Presence Detection System

Abstract

To provide an existence detection system that improves detection accuracy by a steep frequency characteristic.SOLUTION: An existence detection system 100 comprises: a transmission unit 30 that transmits a transmission wave serving as a radio wave; a reception unit 40 that detects a response wave to be output from a resonance tag TG; and a signal processing unit 80 that determines existence of an object on the basis of the response wave. The signal processing unit 80 is configured to: acquire, over a plurality of times, a series of detection data corresponding to the response wave to the transmission wave transmitted while gradually changing a frequency step by step; perform a fast Fourier transformation to the series of detection data by digital processing; sequentially perform smoothing by a weighting addition in consideration for past inclusion with respect to an obtained spectrum distribution; and, when spectrum intensity exceeds a given threshold, determine that the object is detected.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a presence detection system for detecting the presence of a resonant tag attached to a product or the like, and more particularly to a presence detection system incorporating a digital data processing circuit. [Background technology]

[0002] A known conventional presence detection system of this type includes a transmitter, a receiver, a mixer, and a computer (Patent Document 1). In this type of system, the mixer multiplies the output from the receiver and the output from the receiver, A / D converts the multiplied signal, and the A / D converted data is processed by the computer.

[0003] However, increasing the sensitivity of the above device also amplifies noise, and even if a filter is incorporated into the hardware after the mixer, it is not possible to achieve a steep frequency characteristic, and detection accuracy cannot be sufficiently improved. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Special Publication No. 2003-533143 Summary of the Invention

[0005] The present invention has been made in view of the above-mentioned background art, and has as its object to provide a presence detection system that improves detection accuracy by using a steep frequency characteristic.

[0006] To achieve the above purpose of The first presence detection system includes a transmitter that transmits a transmission wave, which is an electric wave, a receiver that detects a response wave output from a resonant tag, and a signal processor that determines whether or not an object is present based on the response wave, and the signal processor acquires a series of detection data corresponding to the response wave in response to the transmitted transmission wave multiple times while gradually changing the frequency in stages; Obtained at each timeA series of detected data is digitally processed using fast Fourier transform, Each time a spectral distribution is obtained by fast Fourier transform, smoothing is performed in sequence. During smoothing, weighted addition is performed on the smoothed results up to that point and the added spectral distribution using the smoothing coefficient corresponding to the number of smoothing operations up to that point and the smoothing coefficient corresponding to the number of smoothing operations including this one. The resulting spectrum is If the intensity of the signal exceeds a predetermined threshold, it is determined that an object has been detected.

[0007] According to the above-described presence detection system, the signal processing unit performs a fast Fourier transform on a series of detection data by digital processing, and then performs smoothing on the obtained spectral distribution by weighted addition taking into account the degree of inclusion. Therefore, not only can the frequency characteristics of detection be made highly sensitive by digital processing and data aggregation, but processing speed can be increased while reducing the amount of memory required for processing.

[0008] According to a specific aspect of the present invention, in the presence detection system, the signal processing unit smoothes a group of data after fast Fourier transform that matches the frequency of the transmission wave among a series of detection data acquired multiple times. In this case, smoothing is performed on a group of data acquired for a specific frequency, making it easier to obtain a determination result that suppresses noise and mitigates fluctuations over time.

[0009] According to another aspect of the present invention, the signal processing unit adds, for each frequency, a series of spectral distributions obtained from a series of detection data corresponding to a transmission wave transmitted while changing the frequency. In this case, even if the frequency of the transmission wave changes, the frequency characteristics of the response of a specific resonant tag attached to an object will be the same, so that sensitivity can be improved by adding them.

[0010] According to another aspect of the present invention, the signal processing unit excludes components of the spectral distribution that are out of the frequency range of the transmitted wave. In this case, calculations for frequency components outside the range expected as the frequency characteristics of the resonant tag attached to the target can be omitted, and processing can be accelerated by focusing on processing related to ensuring sensitivity.

[0011] To achieve the above purpose ofThe second presence detection system includes a transmitter that transmits radio waves, a receiver that detects response waves output from the resonant tag, and a signal processor that determines whether or not an object is present based on the response waves. The signal processor generates a series of detection data corresponding to the response waves in response to the transmitted transmission waves while gradually changing the frequency in stages. multiple times Get Obtained at each time A series of detected data is digitally processed using fast Fourier transform, By fast Fourier transform The obtained series of spectral distributions are summed for each frequency. If the intensity of the spectrum after addition exceeds a predetermined threshold, it is determined that the object has been detected. .

