal output from the frequency analysis unit 120 is converted into the first processing unit 121 and the second processing unit. There is an advantage that processing for adjusting the timing of input to the unit 122 is not necessary.

  Table 1 shows an example of the arrangement of the first processing unit 121 and the second processing unit 122 described above. In Table 1, “first” is the first processing unit 121, and “second” is the second processing unit 122. Further, in Table 1, “Prefix” is a state in which the first processing unit 121 or the second processing unit 122 is placed in front of the frequency analysis unit 120, and “Postfix” is the first processing unit 121 or the second processing unit 122. Represents a state after the frequency analysis unit 120. “E” indicates that the first processing unit 121 or the second processing unit 122 is provided. In Table 1, when the first processing unit 121 or the second processing unit 122 is placed in front of the frequency analysis unit 120, whether to handle an analog signal or a digital signal is not particularly limited.

  By the way, in order to obtain the spatial information of the target space from the frequency spectrum, for example, as shown in FIG. 4, in the power spectrum that is the frequency spectrum with the power as the vertical axis, the frequency bin whose power is the peak value is obtained. Hereinafter, a frequency bin whose power is a peak value is referred to as a “peak frequency”. In the example shown in FIG. 4, the peak frequencies are indicated by f11, f12, and f13. The processing unit 12 having the above-described configuration obtains the frequency component from which the component of the stationary object Ob is removed for each measurement period T1. Therefore, the peak frequencies f11, f12, and f13 in the frequency components obtained by the processing unit 12 are estimated as a frequency difference between the reflected wave and the transmitted wave from the object Ob moving in the target space.

  Here, if the information to be stored in the storage unit 12B of the first processing unit 121 is acquired in a state where there is no moving object in the target space, only the information on the stationary object is stored in the storage unit 12B of the first processing unit 121. Is memorized. Such a stationary object is estimated to be an object Ob that exists for a relatively long time. Hereinafter, the object Ob whose information is stored in the storage unit 12B of the first processing unit 121 is referred to as a “background object”. A signal corresponding to the background object is referred to as a “background signal”. That is, it can be said that the 1st process part 121 outputs the information which remove | eliminated the information of the background object from the information obtained for every measurement period T1. In other words, the first processing unit 121 removes the background signal information from the sensor signal obtained every measurement period T1 and outputs the result.

  After the stationary object information is stored in the storage unit 12B of the first processing unit 121, the information received by the first processing unit 121 for each measurement period T1 in the state where the object Ob other than the background object does not exist in the target space is Ideally, it matches the information stored in the storage unit 12B. In this case, no peak occurs in the power spectrum corresponding to the output of the first processing unit 121. However, when the background object is a tree, a curtain, or the like, and fluctuation occurs in a part of the background object, a peak may occur in the power spectrum corresponding to the output of the first processing unit 121. Similar events may be caused not only by the shape of the background object but also by multipaths of transmitted waves or reflected waves.

  In a state where the object Ob exists in the target space, a difference occurs between the information received by the first processing unit 121 and the information stored in the storage unit 12B for each measurement period T1 due to the reflected wave from the object Ob. Therefore, a peak corresponding to the object Ob that has come to exist in the target space is generated in the power spectrum corresponding to the output of the first processing unit 121.

  However, the power spectrum corresponding to the output of the first processing unit 121 due to the overlap of at least a part of the background object and the object Ob, the multipath of the transmission wave or the reflected wave, and the like has entered the target space. A peak that does not correspond to Ob may occur. For example, when the object Ob overlaps in front of the background object as viewed from the radio wave sensor 10, the reflected wave from the background object decreases, and therefore the frequency bin corresponding to the background object in the power spectrum corresponding to the output of the first processing unit 121. The power of becomes negative. Further, when a multipath occurs in a transmission wave to the object Ob or a reflected wave from the object Ob, a peak due to the multipath may occur in the power spectrum. As described above, the output of the first processing unit 121 may include incorrect information.

  On the other hand, in the second processing unit 122, the information stored in the storage unit 12B is updated every measurement period T1. Here, the subtraction unit 12A of the second processing unit 122 is stationary by obtaining the difference between the information received by the second processing unit 122 and the information stored in the storage unit 12B for each measurement period T1. Assume that the object Ob is configured to remove components. Further, it is assumed that the bin width of the frequency bin is sufficiently small. Under such an assumption, the power spectrum corresponding to the output of the second processing unit 122 changes every measurement period T1 as long as the object Ob moves. That is, in the power spectrum corresponding to the output of the second processing unit 122, a peak corresponding to the distance to the moving object occurs for each measurement period T1.

  When the object Ob moving in the target space is stationary, the information input to the subtraction unit 12A of the second processing unit 122 and the information stored in the storage unit 12B are substantially equal. Therefore, a peak does not occur in the power spectrum corresponding to the output of the second processing unit 122, and the distance to the object Ob is obtained when the object Ob is stationary even though the object Ob exists in the target space. May not be possible.

  As described above, there is a possibility that correct information for measuring the distance to the object Ob may not be obtained depending on the state of the object Ob, regardless of whether the first processing unit 121 or the second processing unit 122 is used. is there. Therefore, the processing unit 12 includes a selection unit 123 that selects information obtained by the first processing unit 121 and information obtained by the second processing unit 122 according to a condition. Here, the selection unit 123 selects not only the output of the first processing unit 121 but also the information on the background object stored in the storage unit 12B of the first processing unit 121 as information obtained by the first processing unit 121. A possible configuration is assumed.

  The selection unit 123 is basically determined to extract information estimated to satisfy the condition of the target object from the information obtained by the first processing unit 121 and the information obtained by the second processing unit 122. It is done. Here, as an example, information included in the mixed signal output from the detection unit 11 includes information on a background object, information on a stationary object other than the background object, information on a moving object, and unnecessary information (unintended Information) is assumed to be classified into four types. Further, it is assumed that the four types of information are information that can be converted into distances. For example, if the information is a peak frequency in the power spectrum, the distance can be obtained by the above-described calculation. Therefore, the information may be a distance value.

  In the following, in order to simplify the description, the case where information is a peak frequency is taken as an example. Further, based on only the peak frequency, the peak frequency corresponding to the output of the first processing unit 121 and the peak frequency corresponding to the output of the second processing unit 122 are not necessary for the background object, the stationary object, the moving object, and the like. Assume that the information is associated with various information.

  When only a background object exists in the target space, ideally, a peak corresponding to the output of the first processing unit 121 does not occur, but a peak corresponding to unnecessary information may occur. When there is an object Ob that is not a background object in the target space, the peak corresponding to the output of the first processing unit 121 includes the peak corresponding to the stationary object, the peak corresponding to the moving object, and unnecessary information. One or more types of peaks may be included. However, it cannot be distinguished whether the peak corresponds to a stationary object, corresponds to a moving object, or is unnecessary information.

  The peak frequency corresponding to the information stored in the storage unit 12B of the first processing unit 121 represents information on the background object. Therefore, the background object information can be distinguished only by the peak frequency corresponding to the output of the first processing unit 121, but the stationary object information, the moving object information, and the unnecessary information cannot be distinguished. In addition, information on stationary objects, information on moving objects, and unnecessary information cannot be distinguished only by peak frequency, but can be distinguished by using power, changes in peak frequency over time, etc. along with peak frequency. It may become.

