JP6830300B2 - Sensor information acquisition device, sensor information generation device and sensor information generation / acquisition system - Google Patents

Description

The present disclosure relates to a sensor information generation / acquisition system including a sensor information generation device for generating sensor information and a sensor information acquisition device for acquiring sensor information.

Prior art documents such as Patent Document 1 exist with respect to a sensor information generation / acquisition system including a sensor information generation device for generating sensor information and a sensor information acquisition device for acquiring sensor information. The contents and issues will be explained.

The transmitter / receiver corresponds to a sensor information acquisition device that acquires sensor information, and transmits a question wave to the wireless response device via an antenna. The wireless response device corresponds to a sensor information generator that generates sensor information, and receives a question wave from the transmission / reception device via an antenna.

The radio response device includes an impedance converter and a SAW (Surface Acoustic Wave) -ID tag. The impedance conversion unit is provided between the antenna and the SAW-ID tag, and changes the impedance in response to an external stimulus.

When there is no external stimulus, the impedance converter does not match the impedance between the antenna and the SAW-ID tag, so the SAW-ID tag does not input a question wave, cannot generate a response wave, and is wireless. The responder cannot transmit the response wave to the transmitter / receiver via the antenna. When there is an external stimulus, the impedance converter matches the impedance between the antenna and the SAW-ID tag, so the SAW-ID tag can input a question wave and generate a response wave, and is a wireless response device. Can transmit the response wave to the transmitter / receiver via the antenna. The transmitter / receiver determines the presence / absence of an external stimulus based on the presence / absence of a response wave.

JP-A-2007-129657

Although not described in Patent Document 1, the transmitter / receiver may calculate the cross-correlation between the question wave and the response wave in the time domain. The transmitter / receiver can then identify the radio responder with a SAW delay time equal to the peak time based on the peak time of cross-correlation in the time domain. Then, the transmission / reception device can determine the presence / absence of an external stimulus in the radio response device based on the amplitude at the peak time of the cross-correlation in the time domain.

However, in reality, the question wave wraps around between the transmitting and receiving antennas. Here, the wraparound wave has a high power level because it is only attenuated by short-distance spatial propagation. On the other hand, the response wave has a low power level because it is attenuated by long-distance spatial propagation and SAW propagation. For the response wave, the long-distance spatial propagation time is much shorter than the SAW propagation time.

The transmitter / receiver then calculates the cross-correlation between the interrogation wave and the received wave (including the response wave and the wraparound wave) in the time domain. Therefore, in the transmitter / receiver, is the peak time of the cross-correlation in the time domain due to the radio response device having the SAW delay time equal to the peak time, or is it due to the wraparound between the transmit / receive antennas of the interrogation wave? Cannot be identified. Then, in the transmitter / receiver, is the amplitude at the peak time of the cross-correlation in the time domain due to the presence or absence of an external stimulus in the radio response device, or is it due to the wraparound between the transmit / receive antennas of the question wave? , Cannot be identified.

Therefore, in order to solve the above-mentioned problems, in the present disclosure, even if there is wraparound between the transmission / reception antennas of the question wave in the sensor information acquisition device, the sensor information generation device is generated from the response wave received from the sensor information generation device. The purpose is to extract the sensor information with high accuracy.

In order to achieve the above object, it was decided to use the FMCW (Frequency Modified Continuous Wave) method used in the radar system in the sensor information generation / acquisition system. In using the FMCW method, the pulse compression ratio TΔF can be increased and the noise immunity can be improved by increasing the signal length T and / or increasing the signal bandwidth ΔF within the band limitation range. By further limiting the bandwidth, the radio channel mask standard can be satisfied without suppressing the power level too much.

Here, it can be said that the spectrum of the FMCW interrogation wave is almost flat in the signal bandwidth. Therefore, the autocorrelation of the FMCW interrogation wave should be zero in a time that is a natural number multiple of the reciprocal of the signal bandwidth. Then, the cross-correlation between the FMCW interrogation wave and the FMCW wraparound wave should be 0 at a time that is a natural number multiple of the reciprocal of the signal bandwidth. This is because it can be said that the spatial propagation time of the FMCW wraparound wave is almost 0 as compared with the spatial propagation time of the FMCW response wave + the intentionally applied delay time (for example, SAW delay time).

Therefore, the delay time applied by the sensor information generator is made equal to the natural number multiple of the reciprocal of the signal bandwidth. Then, the intercorrelation between the FMCW interrogation wave and the FMCW received wave (including the FMCW response wave and the FMCW wraparound wave) is not affected by the contribution from the FMCW wraparound wave in the delay time provided by the sensor information generator. , Should be affected only by the contribution from the FMCW response wave. Therefore, even if the sensor information acquisition device wraps around between the transmitting and receiving antennas of the FMCW question wave, the sensor information generated by the sensor information generating device is extracted with high accuracy from the FMCW response wave received from the sensor information generating device. You should be able to.

However, in reality, the spectrum of the FMCW interrogation wave maintains a flat level around the center of the signal band, but may be attenuated from the flat level at the edge of the signal band. Then, the autocorrelation of the FMCW interrogation wave does not become 0 at a time that is a natural number multiple of the reciprocal of the total signal bandwidth. Then, the cross-correlation between the FMCW interrogation wave and the FMCW wraparound wave does not become 0 at a time of a natural number multiple of the reciprocal of the total signal bandwidth.