[0012] According to the above-mentioned presence detection system, the signal processing unit performs a fast Fourier transform on a series of detection data by digital processing, and adds up a series of spectral distributions obtained from the series of detection data for each frequency. This not only makes it possible to increase the sensitivity of the detection frequency characteristics by digital processing, but also makes it possible to achieve improved sensitivity by adding up, since the frequency characteristics of the response from a specific resonant tag attached to the target remain the same even if the frequency of the transmitted wave changes. [Brief explanation of the drawings]

[0013] [Figure 1] FIG. 1 is a block diagram illustrating a presence detection system according to a first embodiment. [Figure 2] 2 is a diagram illustrating a state in which the presence detection system of FIG. 1 is used. FIG. [Figure 3] 10A and 10B are diagrams illustrating an antenna that constitutes a detection unit. [Figure 4] 4 is a flowchart illustrating the operation of the presence detection system. [Figure 5] 1A is a diagram showing the output timing of a transmission wave, and FIG. 1B is a diagram showing an enlarged view of the time before and after one transmission wave output. [Figure 6] (A) shows an example of the measured spectral distribution and the prior spectral distribution, and (B) shows the difference spectral distribution. [Figure 7] (A) to (I) show the FFT intensity distribution of the received waveform for a specific frequency for each cycle. [Figure 8] (A) is the same as the FFT intensity distribution in Figure 7(G), (B) is the spectral distribution of the moving average up to this cycle, (C) is the spectral distribution after repeated smoothing processing up to this cycle, and (D) is the spectral distribution after smoothing processing compared with the spectral distribution of the moving average. [Figure 9] (A) is the same as the FFT intensity distribution in Figure 7(H), (B) is the spectral distribution of the moving average up to this cycle, (C) is the spectral distribution after repeated smoothing up to this cycle, and (D) is the spectral distribution after smoothing compared with the spectral distribution of the moving average. [Figure 10] (A) is the same as the FFT intensity distribution in Figure 7(I), (B) is the spectral distribution of the moving average up to this cycle, (C) is the spectral distribution after repeated smoothing up to this cycle, and (D) is the spectral distribution after smoothing compared with the spectral distribution of the moving average. [Figure 11] 10 is a flowchart illustrating the operation of the presence detection system according to the second embodiment. [Figure 12] 10A and 10B are diagrams conceptually illustrating the addition of spectral distributions. [Figure 13] (A) to (F) are charts showing FTT intensities before addition in specific examples. [Figure 14] (A) shows the result after adding the FTT intensity, and (B) shows a comparative example without adding the FTT intensity. DETAILED DESCRIPTION OF THE INVENTION

[0014] [First embodiment] Hereinafter, a presence detection system according to a first embodiment of the present invention will be described with reference to FIG.

[0015] The presence detection system 100 comprises a transmitting and receiving antenna 10, a transmitting unit 30 that transmits transmission waves via the antenna 10, a receiving unit 40 that detects reply waves via the antenna 10, a signal processing unit 80 that determines the presence or absence of an object based on the reply waves, and an alarm device 50. The presence detection system 100 transmits multiple pulsed transmission waves accompanied by frequency sweeps to a monitoring zone, receives reply waves from the monitoring zone, and extracts a signal reflecting the electromagnetic characteristics of a resonant tag that is the monitoring object. The resonant tag is attached to an object such as a product, and the presence detection system 100 determines whether an object with the resonant tag attached is present in the monitoring zone.

[0016] The signal processing unit 80 includes a computer 81 that controls the overall operation of the presence detection system 100, and a digital signal processing circuit 83 that performs fast Fourier transform and associated processing at high speed. The digital signal processing circuit 83 can also be realized as a function of the computer 81 rather than being independent from the computer 81.

[0017] The transmitter 30 includes a direct digital synthesizer (hereinafter also referred to as DDS) 31, a switch 32, and an amplifier circuit 33. The DDS 31 is a known circuit that digitally generates any analog wave, and sets the frequency and phase of the transmitted wave to a state suitable for detecting a resonant tag according to instructions from a computer 81 in the signal processor 80. The switch 32 operates under the control of the computer 81 and functions as a gate for the output of the transmitted wave, and the amplifier circuit 33 is an analog circuit that amplifies the intensity of the transmitted wave to an intensity suitable for the antenna 10, resonant tag, etc.