  The first processing unit 121 has the following configuration in order to reduce the influence of multipath. Here, as the background signal, when the object Ob existing in the target space is considered to be stationary, the information on the object Ob is detected with the passage of time when the object Ob is considered stationary. It is determined to increase the rate of removal from the signal. As a specific example, a weighted sum of the sensor signal for each measurement period T1 and the past background signal obtained in the measurement period T1 a predetermined number of times before is used as the background signal.

  Now, S (k) for the sensor signal, B (k−n) for the past background signal, ω1 for the weight coefficient for multiplying the sensor signal S (k), and the weight coefficient for multiplying the background signal B (k−n) for the past. Assuming ω2, the background signal B (k) is expressed as B (k) = ω1 · S (k) + ω2 · B (k−n). Here, k is an integer value of 0 or more that distinguishes the measurement period T1, and corresponds to time. Thus, an increase in k represents the passage of time. Further, n is a positive integer value of 1 or more, and if n = 1, the background signal B (k−1) represents the background signal used in the previous measurement period T1.

  When the background signal B (k) is defined in the above equation, the background signal B (k) includes the information of the sensor signal S (k) and the information of the past background signal B (k−n) as ω1. : Included at a ratio of ω2. It is desirable that the weighting coefficients ω1 and ω2 have a relationship of ω1> ω2 and ω1 + ω2 = 1. As an example, it is possible to define the relationship as ω1 = 0.05 and ω2 = 0.95.

  The background signal B (k) defined by the above equation has the ratio of the weight coefficients ω1 and ω2 when the object Ob is considered to be stationary after the background signal has changed due to the movement of the object Ob. It converges at the corresponding number of times. That is, until the background signal converges, the influence caused by the movement of the object Ob remains in the background signal B (k). In other words, if the object Ob is considered to be stationary, the information of the object Ob considered to be stationary among the information included in the background signal B (k) with the passage of time. Increases the proportion of information. Therefore, the first processing unit 121 removes information on the object Ob considered to be stationary from the information of the sensor signal as time elapses when the object Ob is considered to remain stationary. The ratio increases.

  On the other hand, since the background signal includes the information of the sensor signal, when there is an object Ob moving in the target space, the information of the moving object Ob is a ratio determined by the weighting coefficients ω1 and ω2. Captured in the background signal. The sensor signal also includes information on the object Ob that is regarded as stationary. If the object Ob regarded as stationary is information on the background object, the weighting coefficients ω1 and ω2 are used. It remains in the background signal regardless. That is, since the information of the moving object Ob is smaller in the background signal than the background object, the information that is decreased in the first processing unit 121 is not removed but is extracted as the output of the first processing unit 121.

  As described above, when the background signal B (k) defined by the above expression is used, the first processing unit 121 removes information on the object Ob that is considered to be stationary as time passes. Increase. It is desirable that the time until the information of the object Ob considered to be stationary is completely removed from the sensor signal is set to about 20 [s] to 30 [s]. The period for updating the background signal may be equal to the measurement period T1, but may be determined to be longer than the measurement period T1. The period for updating the background signal may be determined in a range of, for example, about 0.1 [s] to 1 [s]. The background signal B (k) is not limited to the definition of the above equation, and may be a weighted sum of the sensor signal and the past background signal. Therefore, a relationship such as B (k) = ω1 · S (k−1) + ω2 · B (k−n) may be used.

  When the above-described operation is enabled and the object Ob is considered to be stationary, if the rate of removing the information of the object Ob from the sensor signal with the passage of time is increased, Even if a pass occurs, the possibility of erroneous information remaining continuously is reduced. That is, when multipath occurs, the sensor signal includes incorrect information generated by the multipath, and the incorrect information is reflected in the background signal. If erroneous information continues to occur and the erroneous information hardly fluctuates, the proportion of erroneous information contained in the background signal increases with time, as with the stationary object Ob. . As a result, the first processing unit 121 can remove erroneous information from the sensor signal together with information on the stationary object Ob.

  When the object Ob existing in the target space is still stationary, the ratio of the information of the stationary object Ob to the background signal increases, and eventually is stationary from the output of the first processing unit 121. Information on the object Ob is almost removed. Therefore, the period during which the object Ob existing in the target space is still stationary is evaluated, and when the period reaches a predetermined holding period, the stationary object Ob is treated as a background object. Also good. The holding period may be set to about 20 [s] to 30 [s], for example. That is, the object Ob whose stationary state has reached the holding period is estimated as a non-moving object Ob, and thereafter stored in the storage unit 12B of the first processing unit 121 so as to be handled equivalently to a background object. Since the information stored as the background object in the storage unit 12B of the first processing unit 121 is always removed from the sensor signal, the information of the object Ob newly entering the target space can be easily separated.

  The peak frequency corresponding to the output of the second processing unit 122 corresponds to the moving object. Since the second processing unit 122 obtains a difference in information within a time period in which the environment hardly changes, at the peak frequency corresponding to the output of the second processing unit 122, information on the background object, information on the stationary object, It can be considered that unnecessary information is excluded. Therefore, the moving object information and the other three types of information are distinguished by the peak frequency corresponding to the output of the second processing unit 122.

  If the relationship mentioned above is used, it can be said that the 1st process part 121 provides the information corresponding to a background object independently, and the 2nd process part 122 provides the information of a moving object independently. Further, among the peak frequencies corresponding to the output of the first processing unit 121, the peak frequency not included in the peak frequency corresponding to the output of the second processing unit 122 is either stationary object information or unnecessary information. Is estimated to correspond to

  Here, it is estimated that the stationary object existing in the target space was a moving object before being a stationary object. Therefore, if the peak frequency corresponding to the output of the first processing unit 121 corresponds to a stationary object, the peak frequency is considered to have disappeared after existing as the peak frequency corresponding to the output of the second processing unit 122. Therefore, when the peak frequency corresponding to the output of the second processing unit 122 disappears, when the error from the peak frequency corresponding to the output of the first processing unit 121 is within a minute predetermined range, the corresponding peak frequency is Presumed as stationary object information.

  The selection unit 123 can distinguish the four types of information by using the relationship as described above regarding the information obtained from the first processing unit 121 and the information obtained from the second processing unit 122. Therefore, the selection unit 123 selects necessary information from the four types of information described above under conditions according to the use of the information.

  The selection unit 123 may be configured to perform the following operation. Here, in describing the operation of the selection unit 123, the power spectrum corresponding to the output of the first processing unit 121 is expressed as “output of the first processing unit 121” and corresponds to the output of the second processing unit 122. In terms of power spectrum, it is expressed as “output of second processing unit 122”. In other words, each of the “output of the first processing unit 121” and the “output of the second processing unit 122” does not depend on the arrangement of the first processing unit 121 and the second processing unit 122 with respect to the frequency analysis unit 120. But they are considered equivalent.

  The selection unit 123 always selects the output of the first processing unit 121, and when the output of the first processing unit 121 has a peak and the power corresponding to the peak is equal to or greater than a predetermined threshold, the second processing unit The state shifts to a state in which the output 122 is selected. That is, it is monitored whether or not an object Ob that is not a stationary object enters the target space, and when the object Ob enters, the state shifts to a state in which the output of the second processing unit 122 is selected.