Therefore, even if the delay time applied by the sensor information generator is equal to the time that is a natural number multiple of the reciprocal of the total signal bandwidth, the above effect cannot be achieved. That is, the intercorrelation between the FMCW interrogation wave and the FMCW received wave (including the FMCW response wave and the FMCW wraparound wave) is also affected by the contribution from the FMCW wraparound wave in the delay time provided by the sensor information generator. To. Then, in the sensor information acquisition device, when there is a wraparound between the transmission / reception antennas of the FMCW question wave, the sensor information generated by the sensor information generator is extracted with high accuracy from the FMCW response wave received from the sensor information generator. I can't.

Therefore, in the present disclosure, it is decided to cut out only the spectrum around the center of the signal band that maintains a flat level from the spectrum of the FMCW interrogation wave, and calculate the correlation function in the time domain. Then, the autocorrelation of the FMCW interrogation wave becomes 0 at a time that is a natural number multiple of the reciprocal of the flat bandwidth. Then, the cross-correlation between the FMCW interrogation wave and the FMCW wraparound wave becomes 0 at a time that is a natural number multiple of the reciprocal of the flat bandwidth.

Therefore, if the delay time applied by the sensor information generator is equal to the natural number multiple of the reciprocal of the flat bandwidth, the above-mentioned effect can be obtained. That is, the intercorrelation between the FMCW interrogation wave and the FMCW received wave (including the FMCW response wave and the FMCW wraparound wave) is not affected by the contribution from the FMCW wraparound wave in the delay time provided by the sensor information generator. .. Then, even if the sensor information acquisition device wraps around between the transmitting and receiving antennas of the FMCW question wave, the sensor information generated by the sensor information generating device is extracted with high accuracy from the FMCW response wave received from the sensor information generating device. be able to.

Note that the flat bandwidth is narrower than the total signal bandwidth. Therefore, the resolution of the delay time applied by the sensor information generator seems to be apparently deteriorated in the present disclosure. However, the effect of contributions from FMCW wraparound waves is largely non-existent in this disclosure. Therefore, the resolution of the delay time applied by the sensor information generator can be substantially improved in the present disclosure.

Specifically, the present disclosure is a sensor information acquisition device that acquires sensor information from a sensor information generation device, and includes a question wave transmission unit that transmits a band-limited FMCW question wave to the sensor information generation device. The FMCW response wave applied to the FMCW interrogation wave is modulated by the sensor information and delayed by a natural number of times the inverse of the bandwidth in which the spectrum of the FMCW interrogation wave is flat around the center of the bandwidth. FMCW reception including the FMCW interrogation wave, the FMCW response wave, and the FMCW wraparound wave in a bandwidth in which the response wave receiving unit received from the sensor information generator and the spectrum of the FMCW interrogation wave are flat around the band center. The spectrum of the FMCW interrogation wave and / or at a time other than 0 when the amplitude of the intercorrelation in the time region peaks with the intercorrelation calculation unit that calculates the intercorrelation between the wave and the time region. By calculating the amplitude and / or phase of the mutual correlation in the time region at a time that is several times the natural number of the inverse of the bandwidth that is flat around the center of the bandwidth, the FMCW wraparound wave of the FMCW received waves can be obtained. The sensor information acquisition device includes a sensor information extraction unit that extracts the sensor information used for amplitude modulation and / or phase modulation from the FMCW response wave to be removed.

Further, the present disclosure is a sensor information generation device that provides sensor information to a sensor information acquisition device, and includes a question wave receiving unit that receives a band-limited FMCW question wave from the sensor information acquisition device, and the sensor information. The FMCW interrogation wave is modulated by the interrogation wave modulation unit, and the FMCW interrogation wave is delayed by a natural number of times the inverse of the bandwidth in which the spectrum of the FMCW interrogation wave is flat around the center of the bandwidth. The FMCW interrogation wave is subjected to a modulation according to the sensor information and a delay of a natural number of times the inverse of the bandwidth in which the spectrum of the FMCW interrogation wave is flat around the center of the bandwidth. The sensor information generation device includes a response wave transmission unit that transmits the FMCW response wave to the sensor information acquisition device.

Further, the present disclosure is a sensor information generation / acquisition system including the sensor information acquisition device described above and the sensor information generation device described above.

According to this configuration, the sensor information acquisition device appropriately sets the bandwidth for calculating the correlation function in the time domain, and the sensor information generator intentionally sets the delay time (for example, SAW delay time). With proper settings, the above effects can be achieved as a whole system. That is, even if the sensor information acquisition device wraps around between the transmitting and receiving antennas of the FMCW inquiry wave, the sensor information generated by the sensor information generator is extracted with high accuracy from the FMCW response wave received from the sensor information generator. be able to.

As described above, according to the present disclosure, even if the sensor information acquisition device wraps around between the transmission / reception antennas of the question wave, the sensor information generated by the sensor information generation device from the response wave received from the sensor information generation device is used. It can be extracted with high accuracy.