[0018] The receiving unit 40 includes a switch 41, a filter 42, and an AD conversion circuit 43. The switch 41 operates under the control of the computer 81 and functions as a gate for inputting received waves. The filter 42 passes received waves in a frequency band expected for the resonant tag to be monitored. Specifically, if the resonant tag is intended for use in the 8.2 MHz band, a bandpass filter of 7.5 MHz to 10 MHz, for example, is used as the filter 42. The AD conversion circuit 43 operates at a processing speed, resolution, and accuracy that do not interfere with signal processing in the digital signal processing circuit 83.

[0019] In the signal processing unit 80, the computer 81 is, for example, a microcomputer, and has a CPU, memory, etc., and operates based on an installed program. The digital signal processing circuit 83 is a digital circuit that can freely modify the processing content, and has a small memory capacity. The digital signal processing circuit 83 is realized by, for example, an FPGA (field-programmable gate array). Note that the digital signal processing circuit 83 may be integrated with the AD conversion circuit 43 of the receiving unit 40.

[0020] The alarm device 50 includes a speaker and a lamp, and when the computer 81 determines that a resonant tag TG is present, receives a command from the computer 81 and performs an alarm process regarding the detection of the resonant tag TG.

[0021] FIG. 2 is a conceptual diagram showing an application example of the presence detection system 100. In this case, the presence detection system 100 includes a detection unit 1a disposed adjacent to a passage PW, and the passage PW, which is the area near the detection unit 1a, is the monitored zone. The detection unit 1a includes an antenna 10 and a circuit body 100a. The circuit body 100a includes the transmitter 30, receiver 40, and signal processor 80 shown in FIG. 1. The detection unit 1a transmits a transmission wave W1 and detects a response wave W2. Note that multiple detection units similar to the detection unit 1a can be disposed along or across the passage PW.

[0022] As shown in Figures 3(A) and 3(B), in a single detection unit 1a, the antenna 10 includes a first antenna 11 consisting of three antenna loops 11a, 11b, and 11c, and a second antenna 12 consisting of two antenna loops 12a and 12b. The first antenna 11 and the second antenna 12 are driven by a circuit body 100a shown in Figure 2 and other figures, and perform transmission and reception operations at alternating timings to avoid interference. The first antenna 11 has high detection performance for resonant tags TG placed along a horizontal plane at high and low positions, and high detection performance for resonant tags TG placed along a vertical plane at an intermediate position. Furthermore, the second antenna 12 has high detection performance for resonant tags TG placed along a vertical plane at high and low positions, and high detection performance for resonant tags TG placed along a horizontal plane at an intermediate position.

[0023] Returning to Figure 2, the first antenna 11 and the second antenna 12 that make up the antenna 10 are connected to the circuit body 100a, and the first antenna 11 and the second antenna 12 operate alternately under the control of the circuit body 100a to detect the resonant tag TG moving along the passage PW as the resonant tag TG crosses the monitored zone.

[0024] An overview of the operation of the presence detection system 100 will be described with reference to FIG. 4. First, the computer 81 sets an initial frequency value for the swept frequency band (step S11). At this time, the first antenna 11 is set to operate initially. Then, the computer 81 detects background noise immediately before transmission (step S12). At this time, the computer 81 operates the receiver 40 and the digital signal processing circuit 83 to perform fast Fourier transform processing on the received signal and acquire a spectral distribution associating frequency with intensity. The spectral distribution thus acquired is temporarily stored in the digital signal processing circuit 83 as a preliminary spectral distribution. The fast Fourier transform processing is performed using a known technique. The computer 81 excludes components outside the frequency band of the transmitted wave from the spectral distribution obtained by fast Fourier transform. In this case, calculations for frequency components outside the range expected as the frequency characteristics of the resonant tag TG can be omitted, thereby enabling faster processing by focusing on processing related to ensuring sensitivity.