  While the output of the second processing unit 122 is selected, a peak occurs in the output of the second processing unit 122 during the period in which the object Ob is moving. During a period when the output of the second processing unit 122 has a peak, the selection unit 123 selects the output of the second processing unit 122 that is updated every measurement period T1.

  On the other hand, when the peak is not generated in the output of the second processing unit 122 in the state where the second processing unit 122 is selected, the selection unit 123 receives the output of the first processing unit 121 and determines whether the peak is generated. Determine. The occurrence of a peak in the output of the first processing unit 121 indicates that the object Ob exists in the target space. Therefore, if a peak occurs in the output of the first processing unit 121, the selection unit 123 2 The output of the processing unit 122 is received. However, the selection unit 123 maintains and outputs the output of the second processing unit 122 immediately before the peak disappears.

  In this state, the selection unit 123 continues monitoring whether or not a peak occurs in the output of the second processing unit 122 received from the second processing unit 122, and when a peak occurs in the output of the second processing unit 122, The output of the 2nd process part 122 updated for every measurement period T1 is selected. Note that in the period in which the selection unit 123 maintains and outputs the output of the second processing unit 122 immediately before the peak disappears, the selection unit 123 receives the output of the first processing unit 121 at an appropriate timing, and the target space It may be determined whether or not the object Ob exists.

  When the peak disappears from the output of the second processing unit 122 and the peak does not occur in the output of the first processing unit 121, it is considered that the object Ob has left the target space, and the selection unit 123 performs the first processing. It returns to the state which selects the output of the part 121.

  In the operation example described above, the selection unit 123 selects the output of the first processing unit 121 and the output of the second processing unit 122 as follows. That is, it can be determined whether or not the object Ob exists in the target space based on the output of the first processing unit 121, and it is determined whether or not the object Ob moving in the target space exists based on the output of the second processing unit 122. The selection conditions in the selection unit 123 are determined by using what can be done. Since the selection unit 123 selects the information obtained from the output of the first processing unit 121 and the information obtained from the output of the second processing unit 122 so as to complement each other, information that satisfies the condition of the target object Ob is selected. The possibility of extraction increases.

  The operation of the selection unit 123 described above is an example, and other operations can be performed. For example, the selection unit 123 may adopt a configuration in which the selection unit 123 alternately receives the output of the first processing unit 121 and the output of the second processing unit 122.

  By the way, as described above, when a part of the background object shakes or when a multipath occurs in the path of the transmission wave or the reflected wave, a peak occurs in the power spectrum corresponding to the output of the first processing unit 121. There is. The frequency at which such a peak occurs may vary with time. Even if the frequency at which the peak occurs does not change, the mixed signal intensity may vary.

  Assume that the frequency spectrum of the mixed signal obtained at a predetermined point in time has three peaks P11, P12, and P13 as shown by the characteristic F11 in FIG. 7A. Further, it is assumed that the peak frequencies f11, f12, and f13 of the three peaks P11, P12, and P13 correspond to the distance of the background object. In this case, ideally, the three peaks P11, P12, and P13 do not appear in the output of the first processing unit 121. That is, if there is no object Ob in addition to the background object, the residual output from the first processing unit 121 is ideally zero.

  However, as indicated by the characteristic F12, the values of the peaks P21, P22, and P23 fluctuate due to the influence of the object Ob that is not a background object in the target space, and the three peak frequencies f11, f12, and f13 are output to the output of the first processing unit 121. May appear. In other words, in the first processing unit 121, the background information may not be erased simply by subtracting the information stored in the storage unit 12B from the received information. Further, as shown in FIG. 7B, the residual output from the first processing unit 121 tends to increase as the values of the peaks P21, P22, and P23 increase. Although this residual has a relatively small power, the power of the reflected wave from the object Ob that has entered the target space is also relatively small. Therefore, in the power spectrum corresponding to the output of the first processing unit 121, it may be difficult to distinguish the object Ob that has entered the target space from the background object.

  In order to cope with this problem, it is desirable that the processing unit 12 includes a calculation unit 124 at a later stage than the frequency analysis unit 120. The calculation unit 124 is configured to reduce the residual by applying a correction coefficient to the residual of the peak frequencies f11, f12, and f13 corresponding to the background object. Since the correction coefficient is set in association with the peak in the frequency spectrum, it is determined for each frequency bin. Specifically, the calculation unit 124 uses the reciprocal of the power obtained for each frequency bin from the frequency spectrum of the background object as the correction coefficient.

  For example, when the power of the frequency bin f (j) in the frequency spectrum of the background object is represented by p (j), the correction coefficient w (j) in the frequency bin f (j) is w (j) = 1 / p (j ). Here, j is an integer value for distinguishing frequency bins. When the power p (j) is 0, a predetermined fixed value is selected as the correction coefficient w (j). The correction coefficient w (j) determined in this way is a relatively small value for the frequency bin f (j) corresponding to the background object, and for the frequency bin f (j) that is not the background object. A relatively large value.

  The calculation unit 124 applies the correction coefficient w (j) to each frequency bin f (j) with respect to the residual of the power spectrum during the period in which the selection unit 123 selects the output of the first processing unit 121. That is, the power p (j) for each residual frequency bin f (j) is multiplied by the corresponding correction coefficient w (j). When the correction coefficient w (j) is applied to the residual, the power of the frequency bin f (j) corresponding to the background object is relatively small, and the power of the frequency bin (j) that is not the background object is relatively large. .

  When the residual of the power spectrum is multiplied by the correction coefficient by the calculation unit 124, the power p (j) of the frequency bin f (j) corresponding to the background object is substantially eliminated in the power spectrum corresponding to the output of the first processing unit 121. Is done. In the power spectrum corresponding to the output of the first processing unit 121, the power p (j) of the frequency bin f (j) corresponding to the object Ob that has entered the target space is relatively large. In other words, the difference between the power p (j) of the frequency bin f (j) corresponding to the object Ob that has entered the target space and the background object becomes large.

  As described above, if the calculation unit 124 is provided, the power p (j corresponding to the object Ob that has entered the target space using an appropriate threshold value for the power of the power spectrum corresponding to the output of the first processing unit 121. ) And the power p (j) corresponding to the background object can be deleted. That is, the accuracy of detecting the object Ob that has entered the target space is increased by a simple process in which the power is compared with the threshold value for each frequency bin f (j) of the power spectrum corresponding to the output of the first processing unit 121.

  By the way, the processing unit 12 obtains distance data corresponding to the distance from the information selected by the selection unit 123 to the object Ob, and a speed for obtaining speed data corresponding to the moving speed of the object Ob from the distance data. It is desirable to include a processing unit 126. However, the speed processing unit 126 is not essential, and the processing unit 12 may include only the distance processing unit 125 of the distance processing unit 125 and the speed processing unit 126.