It is a figure which shows the outline of the sensor information generation / acquisition system of this disclosure. It is a figure which shows the structure of the sensor information acquisition apparatus of this disclosure. It is a figure which shows the structure of the sensor information generation apparatus of this disclosure. It is a figure which shows the FMCW interrogation wave before the band limitation of this disclosure. It is a figure which shows the FMCW interrogation wave after the band limitation of this disclosure. It is a figure which shows the spectrum of the FMCW interrogation wave of this disclosure. It is a figure which shows the calculation method of the time correlation of the 1st comparative example. It is a figure which shows the calculation result of the autocorrelation of the FMCW interrogation wave of the 1st comparative example. It is a figure which shows the calculation method of the time correlation of the 2nd comparative example. It is a figure which shows the calculation result of the autocorrelation of the FMCW interrogation wave of the 2nd comparative example. It is a figure which shows the 1st method of calculating the time correlation of this disclosure. It is a figure which shows the calculation result of the autocorrelation of the FMCW interrogation wave of the 1st disclosure. It is a figure which shows the 2nd method of calculating the time correlation of this disclosure. It is a figure which shows the calculation result of the autocorrelation of the FMCW inquiry wave of the second present disclosure. It is a figure which shows the spectrum of the FMCW received wave of this disclosure. It is a figure which shows the calculation result of the cross-correlation between the FMCW inquiry wave and the FMCW received wave of this disclosure.

Embodiments of the present disclosure will be described with reference to the accompanying drawings. The embodiments described below are examples of the embodiments of the present disclosure, and the present disclosure is not limited to the following embodiments.

(Outline of the sensor information generation / acquisition system of the present disclosure)
The outline of the sensor information generation / acquisition system of the present disclosure is shown in FIG. The sensor information generation / acquisition system S includes a sensor information acquisition device 1 and sensor information generation devices 2-1, 2-2, and 2-3. The sensor information generation / acquisition system S detects, for example, the vibration of the bridge B. The sensor information acquisition device 1 is installed, for example, on the shoulder of a bridge B. The sensor information generators 2-1, 2-2, and 2-3 are installed, for example, directly under each lane of the bridge B.

The sensor information acquisition device 1 transmits a question wave to the sensor information generation devices 2-1, 2-2, and 2-3. Each sensor information generation device 2-1, 2-2, 2-3 receives a question wave from the sensor information acquisition device 1, modulates according to the vibration of each lane of the bridge B, and delays by a predetermined time (for example, each predetermined time). , SAW delay) is transmitted to the sensor information acquisition device 1 as each response wave. The sensor information acquisition device 1 receives each response wave from each sensor information generation device 2-1, 2-2, 2-3, modulates according to the vibration of each lane of the bridge B, and delays by a predetermined time (each predetermined time). For example, the vibration information of each lane of the bridge B is extracted from each response wave subjected to SAW delay). Information on the vibration of each lane of the bridge B is notified to the administrator directly or via the Internet I or the like.

In this way, each sensor information generation device 2-1, 2-2, 2-3 applies a delay (for example, SAW delay) for each predetermined time to the question wave, so that each sensor information acquisition device 1 The sensor information generators 2-1, 2-2, and 2-3 can be identified. Then, each sensor information generation device 2-1, 2-2, 2-3 applies modulation according to the vibration of each lane of the bridge B to the question wave, so that the sensor information acquisition device 1 can be used for each of the bridge B. Information on lane vibration can be extracted from the response wave. Further, each sensor information generator 2-1, 2-2, 2-3 only passively responds to the question wave and the vibration of each lane of the bridge B, and can operate without a power source. , It is possible to facilitate the installation and maintenance of the bridge B directly under each lane.

In the present disclosure, the FMCW method used in the radar system is used in the sensor information generation / acquisition system S. In using the FMCW method, the pulse compression ratio TΔF can be increased and the noise immunity can be improved by increasing the signal length T and / or increasing the signal bandwidth ΔF within the band limitation range. By further limiting the bandwidth, the radio channel mask standard can be satisfied without suppressing the power level too much.

The configuration of the sensor information acquisition device of the present disclosure is shown in FIG. The sensor information acquisition device 1 is composed of a question wave transmission unit 11, a response wave reception unit 12, a cross-correlation calculation unit 13, and a sensor information extraction unit 14. The cross-correlation calculation unit 13 is composed of an FFT calculation unit 131, a band center cutout unit 132, a complex conjugate unit 133, an FFT calculation unit 134, a band center cutout unit 135, a spectrum multiplication unit 136, and an IFFT calculation unit 137. The sensor information extraction unit 14 includes a peak detection unit 141, an amplitude / phase calculation unit 142, and a vibration information extraction unit 143.

The configuration of the sensor information generator of the present disclosure is shown in FIG. N sensor information generators 2-1 ..., 2-N are installed. The sensor information generation device 2-1 is composed of a question wave receiving unit 21-1, a question wave modulation unit 22-1, a question wave delay unit 23-1, and a response wave transmission unit 24-1. ... The sensor information generator 2-N is composed of a question wave receiving unit 21-N, a question wave modulation unit 22-N, a question wave delay unit 23-N, and a response wave transmitting unit 24-N.

The interrogation wave transmission unit 11 transmits the band-limited FMCW interrogation wave to the sensor information generators 2-1 ..., 2-N. The interrogation wave receiving units 21-1, ..., 21-N receive the band-limited FMCW interrogation wave from the sensor information acquisition device 1.

Each interrogation wave modulation unit 22-1, ..., 22-N applies modulation based on each sensor information to the FMCW interrogation wave. Examples of the sensor information include information such as vibration, pressure, temperature, and humidity (more specifically, information on vibration of each lane of the bridge B). Each interrogation wave delay unit 23-1, ..., 23-N applies a delay of only T ₁ , ..., _{TN for} each predetermined time to the FMCW interrogation wave. Examples of each interrogation wave delay unit include a sound wave propagation circuit (more specifically, a SAW delay circuit) that converts an electromagnetic wave into a sound wave or the like, propagates the sound wave or the like, and reconverts the sound wave or the like into an electromagnetic wave. ..