[0025] Immediately after the operation of acquiring the preliminary spectral distribution, or a predetermined time thereafter, the computer 81 operates the DDS 31 to generate a transmission signal at the initial frequency value set in step S11. This causes a transmission wave to be output from the first antenna 11 at the initial frequency value, and a response wave is output from the resonant tag TG that receives the transmission wave (step S13). After or in parallel with the generation of the transmission signal, the computer 81 operates the receiver 40 and the digital signal processing circuit 83 to detect the received wave, temporarily stores a received signal corresponding to the received wave in memory, and performs fast Fourier transform processing on the received signal to acquire a measured spectral distribution that associates frequency and intensity (step S13). The computer 81 then calculates the difference between the measured spectral distribution acquired in step S13 and the preliminary spectral distribution acquired in step S12 (hereinafter also referred to as the difference spectral distribution), and smooths the data by averaging the obtained difference spectral distribution with the previous averaged spectral distribution obtained from the measured spectral distribution already detected at the same transmission frequency (step S14). The new averaged spectral distribution obtained by averaging is temporarily stored in the memory of the digital signal processing circuit 83. Here, averaging is performed not between measured spectral distributions with gradually increasing transmission frequencies, but between measured spectral distributions with the same transmission frequency. This corresponds to smoothing the detection signal of the resonant tag TG, which gradually shifts as the detection cycle is repeated. Note that the first time a measured spectral distribution is acquired at each transmission frequency, there is no previously detected measured spectral distribution, so the difference spectral distribution between the measured spectral distribution and the previous spectral distribution is used as the new averaged spectral distribution. The computer 81 then checks whether the frequency sweep is complete (step S15). If the frequency sweep is not complete (No in step S15), the computer 81 updates the frequency by increasing it by a predetermined amount Δ and sets the frequency to be measured next (step S16).Thereafter, the process returns to step S12 to detect background noise, and a measured spectral distribution is obtained by detecting the received wave while outputting the transmitted wave (step S13), a differential spectral distribution between the measured spectral distribution and the prior spectral distribution is calculated, and the obtained differential spectral distribution is averaged with a previous averaged spectral distribution obtained from the measured spectral distribution already detected at the same transmission frequency to smooth the data (step S14). In the above, the received signals obtained from the received wave obtained while gradually changing the frequency of the transmitted wave in stages are a series of detection data.

[0026] If the frequency sweep is completed (Yes in step S15), the computer 81 determines whether a series of detection data has been acquired by the frequency sweep using both the first and second antennas 11 and 12, that is, whether the detection by the second antenna 12, which is performed after the detection by the first antenna 11, has been completed (step S18). If the detection by the second antenna 12 has not been completed (No in step S18), the computer 81 switches from the first antenna 11 to the second antenna 12, returns to step S11, and repeats the processes of steps S12 to S16.

[0027] If detection by the second antenna 12 is complete (Yes in step S18), the computer 81 determines whether the process of obtaining a new averaged spectral distribution while sweeping the frequency has been performed a specified number of times for the first and second antennas 11 and 12 (step S19). If the process has not been performed the specified number of times (No in step S19), the computer 81 returns to step S11 and repeats the processes of steps S12 to S16. Detecting received waves while outputting transmitted waves while increasing the frequency by an amount of change Δ until the frequency sweep is complete means acquiring a series of detection data corresponding to response waves to transmitted transmitted waves while gradually changing the frequency in stages, over multiple times. This detection operation is called a detection cycle. Calculating a measured spectral distribution or a differential spectral distribution from received signals obtained by detecting received waves while gradually increasing the frequency of the transmitted waves corresponds to performing a fast Fourier transform on the series of detection data by digital processing. Furthermore, smoothing the data by averaging the difference spectral distribution with the old averaged spectral distribution obtained from the already detected measured spectral distribution is equivalent to sequentially smoothing a group of spectral distributions obtained from the received signal (specifically, a group of data after fast Fourier transform that matches the frequency of the transmitted wave) by weighting and adding them taking into account the degree of inclusion.

[0028] If the process of obtaining a new averaged spectral distribution has been repeated a predetermined number of times (Yes in step S19), the computer 81 analyzes the new averaged spectral distribution obtained in step S14 (hereinafter, the final one is also referred to as the averaged spectral distribution) (step S20). Here, the averaged spectral distribution is obtained by smoothing the temporal change in intensity in the spectral distribution obtained at each swept frequency a predetermined number of times (i.e., the number of cycles), thereby reducing noise. In analyzing the averaged spectral distribution, specifically, the computer 81 determines whether or not a peak exists in the averaged spectral distribution obtained at a predetermined number of steps at different frequencies. If a peak exists, the computer 81 detects the frequency value and peak intensity value of the peak. If the frequency values ​​of the intensity peaks match and the peak intensity values ​​exceed a predetermined threshold or more, the computer 81 determines that the spectral intensity exceeds the predetermined threshold, determines that the resonant tag TG is present, and determines that the target OB to which the resonant tag TG is attached has been detected.