  The distance processing unit 125 obtains a peak frequency (corresponding to f11, f12, and f13 in FIG. 4) from the frequency component output from the selection unit 123, and obtains the peak frequency as distance data corresponding to the distance to the object Ob. That is, the distance processing unit 125 obtains a frequency bin value that is a local maximum value among the frequency components output from the selection unit 123 as distance data. When a plurality of peak frequencies are extracted from the frequency component output by the selection unit 123, the distance processing unit 125 sets each of the plurality of peak frequencies as distance data.

  The speed processing unit 126 calculates a difference in distance data at different times from the time series of distance data as speed data. For example, when the distance data is represented by L1 (i) and the speed data is represented by V1 (i), the distance data L1 (i) and the speed data V1 (i) are V1 (i) = L1 (i) −L1. It is represented by the relationship (i + 1). Here, the variable i represents the order in which the distance data and the speed data are arranged in time series, and corresponds to the measurement period T1 on a one-to-one basis. In other words, the variable i corresponds to the time.

  The speed data described above is classified into one of positive value, zero, and negative value. The velocity data being zero indicates that there is no change in the distance to the object Ob. Further, the velocity data defined as described above indicates that the distance to the object Ob is increased if the value is negative, that is, the object Ob is moving away from the radio wave sensor 10, and the object if the value is positive. This represents that Ob is approaching the radio wave sensor 10. The absolute value of the speed data is the distance that the object Ob has moved in the measurement period T1, and therefore corresponds to the speed at which the object Ob has moved. In short, the speed data includes information on the direction and speed in which the object Ob is moving.

  It is not essential that the speed data is a difference between two distance data adjacent in time series, and the speed data can be obtained using three or more distance data. The distance data and the speed data may have a relationship of V1 (i) = L1 (i−1) −L1 (i + 1), for example. Further, the distance data and the speed data may be in a relationship of V1 (i) = {L1 (i-1) -L1 (i + 1)} / 2, and V1 (i) = [{L1 (i) -L1 (i + 1). )} + {L1 (i-1) -L1 (i))}] / 2. Since the speed data is information on the speed of the object Ob, the speed data is not limited to these examples, and can be determined as appropriate. In addition, the frequency of the transmission wave is changed so as to have a period in which the frequency of the transmission wave is increased with the lapse of time and a period in which the frequency of the transmission wave is decreased with the lapse of time. The speed of the object Ob may be obtained from the frequency difference.

  By the way, the processing unit 12 determines the range of the target space based on the value of the distance data. In addition, the processing unit 12 includes a section setting unit 127 that determines one or more specific sections divided by distance as at least a part of the target space. That is, the section setting unit 127 has a case where all or part of the target space is defined as a specific section, and a case where a plurality of specific sections are defined in the target space. One specific section is determined by the range of distance. The object Ob existing in the specific section is treated as a background object. That is, when the distance data obtained by the distance processing unit 125 is included in a specific section determined by the section setting unit 127, a signal corresponding to the object Ob existing in the specific section is handled as background object information.

  The signal corresponding to the object Ob existing in the specific section is updated when the reacquisition of the background signal is instructed. Although the background signal can be automatically updated at a specific date and time, it may be performed by a user's instruction. For example, when a user places a fixture in a specific section, the fixture in the specific section is recognized as a background object by instructing reacquisition of the background signal.

  In order to determine the identity of the object Ob for which distance data is obtained in the target space, it is necessary to set ranges for the distance and speed, and to set the distance and speed ranges relatively wide. In order to determine whether or not the distance data corresponds to a single object Ob, the speed processing unit 126 is configured to perform a tracking process using the distance data. Further, when performing the tracking process, corrected distance data obtained by correcting the distance data output from the distance processing unit 125 is obtained, and the speed data is obtained based on the corrected distance data. When it can be determined that the distance data and speed data obtained for each measurement period T1 correspond to a single object Ob, the speed processing unit 126 gives a label to the distance data and speed data of the corresponding object Ob.

  The processing unit 12 can regard the distance data to which the label is attached as equivalent to the frequency difference between the transmission wave and the reflected wave, and can calculate the distance based on the distance data. If there is speed data with a label, the processing unit 12 can calculate the speed based on the speed data. Since the label is determined in association with the object Ob, the same label is given to the distance data and the velocity data corresponding to the single object Ob. Note that the calculation for obtaining the distance from the distance data or the calculation for obtaining the speed from the speed data is not essential and is appropriately performed according to the use of the radio wave sensor 10.

  By the way, the peak frequency in the frequency spectrum may fluctuate even for a single object Ob that is stationary. That is, when the object Ob is a curtain, a tree, etc., and a part of the object Ob shakes, or when multipath occurs in at least one of the transmission wave and the reflected wave, the peak frequency may fluctuate. There is sex. In other words, even the distance data obtained for a single object Ob may have a relatively large degree of variation. If the variation of the distance data is large, the variation of the speed data is also large.

  When the variation of the distance data is large, it is necessary to widen the range recognized as a single object Ob based on the distance data. However, if the range of the distance data to be determined as a single object Ob is expanded, there is a high possibility that an erroneous label is given to the distance data when there are a plurality of objects Ob in the target space. For example, assuming that two objects Ob having a relatively small speed difference exist in the target space, and the distance data values for the two objects Ob are substantially equal, it is difficult to distinguish the two objects Ob. .

  Therefore, it is desirable that the speed processing unit 126 is configured to suppress variations in distance data and speed data by performing the following processing. Although it is not essential for the speed processing unit 126 to perform the following process, it is desirable to perform the following process in order to reduce the possibility of assigning an erroneous label to the distance data and the speed data.

The speed processing unit 126 tracks the distance data and speed data obtained from the same object so as to approach the true value by using the following relational expression, for example. In other words, the speed processing unit 126 obtains corrected distance data that is close to a true value by suppressing variation in the distance data output from the distance processing unit 125, and obtains speed data using the corrected distance data.
Xs (k) = Xp (k) + α {Xm (k) −Xp (k)}
Vs (k) = Vs (k-1) + [beta] {Xm (k) -Xp (k)} / T1
Xp (k + 1) = Xs (k) + Vs (k) · T1
Here, Xm (k) means the kth distance data output by the distance processing unit 125, and Xs (k) is the kth corrected distance data obtained by calculation. Vs (k) is speed data obtained by calculation, and Xp (k) is a predicted value of the kth distance data. The distance data can be read as distance, and the speed data can be read as speed.

Here, the constant α and the constant β are determined to have a relationship of β = α ² / (2-α), for example. The constant α and the constant β are not limited to this relationship, but relatively good results are obtained by using this relationship. The corrected distance data Xs (k) and velocity data Vs (k) are less susceptible to noise as the constants α and β are smaller, and are less susceptible to steep changes as the constants α and β are larger. Therefore, the magnitudes of the constants α and β are determined according to the purpose.

  In order to use the above-described relational expression, the initial value Xp (1) related to the predicted value Xp (k) of the distance data and the initial value Vs (0) related to the speed data Vs (k) are necessary. Here, the initial value Xp (1) regarding the predicted value Xp (k) of the distance data is substituted with the distance data Xm (1) (that is, Xp (1) = Xm (1)). The initial value Vs (1) related to the speed data Vs (k) employs a difference between the distance data Xm (2) and the distance data Xm (1) (that is, Vs (1) = Xm (1) −Xm (2)).