Each response wave transmitter 24-1, ..., 24-N is subjected to modulation by each sensor information and delay of T ₁ , ..., _{TN for} each predetermined time on the FMCW interrogation wave. Each FMCW response wave is transmitted to the sensor information acquisition device 1. The response wave receiving unit 12 transmits each FMCW response wave applied to the FMCW interrogation wave by the modulation by each sensor information and the delay of T ₁ , ..., _{TN for} each predetermined time. Received from 24-1, ..., 24-N.

In reality, the FMCW interrogation wave wraps around between the interrogation wave transmitting unit 11 and the response wave receiving unit 12. Here, the FMCW wraparound wave has a high power level because it is only attenuated by short-distance spatial propagation. On the other hand, each FMCW response wave has a low power level because it is subjected to long-distance spatial propagation and attenuation by each interrogation wave delay unit 23-1, ..., 23-N. As shown in FIGS. 8, 10, 12, 14, and 16 for each FMCW response wave, the time for long-distance spatial propagation is the propagation of each interrogation wave delay unit 23-1, ..., 23-N. Much shorter than the time of.

The cross-correlation calculation unit 13 calculates the cross-correlation in the time region between the FMCW interrogation wave and the FMCW received wave including each FMCW response wave and the FMCW wraparound wave. The sensor information extraction unit 14 extracts each sensor information used for each modulation from each FMCW response wave excluding the FMCW wraparound wave among the FMCW received waves. The cross-correlation calculation unit 13 and the sensor information extraction unit 14 will be described in detail in the following paragraphs while comparing a comparative example with the present disclosure.

In the following paragraphs, the equivalent low frequency system will be used for explanation. The FMCW interrogation wave before the band limitation of the present disclosure is shown in FIG. The interrogation wave transmission unit 11 generates the FMCW interrogation wave shown in FIG. 4 by sweeping the frequency from −500 kHz to +500 kHz while keeping the amplitude constant over 1 msec. The FMCW interrogation wave after the band limitation of the present disclosure is shown in FIG. The interrogation wave transmission unit 11 generates the FMCW interrogation wave shown in FIG. 5 by multiplying the FMCW interrogation wave shown in FIG. 4 by a window function for reducing the power outside the signal band in the time domain. Examples of the window function for reducing the power outside the signal band include a Cosine-Tukey window function, a Hamming window function, a Hanning window function, and a Blackmann window function. The spectrum of the FMCW interrogation wave of the present disclosure is shown in FIG. The vertical axis of FIG. 6 indicates the power level of the spectrum of the FMCW interrogation wave in logarithmic display dB. The power level of the spectrum of the FMCW interrogation wave is proportional to the U (k) described in Equations 1 and 5 described later. The signal bandwidth after band limitation is 1 MHz.

As shown in FIG. 6, it can be said that the spectrum of the FMCW interrogation wave is substantially flat at a signal bandwidth of 1 MHz except for the signal band end. Therefore, the autocorrelation of the FMCW interrogation wave should be 0 at the reciprocal 1 μsec of the signal bandwidth of 1 MHz, the natural number 1, ..., N times the time 1, ..., N μsec. The cross-correlation between the FMCW interrogation wave and the FMCW wraparound wave should be 0 at the reciprocal 1 μsec of the signal bandwidth of 1 MHz, the natural number 1, ..., N times the time 1, ..., N μsec. is there. This is because it can be said that the spatial propagation time of the FMCW wraparound wave is almost 0 as compared with each spatial propagation time of each FMCW response wave + each intentionally applied delay time (for example, each SAW delay time). Is.

Therefore, the time T ₁ , ..., _TN of each delay applied by each sensor information generator 2-1, ..., 2-N is a natural number 1, ..., The reciprocal of 1 μsec with a signal bandwidth of 1 MHz. Equal to N times the time 1, ..., Nμsec. Then, the mutual correlation between the FMCW interrogation wave and the FMCW received wave (including each FMCW response wave and FMCW wraparound wave) is the delay applied by each sensor information generator 2-1 ..., 2-N. At times T ₁ , ..., _TN , it should not be affected by the contribution from the FMCW wraparound wave, but only by the contribution from each FMCW response wave. Therefore, in the sensor information acquisition device 1, even if there is a wraparound between the question wave transmission unit 11 of the FMCW question wave and the response wave reception unit 12, each sensor information generation device 2-1 ..., 2- From each FMCW response wave received from N, each sensor information generated by each sensor information generator 2-1 ..., 2-N should be able to be extracted with high accuracy. However, it will be described first as a comparative example that the above-mentioned effects cannot be actually achieved. Then, the method for achieving the above-mentioned effect will be described next as the present disclosure.

(Calculation method of time correlation of comparative example and calculation result of autocorrelation of FMCW question wave)
In the comparative example, in the spectrum of the FMCW interrogation wave, the correlation function in the time domain is calculated in the bandwidth that is flat around the center of the band and the bandwidth that is attenuated at the signal band edge.

As shown in Equation 1, the FFT calculation unit 131 Fourier transforms the FMCW interrogation wave u (n) in the time domain into the FMCW interrogation wave U (k) in the frequency domain. Here, n represents a time normalized using a sampling period. As shown in FIGS. 7 and 9, the band center cutout portion 132 has a bandwidth that is flat around the band center and a bandwidth that is attenuated at the signal band edge from the FMCW interrogation wave U (k) in the frequency domain. , cut out FMCW interrogation wave _U B (k) in the frequency domain. Complex conjugate unit 133, a FMCW interrogation wave _U B (k) in the frequency domain, to the complex conjugate of the FMCW interrogation wave _U ^B * (k) in the frequency domain.