  If the above-described initial values are applied, the values of the above-described three expressions can be obtained based on the distance data Xm (k). That is, based on the distance data Xm (k) output from the distance processing unit 125, the speed processing unit 126 suppresses variations in the distance data Xm (k) by obtaining the predicted value Xp (k) of the distance data. Corrected distance data Xs (k) is obtained. Furthermore, the speed processing unit 126 obtains speed data Vs (k) with suppressed variation based on the corrected distance data Xs (k). In this way, when variation in distance data and speed data is suppressed, even if there are a plurality of objects Ob in the target space, a plurality of objects Ob can be obtained by rationally associating the distances and speeds of the objects Ob. There is a high possibility that a label can be assigned to each object Ob without error.

  When the correction distance data described above is obtained, the correction distance data may be used instead of the distance data. For example, in the case where fluctuation occurs even though the object Ob is arranged at a fixed position such as a curtain or a tree, the range in which the fluctuation of the distance data is allowed may be determined based on the corrected distance data. Good. The section setting unit 127 may determine a specific section based on the correction distance data.

  Now, as shown in FIG. 8A, it is assumed that the position of the object Ob changes with time. In the example of FIG. 8A, the object Ob approaches the radio wave sensor 10 from a distance of about 11 [m], approaches the radio wave sensor 10 and then moves away from the radio wave sensor 10. In FIG. 8A, the moving speed of the object Ob is substantially constant and is about 1.5 [m / s]. When the above-described calculation was performed with the constant α set to 0.3 and the constant β set to 0.0529, the results shown in FIGS. 8B and 8C were obtained. In FIG. 8B, the dot Xm represents the error of the distance measurement value Xm (k), and the characteristic Xs represents the error of the distance calculation value Xs (k). That is, assuming that the true value of the distance from the radio wave sensor 10 to the object Ob is Xt (k), the dot Xm represents Xm (k) −Xt (k), and the characteristic Xs represents Xs (k) −Xt (k). Represent. In FIG. 8C, a characteristic Vm represents a speed measurement value obtained from a distance measurement value Vm (k), and a characteristic Vs represents a speed calculation value Vs (k). According to FIGS. 8B and 8C, it can be seen that the variation of the distance data and the speed data is suppressed by the speed processing unit 126 performing the above-described processing.

  The above-described relational expression used for suppressing variation in distance data and velocity data is merely an example showing the principle of suppressing variation. Here, a relational expression that multiplies the difference between the measured value and the predicted value by constants α and β is used. Therefore, the degree of variation in distance data and speed data can be adjusted by appropriately determining the values of constants α and β. Is possible.

  A configuration for performing this type of processing is called a tracking filter. In addition to the αβ filter using the above-described relational expression, several types of tracking filters are known. That is, another tracking filter can be used for the speed processing unit 126. The speed processing unit 126 may suppress variations in distance data and speed data using other relational expressions.

  As described above, when the corrected distance data is obtained by tracking the distance data, the variation of the distance data is suppressed even for the object Ob that fluctuates at a fixed position, such as a curtain or a tree. Therefore, the processing unit 12 measures the time during which the corrected distance data obtained for each measurement period T1 is in a predetermined allowable range, and when this time reaches a predetermined determination period, You may judge. That is, the processing unit 12 may exclude data within the allowable range as background object data after the state in which the corrected distance data is within the allowable range continues and reaches the determination period.

  By the way, in the above-described operation example, the case where only one object Ob exists in the target space in addition to the background object has been described. However, in addition to the background object, a plurality of objects Ob may exist in the target space. Hereinafter, an operation in the case where a plurality of objects Ob in addition to the background object exist in the target space will be described. As described above, for the object Ob existing in the target space, variation in distance data and speed data is suppressed by the tracking process.

  Here, it is assumed that the distance data has changed as time passes as shown in FIG. FIG. 9 is an example of a graph showing the time variation of the distance data obtained for each measurement period T1, where the horizontal axis represents time and the vertical axis represents the distance data value. FIG. 9 shows information generated in the background object and information on the object Ob moving in the target space. The information generated by the background object or the like is represented by the characteristic L11 in FIG. 9, and the value of the distance data hardly changes over time, and the change of the value of the distance data is within the error range. On the other hand, the characteristic L12 and the characteristic L13 in FIG. 9 have a period in which the distance hardly changes temporarily, but the distance changes with the passage of time in most of the periods. That is, it is estimated that the characteristic L12 and the characteristic L13 correspond to two different objects Ob existing in the target space.

  That is, when a graph as shown in FIG. 9 is obtained, if a person sees, a characteristic indicating time variation such as L11 among these characteristics corresponds to unnecessary information generated in a background object or the like, and a characteristic L12 Therefore, it can be easily determined that two different moving objects have different characteristics L13. However, it is not easy for the processing unit 12 to recognize that the characteristics L12 and the characteristics L13 correspond to two moving objects. Therefore, the processing unit 12 includes a determination unit 128.

  The determination unit 128 tracks distance data and velocity data obtained every measurement period T1, and recognizes the same object Ob when the tracking is possible. The determination unit 128 actually uses the corrected distance data, but is simply referred to as distance data below. The determination unit 128 is on condition that the time change of the distance data can be reasonably determined as the same object Ob, and the time change of the speed data can be rationally determined as the same object Ob. When tracking distance data and velocity data, the determination unit 128 basically evaluates the relationship between two distance data adjacent in the time series and the relationship between two velocity data adjacent in the time series. However, the determination unit 128 may perform tracking by evaluating three or more relationships with respect to at least one of the distance data and the speed data.

  An example of conditions used by the determination unit 128 will be described below. Here, a case where the determination unit 128 evaluates the relationship between two distance data adjacent in the time series and the relationship between two speed data adjacent in the time series will be described as an example. In addition, a case where the distance data changes as time passes as shown in FIG. 9 is taken as an example. In the characteristics L12 and L13 shown in FIG. 9, there is a period Tn in which the distance data to the object Ob hardly changes, and there are two intersecting portions Px. Here, the crossing of the distance data does not necessarily indicate that the two objects Ob are approaching, but simply indicates that the distances from the radio wave sensor 10 are substantially equal.

  In a period in which the characteristic L12 and the characteristic L13 do not intersect, that is, a period in which the condition that the difference between the two distance data at the same time is larger than a predetermined determination threshold is satisfied, the distance data corresponding to each object Ob Can be tracked individually. That is, if the distance data and the speed data satisfy the tracking condition described above, the determination unit 128 estimates that the distance data that satisfies the condition corresponds to the same object Ob.

  On the other hand, it is confirmed that the two objects Ob are present at substantially the same distance from the radio wave sensor 10 during a period in which the condition that the difference between the two distance data at the same time is equal to or less than a predetermined determination threshold is satisfied. Since it represents, the probability that distance data will cross will become high. When the distance data intersects, the correspondence between the two distance data before the intersection and the two distance data after the intersection must be guaranteed. As described above, a label is assigned to the distance data. That is, it is necessary to correctly label the two distance data before the intersection and the two distance data after the intersection.