The FMCW received wave x (n) in the time domain is a superposition of each FMCW response wave v (n) and the FMCW wraparound wave w (n) in the time domain. As shown in Equation 2, the FFT calculation unit 134 sets the FMCW received wave x (n) = v (n) + w (n) in the time domain and the FMCW received wave X (k) = V (k) in the frequency domain. ) + W (k) Fourier transform. Band center cutting unit 135, the FMCW reception wave X in the frequency domain (k) = V (k) + W (k), FMCW interrogation wave band center cutting unit 132 in the frequency domain _U B (k) of in the same bandwidth and the bandwidth for cutting, cutting out the FMCW reception wave _X B in the frequency domain _{(k) = V B (k} ) + W B (k).

Spectral multiplication unit 136, as shown in Equation 3, and * FMCW interrogation wave _U ^B in the frequency domain (k), FMCW reception wave _X B in the frequency domain _{(k) = V B (k} ) + W B (k ) And, and the cross-correlation Y _B (k) in the frequency domain is calculated. As shown in Equation 4, the IFFT calculation unit 137 inverse-Fourier transforms the cross-correlation Y _B (k) in the frequency domain into the cross-correlation y _B (n) in the time domain. The cross-correlation y _B (n) in the time domain includes y _{B, v} (n) and y _{B, w} (n). y _{B, v} (n) represent the cross-correlation between the FMCW interrogation wave and each FMCW response wave in the time domain. y _{B, w} (n) represent the cross-correlation between the FMCW interrogation wave and the FMCW wraparound wave in the time domain, and are proportional to the autocorrelation of the FMCW interrogation wave in the time domain.

Here, we want to calculate the cross-correlation y _{B, v} (n) in the time domain between the FMCW inquiry wave and each FMCW response wave, and extract the delay time and sensor information of each FMCW response wave. Then, it is necessary to reduce the cross-correlation y _{B, w} (n) in the time domain between the FMCW interrogation wave and the FMCW wraparound wave. That is, it is necessary to reduce the autocorrelation of the FMCW interrogation wave in the time domain.

FIG. 7 shows a method of calculating the time correlation of the first comparative example. The vertical axis of FIG. 7 shows the power level of the spectrum of the FMCW interrogation wave in logarithmic display dB. The power level of the spectrum of the FMCW interrogation wave is proportional to the U (k) described in Equation 1 above. The bandwidth used to calculate the time correlation is 1 MHz, including the bandwidth that is flat around the center of the band and the bandwidth that is attenuated at the signal band edge. The sampling frequency is 4 MHz.

The calculation result of the autocorrelation of the FMCW interrogation wave of the first comparative example is shown in FIG. The vertical axis of FIG. 8 shows the power level of the autocorrelation of the FMCW interrogation wave of the first comparative example in logarithmic display dB. The autocorrelation power level of the FMCW interrogation wave of the first comparative example is proportional to y _{B, w} (n) described in Equation 4 described above. As described above, the autocorrelation of the FMCW interrogation wave does not become 0 at the reciprocal 1 μsec of the total signal bandwidth of 1 MHz, the natural number 1, ..., N times the time 1, ..., N μsec. That is, the cross-correlation y _{B, w} (n) between the FMCW interrogation wave and the FMCW wraparound wave is a natural number 1, ..., N times the time 1, ..., The reciprocal 1 μsec of the total signal bandwidth of 1 MHz. It does not become 0 in Nμsec.

FIG. 9 shows a method of calculating the time correlation of the second comparative example. The vertical axis of FIG. 9 shows the power level of the spectrum of the FMCW interrogation wave in logarithmic display dB. The power level of the spectrum of the FMCW interrogation wave is proportional to the U (k) described in Equation 1 above. The bandwidth used to calculate the time correlation is 1 MHz, including the bandwidth that is flat around the center of the band and the bandwidth that is attenuated at the signal band edge. The sampling frequency is also 1 MHz.

The calculation result of the autocorrelation of the FMCW interrogation wave of the second comparative example is shown in FIG. The vertical axis of FIG. 10 shows the power level of the autocorrelation of the FMCW interrogation wave of the second comparative example in logarithmic display dB. The autocorrelation power level of the FMCW interrogation wave of the second comparative example is proportional to y _{B, w} (n) described in Equation 4 described above. As described above, the autocorrelation of the FMCW interrogation wave does not become 0 at the natural number 1, ..., N times the time 1, ..., Nμsec, which is the reciprocal of 1 μsec of the total signal bandwidth of 1 MHz. That is, the cross-correlation y _{B, w} (n) between the FMCW interrogation wave and the FMCW wraparound wave is a natural number 1, ..., N times the time 1, ..., The reciprocal 1 μsec of the total signal bandwidth of 1 MHz. It does not become 0 in Nμsec.

Therefore, the time T ₁ , ..., _TN of each delay applied by each sensor information generator 2-1, ..., 2-N is a natural number 1, ..., The reciprocal of 1 μsec with a total signal bandwidth of 1 MHz. , N times the time 1, ..., Even if it is equal to Nμsec, the above-mentioned effect cannot be obtained. That is, the cross-correlation y _B (n) between the FMCW interrogation wave and the FMCW reception wave is the time of each delay applied by each sensor information generator 2-1 ..., 2-N 1, ..., At Nμsec, it is also affected by the contribution from the FMCW wraparound wave. Then, in the sensor information acquisition device 1, when there is a wraparound between the question wave transmitting unit 11 of the FMCW question wave and the response wave receiving unit 12, each sensor information generating device 2-1 ..., 2-N From each FMCW response wave received from, each sensor information generated by each sensor information generator 2-1, ..., 2-N cannot be extracted with high accuracy.

(Calculation method of time correlation of the present disclosure and calculation result of autocorrelation of FMCW question wave)
In the present disclosure, the correlation function in the time domain is calculated in the bandwidth flat around the center of the band, excluding the bandwidth attenuated at the signal band edge in the spectrum of the FMCW interrogation wave.

As shown in Equation 5, the FFT calculation unit 131 Fourier transforms the FMCW interrogation wave u (n) in the time domain into the FMCW interrogation wave U (k) in the frequency domain. Here, n represents a time normalized using a sampling period. As shown in FIGS. 11 and 13, the band center cutout portion 132 is a band that is flat around the band center from the FMCW interrogation wave U (k) in the frequency domain, except for the bandwidth that is attenuated at the signal band end. in width, cut out FMCW interrogation wave _U C (k) in the frequency domain. Complex conjugate unit 133, a FMCW interrogation wave _U C (k) in the frequency domain, to the complex conjugate of the FMCW interrogation wave _U ^C * (k) in the frequency domain.

The FMCW received wave x (n) in the time domain is a superposition of each FMCW response wave v (n) and the FMCW wraparound wave w (n) in the time domain. As shown in Equation 6, the FFT calculation unit 134 sets the FMCW received wave x (n) = v (n) + w (n) in the time domain and the FMCW received wave X (k) = V (k) in the frequency domain. ) + W (k) Fourier transform. Band center cutting unit 135, the FMCW reception wave X in the frequency domain (k) = V (k) + W (k), FMCW interrogation wave band center cutting unit 132 in the frequency domain _U C a (k) in the same bandwidth and the bandwidth for cutting, cutting out the FMCW reception wave in the frequency domain _{X C (k) = V C} (k) + W C (k).

Spectral multiplication unit 136, as shown in Expression 7, the FMCW query wave _U ^C * (k) in the frequency domain, FMCW reception wave in the frequency domain _{X C (k) = V C} (k) + W C (k ) And, and the mutual correlation Y _C (k) in the frequency domain is calculated. As shown in Equation 8, the IFFT calculation unit 137 inverse-Fourier transforms the cross-correlation Y _C (k) in the frequency domain into the cross-correlation y _C (n) in the time domain. The cross-correlation y _C (n) in the time domain includes y _{C, v} (n) and y _{C, w} (n). y _{C, v} (n) represent the cross-correlation between the FMCW interrogation wave and each FMCW response wave in the time domain. y _{C, w} (n) represents the cross-correlation between the FMCW interrogation wave and the FMCW wraparound wave in the time domain, and is proportional to the autocorrelation of the FMCW interrogation wave in the time domain.

Here, we want to calculate the cross-correlation y _{C, v} (n) in the time domain between the FMCW inquiry wave and each FMCW response wave, and extract the delay time and sensor information of each FMCW response wave. Then, it is necessary to reduce the cross-correlation y _{C, w} (n) in the time domain between the FMCW interrogation wave and the FMCW wraparound wave. That is, it is necessary to reduce the autocorrelation of the FMCW interrogation wave in the time domain.

The first method of calculating the time correlation of the present disclosure is shown in FIG. The vertical axis of FIG. 11 shows the power level of the spectrum of the FMCW interrogation wave in logarithmic display dB. The power level of the spectrum of the FMCW interrogation wave is proportional to the U (k) described in Equation 5 above. The bandwidth used to calculate the time correlation is 0.8 MHz, including the bandwidth that is flat around the center of the band, excluding the bandwidth that is attenuated at the signal band edge. The sampling frequency is 4 MHz.

The calculation result of the autocorrelation of the first FMCW interrogation wave of the present disclosure is shown in FIG. The vertical axis of FIG. 12 shows the power level of the autocorrelation of the first disclosed FMCW interrogation wave in logarithmic display dB. The autocorrelation power level of the first FMCW interrogation wave of the present disclosure is proportional to y _{C, w} (n) described in Equation 8 described above. As described above, the autocorrelation of the FMCW interrogation wave is a natural number 1, 2, ..., N times the time 1.25, 2.50, ..., With a flat bandwidth of 0.8 MHz and a reciprocal of 1.25 μsec. It becomes almost 0 at 1.25 Nμsec. That is, the cross-correlation y _{C, w} (n) between the FMCW interrogation wave and the FMCW wraparound wave is a natural number 1, 2, ..., N times the time of the reciprocal 1.25 μsec with a flat bandwidth of 0.8 MHz. At 1.25, 2.50, ..., 1.25 Nμsec, it becomes almost 0.

The second method of calculating the time correlation of the present disclosure is shown in FIG. The vertical axis of FIG. 13 indicates the power level of the spectrum of the FMCW interrogation wave in logarithmic display dB. The power level of the spectrum of the FMCW interrogation wave is proportional to the U (k) described in Equation 5 above. The bandwidth used to calculate the time correlation is 0.8 MHz, including the bandwidth that is flat around the center of the band, excluding the bandwidth that is attenuated at the signal band edge. The sampling frequency is also 0.8 MHz.