  Under the condition that the difference between the two distance data at the same time is equal to or less than the determination threshold value, the determination unit 128 includes the two speed data corresponding to the two distance data on a one-to-one basis and the measurement period T1 before the temporary point. Compare with two velocity data. At the time when the difference between the two distance data at the same time becomes equal to or less than the determination threshold value, the two speed data corresponding to the two distance data in a one-to-one manner are not labeled, but the previous measurement period Each of the two speed data at T1 is provided with a label. Therefore, the determination unit 128 compares each of the two speed data without a label by comparing the two speed data in the previous measurement period T1 with the two speed data without the label. Give a label.

  Here, speed data without a label is referred to as “target data”, and speed data in the immediately preceding measurement period T1 is referred to as “immediate data”. A label is assigned to the immediately preceding data. Since the determination unit 128 grasps the two labels corresponding to the two immediately preceding data, by comparing each of the two immediately preceding data with the two target data, one label is assigned to each of the two target data. Can be associated.

  Specifically, the determination unit 128 associates a label with each of the two target data according to an empirical rule that, for one object Ob, the difference between adjacent velocity data in the time series is relatively small. According to this rule of thumb, the smaller the difference between the target data and the immediately preceding data, the higher the possibility that the target data and the immediately preceding data correspond to the same object Ob. Therefore, the determination unit 128 obtains a difference between each of the two target data and one of the two immediately preceding data, and associates the same target label with the same label as the immediately preceding data for which the difference is obtained. The determination unit 128 may associate the other label of the two immediately preceding data with the remaining target data.

  The determination unit 128 gives a label to the target data by the above-described process during a period in which the difference between the two distance data at the same time is equal to or less than the determination threshold. With this process, even if the two distance data intersect, the labels are associated with each other without error, and the possibility that the two distance data can be individually tracked increases. That is, the same result as that obtained when a person tracks distance data is obtained. In the operation example shown in FIG. 9, two distance data have been described for easy understanding. However, the same processing is performed when three or more distance data are generated at the same time. That is, the difference between the two distance data is evaluated, and if any difference is equal to or less than the determination threshold, the distance data may be tracked based on the time change of the speed data.

  In the configuration example described above, the distance processing unit 125 and the speed processing unit 126 are placed after the selection unit 123, but the distance processing unit 125 and the speed processing unit 126 are placed after the selection unit 123. May be.

  The processing unit 12 described above is configured with a microcomputer as a main hardware element. The microcomputer is configured as a one-chip device including a processor that operates according to a program, a memory that stores a program that operates the processor, and a working memory. The processing unit 12 may be configured by a device selected from an FPGA (Field-Programmable Gate Array), a DSP (Digital Signal Processor), a PIC (Peripheral Interface Controller), or the like instead of a microcomputer. Alternatively, the processing unit 12 may include a processor such as an MPU (Micro Processing Unit), and may be configured by a device used by connecting a memory.

  When the processing unit 12 includes a plurality of processors, the first processing unit 121 and the second processing unit 122 can be operated simultaneously. In addition, when adopting a configuration in which the first processing unit 121 and the second processing unit 122 are provided in the subsequent stage of the frequency analysis unit 120 (configuration 3 in Table 1), the first processing unit 121 and the second processing unit 122 have the frequency. The analysis unit 120 can be shared. Since the amount of information is compressed in the subsequent stage of the frequency analysis unit 120 than in the previous stage, providing the first processing unit 121 and the second processing unit 122 in the subsequent stage of the frequency analysis unit 120 leads to a reduction in processing load.

  When the processing unit 12 is configured by a single processor, the processing unit 12 switches between the processes of the first processing unit 121 and the second processing unit 122 and executes them. That is, the processing unit 12 has a multitasking function. The processes of the first processing unit 121 and the second processing unit 122 can be switched alternately, but the process of the first processing unit 121 and the process of the second processing unit 122 are performed at an appropriate ratio. It may be. For example, the ratio of the process of the second processing unit 122 to the process of the first processing unit 121 can be set to 1:10 or the like. Alternatively, a condition for switching the process of the first processing unit 121 and the process of the second processing unit 122 is determined, and the process of the first processing unit 121 and the process of the second processing unit 122 are switched when the condition is satisfied. May be. The conditions for switching the process of the first processing unit 121 and the process of the second processing unit 122 are determined according to the conditions used by the selection unit 123 described above, for example. Note that the minimum time assigned to the process of the first processing unit 121 and the process of the second processing unit 122 is the measurement period T1, and the minimum time of the measurement period T1 is the processing cycle T0.

  The program can be provided in a state of being stored in a ROM (Read Only Memory) of the memory, or can be provided on a recording medium such as an optical disk readable by a computer or an external storage device. The program may be provided through an electric communication line such as the Internet. A program provided through a storage medium or a telecommunication line is stored in a rewritable nonvolatile memory.

  In addition, the section setting unit 127 may be configured as an interface unit that communicates with an external device operated by the user. This type of external device is selected from a general-purpose personal computer, a smartphone, a tablet terminal, and the like in addition to a dedicated setting device.

  By the way, as shown in FIG. 10, the radio wave sensor 10 mentioned above may be provided in the equipment 20. The facility device 20 includes a load device 21 and a radio wave sensor 10 and is configured to instruct the operation of the load device 21 according to the state of the object Ob monitored by the radio wave sensor 10. For example, if the load device 21 is a lighting device and the radio wave sensor 10 monitors whether or not a person as an object Ob exists in the target space, the lighting device is turned on during a period in which the person exists in the target space. Thus, the equipment 20 can be configured as described above.

  The radio wave sensor 10 described above includes a detection unit 11 and a processing unit 12. The detection unit 11 radiates a radio wave whose frequency changes over time within a predetermined measurement period T1 to the target space, and receives the radio wave from the target space. In each of the plurality of measurement periods T <b> 1, the processing unit 12 performs processing up to the object Ob existing in the target space based on the sensor signal including information on the frequency difference between the signal corresponding to the radiated radio wave and the signal corresponding to the received radio wave. Measure distance. Further, the processing unit 12 has a function of measuring the distance to the object Ob existing in the target space after removing information included in the background signal from the sensor signal obtained in each of the plurality of measurement periods T1. When the object Ob existing in the target space is considered to be stationary, the background signal removes the information of the object Ob from the sensor signal as time elapses when the object Ob is regarded as stationary. It is stipulated to increase the percentage of

  According to this configuration, since the information on the stationary object Ob is removed from the sensor signal as time elapses, the information on the moving object Ob in the target space is highlighted, and the moving object It becomes easy to extract a sensor signal related to Ob. In addition, even when multipath occurs due to reflection of radio waves by a plurality of objects Ob existing in the target space, erroneous information due to multipath can be easily excluded.

  For example, it is assumed that an object Ob stationary in the target space and a moving object Ob are mixed. In this case, there may be a path in which the moving object Ob is interposed between the radio wave sensor 10 and the stationary object Ob. Even if the information of the stationary object Ob can be removed, the moving object Ob Incorrect information that cannot be excluded may remain due to changes in the position of. For this reason, if information on the stationary object Ob is fixedly handled as temporary point information, erroneous information that cannot be excluded may remain during a period in which the moving object Ob exists in the target space. On the other hand, since the ratio of removing the information of the stationary object Ob increases with the passage of time, the change of the stationary object Ob with the passage of time is captured, There is a high possibility that the influence of erroneous information is reduced.