FIG. 14 shows the calculation result of the autocorrelation of the second FMCW interrogation wave of the present disclosure. The vertical axis of FIG. 14 shows the power level of the autocorrelation of the second FMCW interrogation wave of the present disclosure in logarithmic display dB. The power level of the autocorrelation of the second FMCW interrogation wave of the present disclosure is proportional to y _{C, w} (n) described in Equation 8 described above. As described above, the autocorrelation of the FMCW interrogation wave is a natural number 1, 2, ..., N times the time 1.25, 2.50, ..., With a flat bandwidth of 0.8 MHz and a reciprocal of 1.25 μsec. It becomes almost 0 at 1.25 Nμsec. That is, the cross-correlation y _{C, w} (n) between the FMCW interrogation wave and the FMCW wraparound wave is a natural number 1, 2, ..., N times the time of the reciprocal 1.25 μsec with a flat bandwidth of 0.8 MHz. At 1.25, 2.50, ..., 1.25 Nμsec, it becomes almost 0.

Therefore, the delay times T ₁ , T ₂ , ..., _TN applied by each sensor information generator 2-1, 2-2, ..., 2-N are set to a flat bandwidth of 0.8 MHz. The above effect can be obtained by equalizing the reciprocal 1.25 μsec to the natural numbers 1, 2, ..., N times the time 1.25, 2.50, ..., 1.25 Nμsec. That is, the cross-correlation y _C (n) between the FMCW interrogation wave and the FMCW reception wave is the time of each delay applied by each sensor information generator 2-1, 2-2, ..., 2-N. At 25, 2.50, ..., 1.25 Nμsec, it is not affected by the contribution from the FMCW wraparound wave. Then, in the sensor information acquisition device 1, even if there is a wraparound between the question wave transmitting unit 11 of the FMCW question wave and the response wave receiving unit 12, each sensor information generating device 2-1 ..., 2- From each FMCW response wave received from N, each sensor information generated by each sensor information generator 2-1, ..., 2-N can be extracted with high accuracy.

The flat bandwidth of 0.8 MHz is narrower than the total signal bandwidth of 1 MHz. Therefore, the resolution of each delay time T ₁ , ..., _TN applied by each sensor information generator 2-1, ..., 2-N seems to be apparently deteriorated in the present disclosure. (1 μsec in the comparative example → 1.25 μsec in the present disclosure). However, the effect of contributions from FMCW wraparound waves is largely non-existent in this disclosure. Therefore, the resolution of each delay time T ₁ , ..., _TN applied by each sensor information generator 2-1, ..., 2-N can be substantially improved in the present disclosure.

(Calculation result of cross-correlation between FMCW inquiry wave and FMCW received wave of the present disclosure)
The spectrum of the FMCW received wave of the present disclosure is shown in FIG. The vertical axis of FIG. 15 indicates the power level of the spectrum of the FMCW received wave in logarithmic display dB. The power level of the spectrum of the FMCW received wave is proportional to X (k) according to Equation 6 described above. The FMCW received wave x (n) in the time domain is a superposition of each FMCW response wave v (n) and the FMCW wraparound wave w (n) in the time domain. As each FMCW response wave v (n) in the time domain, only the FMCW response wave to which the sensor information generator 2-3 has been subjected to a delay time T ₃ = 3.75 μsec is taken up.

The calculation result of the cross-correlation between the FMCW interrogation wave and the FMCW reception wave of the present disclosure is shown in FIG. The vertical axis of FIG. 16 shows the power level of the cross-correlation between the FMCW interrogation wave and the FMCW reception wave of the present disclosure in logarithmic display dB. The power level of the cross-correlation between the FMCW interrogation wave and the FMCW received wave of the present disclosure is proportional to y _C (n) = y _{C, v} (n) + y _{C, w} (n) described in Equation 8 described above. To do. The bandwidth used to calculate the time correlation is 0.8 MHz, excluding the bandwidth attenuated at the signal bandwidth edge in the FMCW interrogation wave, and including the bandwidth flat around the band center in the FMCW interrogation wave. The sampling frequency is also 0.8 MHz.

The peak detection unit 141 detects the peak position of the amplitude | y _C (n) | of the cross-correlation y _C (n) between the FMCW interrogation wave and the FMCW reception wave. In general, the peak position of | y _C (n) | is the normalized time n = n ₁ , ... n _N corresponding to the time T = T ₁ , ... _TN . In FIG. 16, the peak position of | y _C (n) | is the normalized time n = n ₃ corresponding to the time T = T ₃ . However, as the peak position of | y _C (n) |, the peak position of the autocorrelation amplitude of the FMCW interrogation wave as shown in FIG. 12 should not be adopted, that is, the time _TN-1 <T < T normalized corresponding to the _n time _n n-1 _<n <should not adopt _{n n.}

Next, the amplitude / phase calculation unit 142 sets the normalization time n = other than 0 at which the amplitude | y _C (n) | of the mutual correlation y _C (n) between the FMCW interrogation wave and the FMCW received wave peaks. At n ₁ , ... n _N , the amplitude | y _C (n) | and / or the phase θ _C (n) of the mutual correlation y _C (n) between the FMCW inquiry wave and the FMCW received wave is calculated. Alternatively, the amplitude / phase calculation unit 142 has a normalization time n = n ₁ , ... n _N , which is a natural number multiple of the inverse of the bandwidth in which the spectrum U (k) of the FMCW interrogation wave is flat around the center of the bandwidth. , The amplitude | y _C (n) | and / or the phase θ _C (n) of the mutual correlation y _C (n) between the FMCW interrogation wave and the FMCW received wave is calculated.