  The background signal is preferably a weighted sum of the sensor signal obtained in each of the plurality of measurement periods T1 and the past background signal obtained in the measurement period T1 a predetermined number of times before.

  According to this configuration, since the background signal is updated while leaving a part of the information of the background signal in the past, information that does not change with the passage of time is detected as information corresponding to the object Ob that is still stationary. It can be removed from the signal. In addition, when there is a moving object Ob, the information corresponding to the moving object Ob is only incorporated into a part of the background signal, so even if the information included in the background signal is removed from the sensor signal, the information moves. Information on the object Ob that remains is left. That is, it is possible to extract information on the moving object Ob. Further, the information of the object Ob that is temporarily stationary remains if the information included in the background signal is removed from the sensor signal if the stationary time is short, and if the stationary time is long, the sensor Ob Removed from the signal. That is, it is possible to distinguish between the object Ob that is temporarily stationary and the object Ob that is continuously stationary.

  Here, in the weighted sum, the weight coefficient for the past background signal is larger than the weight coefficient for the sensor signal, and the sum of the two weight coefficients is preferably 1.

  That is, by leaving more information on the past background signal in the updated background signal than in the sensor signal, information on the stationary object Ob is likely to remain in the updated background signal. If the weighting coefficient is adjusted here, the time until the information of the stationary object Ob can be removed from the sensor signal is adjusted.

  The background signal may be determined so that the information of the object Ob is removed from the sensor signal after the period in which the object Ob existing in the target space is considered to be stationary still reaches a predetermined holding period. desirable.

  According to this configuration, the object Ob that is still stationary can be previously removed from the sensor signal as a background object Ob, and information on the object Ob that has entered the target space is separated from the background information. It becomes easy to do.

  The processing unit 12 preferably includes a first processing unit 121, a second processing unit 122, and a selection unit 123. The first processing unit 121 removes information on the background signal from the sensor signal obtained in each of the plurality of measurement periods T1. The second processing unit 122 removes information on the object Ob that is stationary between the sensor signals obtained in each of the plurality of measurement periods T1. The selection unit 123 selects information satisfying the condition of the target object Ob from the information obtained from the output of the first processing unit 121 and the information obtained from the output of the second processing unit 122.

  That is, the information obtained from the output of the first processing unit 121 can be used to exclude information on the object Ob that is constantly present in the target space and obtain information on the object Ob that has newly entered the target space. . In addition, information on the object Ob moving in the target space can be obtained by excluding information on the object Ob stationary in the target space by the information obtained from the output of the second processing unit 122. And since the selection part 123 selects the information obtained from the output of the 1st process part 121 and the information obtained from the output of the 2nd process part 122 according to conditions, it is obtained from the output of the 1st process part 121. The information and the information obtained from the output of the second processing unit 122 are complemented. As a result, there is a high possibility that information regarding the object Ob can be correctly extracted even when the object Ob exists in front of the background object or when the object Ob is stationary in the target space.

  The processing unit 12 further includes a calculation unit 124 that multiplies the frequency component of the output of the first processing unit 121 by a correction coefficient, and this correction coefficient becomes smaller as the power or amplitude value of the frequency component of the background signal increases. It is desirable that

  When the information included in the background signal is removed from the sensor signal in the first processing unit 121, the information corresponding to the stationary object Ob is included even if the background signal includes only the stationary object Ob. There may be a residual in the frequency component. This residual tends to increase as the power or amplitude value of the frequency component of the background signal increases. Therefore, by multiplying the frequency component of the output of the first processing unit 121 by a correction coefficient that is smaller as the power or amplitude value of the frequency component of the background signal is larger, the frequency component corresponding to the stationary object Ob is obtained. It is possible to reduce the residual. That is, there is a high possibility that information on the stationary object Ob can be excluded from the output of the first processing unit 121, and information on the object Ob that has entered the target space can be easily distinguished.

  As the sensor signal, a time-domain signal obtained by mixing a signal corresponding to the radiated radio wave and a signal corresponding to the received radio wave can be used. The sensor signal may be a frequency domain signal obtained by performing frequency analysis after mixing a signal corresponding to the radiated radio wave and a signal corresponding to the received radio wave.

  Regardless of whether the sensor signal is in the time domain or the frequency domain, the first processing unit 121 determines the difference between the sensor signal obtained in advance at a predetermined time and the sensor signal obtained in each of the plurality of measurement periods T1. Can be configured to determine Further, whether the sensor signal is in the time domain or the frequency domain, the second processing unit 122 may be configured to obtain a difference between the sensor signals obtained in each of the plurality of measurement periods T1. Is possible.

  These configurations are specific examples for excluding information regarding unnecessary objects Ob. When the sensor signal is a signal in the time domain, the component of the object Ob that is stationary in the time domain is removed. That is, since it is only necessary to obtain the frequency spectrum of the mixed signal from which the background component is removed, there is a possibility that the processing load when obtaining the frequency spectrum is reduced.

  On the other hand, when the sensor signal is a signal in the frequency domain, the component of the object Ob that is stationary in the frequency domain is removed. When there is a multipath of a transmission wave or a reflected wave, there is a possibility that the component corresponding to the stationary object Ob will fluctuate in the mixed signal, but such fluctuation in the process of obtaining the frequency spectrum from the mixed signal. Therefore, the component corresponding to the stationary object Ob can be easily removed.

  The processing unit 12 preferably includes a distance processing unit 125 and a speed processing unit 126. The distance processing unit 125 obtains distance data corresponding to the distance from the frequency difference to the object Ob in each of the plurality of measurement periods T1. The speed processing unit 126 obtains speed data corresponding to the speed of the object Ob from the plurality of distance data obtained in each of the plurality of measurement periods T1 and the length of the measurement period T1. Further, the speed processing unit 126 is configured to obtain corrected distance data in which variation is suppressed so as to approach a true value based on a time series of distance data, and to obtain speed data using the corrected distance data.

  According to this configuration, variation in distance data due to fluctuations in part of the object Ob, variation in distance data due to multipath of transmitted waves or reflected waves, and the like are suppressed. That is, the process of associating the distance data with the object Ob is facilitated. Further, since the speed data is obtained using the corrected distance data, variations in the speed data are also suppressed.

  The processing unit 12 preferably includes a determination unit 128. The determination unit 128 outputs a plurality of correction distances during a period in which the speed processing unit 126 outputs a plurality of correction distance data at the same time and a difference between the plurality of correction distance data at the same time is equal to or less than a predetermined determination threshold value. Each data is individually tracked based on time changes.

  According to this configuration, when a plurality of correction distance data are obtained at the same time for a plurality of objects Ob, the correction distance data obtained from the individual objects Ob can be individually tracked. That is, when the difference between the plurality of correction distance data obtained at the same time is equal to or less than the determination threshold value, each of the plurality of correction distance data is individually tracked, so that the correction distance data can be correctly associated with the object Ob. It becomes possible. In this way, even when the plurality of objects Ob cannot be distinguished only by the correction distance data, the possibility that the plurality of objects Ob can be individually distinguished using the temporal change of the correction distance data is increased. Can be associated correctly.