The vibration information extraction unit 143 has an amplitude | y _C (n) | and / or a phase θ _C (n) (n = n ₁ ,) of the mutual correlation y _C (n) between the FMCW inquiry wave and the FMCW received wave. ... Based on n _N ), each sensor information s (n) (n = n) used for each amplitude modulation and / or each phase modulation from each FMCW response wave excluding the FMCW wraparound wave among the FMCW received waves. ₁ , ... n _N ) is extracted.

Then, the vibration information extraction unit 143 extracts each sensor information s (n) (n = n ₁ , ... n _N ) extracted a single time from each sensor information generator 2-1 ... It is extracted as information about the acceleration directly under each lane of the bridge B where −N is installed. Further, the vibration information extraction unit 143 outputs the FFT results for each sensor information s (n) (n = n ₁ , ... n _N ) extracted a plurality of times, respectively, to each sensor information generator 2-1. -Extract as information about the vibration directly under each lane of the bridge B where 2-N is installed. Information on acceleration and vibration directly under each lane of the bridge B on which each sensor information generator 2-1, ..., 2-N is installed can be obtained directly or via the Internet I or the like. Will be notified.

Here, in the sensor information acquisition device 1, when the amplitude of each FMCW response wave is demolished, the amplitude of each FMCW response wave is spatially propagated between each sensor information generation device 2-1 ..., 2-N. It is desirable to distinguish whether it is attenuated by or generated by each sensor information in each sensor information generator 2-1 ..., 2-N. Therefore, it is desirable that each sensor information generator 2-1, ..., 2-N not only has the configuration shown in FIG. 3 but also has a reference element unrelated to each sensor information. Then, it is desirable that the sensor information acquisition device 1 measures the difference between each FMCW response wave from the configuration shown in FIG. 2 and a reference signal from a reference element unrelated to each sensor information.

The sensor information acquisition device, sensor information generation device, and sensor information generation / acquisition system of the present disclosure can be applied to applications such as inspection of structures such as bridges in which sensor information is wirelessly communicated from a responder to an interrogator. ..

S: Sensor information generation / acquisition system B: Bridge I: Internet 1: Sensor information acquisition device 2-1, 2-2, 2-3, 2-N: Sensor information generation device 11: Question wave transmitter 12: Response wave Reception unit 13: Cross-correlation calculation unit 14: Sensor information extraction unit 21-1, 21-N: Question wave receiver 22-1, 22-N: Question wave modulation unit 23-1, 23-N: Question wave delay unit 24-1, 24-N: Response wave transmission unit 131: FFT calculation unit 132: Band center cutout unit 133: Complex conjugate unit 134: FFT calculation unit 135: Band center cutout unit 136: Spectrum multiplication unit 137: Amplitude calculation Unit 141: Peak detection unit 142: Amplitude / phase calculation unit 143: Vibration and other information extraction unit

Claims (3)

A sensor information acquisition device that acquires sensor information from a sensor information generator.
A band-restricted FMCW (Frequency Modulated Continuous Wave) interrogation wave is transmitted to the sensor information generator, and a interrogation wave transmitter.
The FMCW response wave applied to the FMCW interrogation wave is modulated by the sensor information and delayed by a natural number several times the inverse of the bandwidth in which the spectrum of the FMCW interrogation wave is flat around the center of the bandwidth. The response wave receiver received from the sensor information generator and
Calculate the cross-correlation in the time region between the FMCW interrogation wave and the FMCW received wave including the FMCW response wave and the FMCW wraparound wave in a bandwidth in which the spectrum of the FMCW interrogation wave is flat around the center of the band. Cross-correlation calculation unit and
At times other than 0 when the amplitude of the intercorrelation peaks in the time region, and / or at times of natural multiples of the inverse of the bandwidth where the spectrum of the FMCW interrogation wave is flat around the center of the bandwidth. By calculating the amplitude and / or phase of the mutual correlation in the time region, the sensor information used for amplitude modulation and / or phase modulation from the FMCW response wave excluding the FMCW wraparound wave among the FMCW received waves. Sensor information extraction unit that extracts
A sensor information acquisition device characterized by comprising. A sensor information generator that provides sensor information to a sensor information acquisition device.
A band-restricted FMCW (Frequency Modulated Continuous Wave) interrogation wave is received from the sensor information acquisition device, and a interrogation wave receiver.
A interrogation wave modulator that applies modulation based on the sensor information to the FMCW interrogation wave,
The intercorrelation in the time region between the FMCW interrogation wave and the FMCW received wave including the FMCW response wave and the FMCW wraparound wave in the bandwidth where the spectrum of the FMCW interrogation wave is flat around the center of the band is described. The natural inverse of the bandwidth in which the spectrum of the FMCW interrogation wave is flat around the center of the band so that the delay time applied by the sensor information generator to the FMCW interrogation wave is not affected by the contribution from the FMCW wraparound wave. A question wave delay section that applies a delay of several times the time to the FMCW question wave, and
And modulation by the sensor information, and the FMCW interrogation wave spectrum is only natural number times the reciprocal of the bandwidth is flat around band center delay, the said FMCW interrogation waves to the FMCW response wave which has been subjected , The response wave transmitter that transmits to the sensor information acquisition device,
A sensor information generator characterized by the present invention. The sensor information acquisition device according to claim 1 and
The sensor information generator according to claim 2 and
A sensor information generation / acquisition system characterized by being equipped with.

Images