  The processing unit 12 excludes the corrected distance data within the allowable range after the plurality of corrected distance data obtained in each of the plurality of measurement periods T1 is within the predetermined allowable range after the predetermined determination period is reached. It may be configured to.

  According to this configuration, after the state in which the distance to the object Ob does not change reaches the determination period, the information on the corresponding object Ob is excluded. That is, the object Ob whose distance does not change is regarded as a stationary background, and information on the distance to the object Ob is excluded. Therefore, it becomes easy to distinguish the information of the moving object Ob that has entered the target space.

  The processing unit 12 further includes a section setting unit 127 that determines one or more specific sections divided by distance in at least a part of the target space. The background signal is determined so as to remove the information of the object Ob existing in the specific section from the sensor signal, and is updated when the reacquisition of the background signal corresponding to the object Ob existing in the specific section is instructed.

  According to this configuration, it is possible to set a specific section in the target space, and it is also possible to set the target space by dividing it into a plurality of specific sections. Further, since the information of the object Ob existing in the specific section is removed from the sensor signal as a background signal, the object Ob existing in the specific section is treated as the background. In the specific section, the background signal is updated when the background signal re-acquisition is instructed. Therefore, when the background signal re-acquisition is instructed when the background object Ob is changed in the specific section, the background signal is updated. Is updated.

  The equipment 20 described above includes a radio wave sensor 10 and a load device 21 that is instructed to operate according to the state of the object Ob monitored by the radio wave sensor 10.

According to this configuration, the operation of the load device 21 can be instructed according to the state of the object Ob.
The state of the object Ob includes, for example, a state of whether or not the object Ob exists at a predetermined distance, a state of whether the object Ob is approaching or moving away from the radio wave sensor 10, and the like. Therefore, it becomes possible to control operations of various load devices 21 such as a lighting device, an air conditioner, a display device, and an automatic door according to the state of the object Ob. For example, when the load device 21 is a lighting device such as a street lamp, the radio wave sensor 10 monitors a region where a person exists and a region where a person does not exist, and controls the lighting state of the lighting device corresponding to each region. Is possible.

  The above-described embodiment is an example of the present invention. For this reason, the present invention is not limited to the above-described embodiment, and various modifications can be made according to design and the like as long as the technical idea according to the present invention is not deviated from this embodiment. Of course, it can be changed.

DESCRIPTION OF SYMBOLS 10 Radio wave sensor 11 Detection part 12 Processing part 20 Equipment 21 Load device 121 1st processing part 122 2nd processing part 123 Selection part 124 Calculation part 125 Distance processing part 126 Speed processing part 127 Section setting part 128 Judgment part Ob Object T1 Measurement period

Claims (14)

A detector that radiates radio waves whose frequency changes over time within a predetermined measurement period to the target space and receives radio waves from the target space;
Processing for measuring a distance to an object existing in the target space based on a sensor signal including information on a frequency difference between a signal corresponding to a radiated radio wave and a signal corresponding to a received radio wave in each of the plurality of measurement periods With
The processor is
It has a function of measuring a distance to an object existing in the target space after removing information included in a background signal from a plurality of sensor signals obtained in each of the measurement periods,
When the object existing in the target space is considered to be still still, the background signal indicates the information of the object as the sensor signal as time elapses when the object is considered to be stationary. Radio wave sensor characterized in that it is determined to increase the rate of removal from the sensor. The radio wave sensor according to claim 1, wherein the background signal is a weighted sum of a sensor signal obtained in each of the plurality of measurement periods and a past background signal obtained in a predetermined measurement period. The radio wave sensor according to claim 2, wherein, in the weighted sum, a weighting factor for the past background signal is larger than a weighting factor for the sensor signal, and a sum of the two weighting factors is 1. The background signal is determined to remove information on the object from the sensor signal after a period during which an object existing in the target space is considered to be stationary still reaches a predetermined holding period. The radio wave sensor according to any one of claims 1 to 3. The processor is
A first processing unit for removing information of the background signal from the sensor signal obtained in each of a plurality of measurement periods;
A second processing unit for removing information on an object stationary between the sensor signals obtained in each of the plurality of measurement periods;
5. A selection unit that selects information satisfying a condition of a target object from information obtained from the output of the first processing unit and information obtained from the output of the second processing unit. The radio wave sensor according to item 1. The processor is
A calculation unit that multiplies the frequency component of the output of the first processing unit by a correction coefficient;
The radio wave sensor according to claim 5, wherein the correction coefficient is determined to be a smaller value as the power or amplitude value of the frequency component of the background signal is larger. The sensor signal is a time-domain signal obtained by mixing a signal corresponding to the radio wave radiated by the detection unit and a signal corresponding to the received radio wave,
The radio wave sensor according to claim 5 or 6, wherein the first processing unit and the second processing unit are configured to remove information on a stationary object in a time domain. The sensor signal is a frequency domain signal obtained by performing frequency analysis after mixing a signal corresponding to the radio wave radiated by the detection unit and a signal corresponding to the received radio wave,
The radio wave sensor according to claim 5 or 6, wherein the first processing unit and the second processing unit are configured to remove information on a stationary object in a frequency domain. The sensor signal is
A time-domain signal obtained by mixing a signal corresponding to the radio wave radiated by the detection unit and a signal corresponding to the received radio wave, a signal corresponding to the radio wave radiated by the detection unit, and a signal corresponding to the received radio wave. A frequency domain signal that has been subjected to frequency analysis after mixing,
The first processing unit is configured to remove information on a stationary object in the time domain, and the second processing unit is configured to remove information on a stationary object in the frequency domain. The radio wave sensor according to claim 5 or 6. The processor is
A distance processing unit that obtains distance data corresponding to the distance from the frequency difference to the object in each of the plurality of measurement periods;
A speed processing unit for obtaining speed data corresponding to the speed of the object from a plurality of the distance data obtained in each of the plurality of measurement periods and the length of the measurement period;
The speed processing unit is configured to obtain correction distance data in which variation is suppressed so as to approach a true value based on a time series of the distance data, and to obtain speed data using the correction distance data. Item 10. The radio wave sensor according to any one of Items 1 to 9. The processor is
Each of the plurality of correction distance data is output in a period in which the speed processing unit outputs a plurality of correction distance data at the same time and a difference between the plurality of correction distance data at the same time is a predetermined determination threshold value or less. The radio wave sensor according to claim 10, further comprising a determination unit that individually tracks based on a time change. The processor is
A plurality of the correction distance data obtained in each of the plurality of measurement periods are excluded from the correction distance data within the allowable range after the state where the correction distance data is within the predetermined allowable range reaches a predetermined determination period. The radio wave sensor according to claim 10 or 11, wherein the radio wave sensor is configured. The processor is
A section setting unit that defines one or more specific sections divided by distance in at least a part of the target space,
The background signal is determined to remove information on an object existing in the specific section from the sensor signal, and is updated when an instruction to reacquire a background signal corresponding to the object existing in the specific section is given. Item 13. The radio wave sensor according to any one of items 1 to 12. The radio wave sensor according to any one of claims 1 to 13,
A facility device including a radio wave sensor, comprising: a load device whose operation is instructed according to a state of the object monitored by the radio wave sensor.

Images