[0029] If it is determined that the resonant tag TG does not exist (No in step S21), the computer 81 returns to step S11 and repeats the processes of steps S12 to S16, etc. On the other hand, if it is determined that the resonant tag TG exists (Yes in step S21), the computer 81 activates the alarm device 50 to issue an alarm to notify the surrounding area of ​​the presence of the resonant tag TG by sound or the like, or reports the presence of the resonant tag TG to a higher-level management device (step S22).

[0030] Figure 5 is a diagram for explaining specific operation timing. Figure 5(A) shows the output timing of the transmission wave, and for example, a detection cycle is repeated in which 25 bursts of transmission wave are output from the first antenna 11 of the detection unit 1a while the frequency is gradually increased. The frequency f k =f0~f 24 can be set to a desired value by the computer 81, and is set to, for example, 7.80 MHz to 8.56 MHz. In a specific embodiment, the center frequency f cis set to, for example, 8.0 MHz to 10.5 MHz. w can be set to a desired value by the computer 81, for example, ±400 kHz, ±500 kHz, etc. In other words, the center frequency f c = 8.0 MHz, the frequency of the transmission wave will be 8.0 ± 0.4 MHz or 8.0 ± 0.5 MHz. Furthermore, the frequency of the detection cycle is called the synchronization frequency fs, and can be set to a desired value by the computer 81, for example, 84 to 129 Hz. In the above, the transmission wave output from the first antenna 11 of the detection unit 1a and the transmission wave output from the second antenna 12 of the detection unit 1a are output alternately with a shift in timing.

[0031] As shown in Fig. 5(B), input and sampling of the received wave are performed before and after each output of the transmitted wave. The sampling performed before the output of the transmitted wave obtains the preliminary spectral distribution (see step S12 in Fig. 4), and the sampling performed after the output of the transmitted wave obtains the measured spectral distribution (see step S13 in Fig. 4).

[0032] Figure 6(A) shows an example of a measured spectral distribution and an example of a preliminary spectral distribution. Figure 6(B) shows the difference spectral distribution ΔFFT obtained by subtracting the preliminary spectral distribution from the measured spectral distribution. The difference spectral distribution has a large intensity peak in one location.

[0033] The specific calculation method is explained below. In the transmission operation, the following 25 transmission frequencies f i Performs sending and receiving processing. f i =f c -(2f w / step)*i … (1) i=0~(step-1),step=25 where: f c is the center frequency fw is the bandwidth (one side) step is the number of frequencies swept in one cycle is. Each transmission frequency f i The ADC sample data (d i (t,c)), apply a fast Fourier transform (FFT) to each, and take the difference of the absolute values ​​of the results. iafter (t,c)(FFT) → F iafter (f,c) t=0,1 / (N f smp ),2 / (N f smp ), …,(N-1) / (N f smp ) … (2.1) d ibefore (t,c)(FFT) → F ibefore (f,c) t=0,1 / (N f smp ),2 / (N f smp ), …,(N-1) / (N f smp ) … (2.2) y i =|F iafter (f,c)|-|F ibefore (f,c)| … (3) where: F iafter (f,c) is the FFT data before transmission (including real and imaginary parts) F ibefore (f,c) is the FFT data after transmission (including real and imaginary parts) N is the number of FFT stages f smp is the sampling frequency c is the cycle number is. The difference y above i To reduce the memory capacity required, a smoothing process is performed on the difference spectrum distribution (i.e., the difference spectrum distribution). The smoothing process is performed by setting a smoothing coefficient A (A-1 corresponds to the degree of inclusion) as shown in the following equation. TIFF0007764184000001.tif12168The smoothing target y(f,c) is the difference y iFurthermore, D(f,c-1) is the spectral distribution obtained by smoothing up to the previous cycle (c-1th). This method means that only a one-stage buffer is required to store FFT data, regardless of the number of cycles. Furthermore, the FFT frequency characteristic data (i.e., the differential spectral distribution) only needs to use the vicinity of the center frequency (specifically, 8.2 MHz), which allows for further reduction in memory resources. As shown in equation (2.2), the frequency resolution of the FFT is f smp = 50MHz and N = 512, f smp / N=97.7 kHz. If the center frequency is, for example, 8.2 MHz, and only the region from 7.7 MHz to 8.7 MHz is analyzed and the outside of that is excluded, the number of data points or time data will be 11 points. In other words, the buffer size required for smoothing processing is, assuming that the unit memory built into the digital signal processing circuit 83 is 4 bytes, there are two antennas, and the number of frequency steps or stages is 25, 4(Byte)×2×25×11=2.15(kByte) Thus, the buffer size can be compressed. Although not explained further, if memory is allocated for the number of cycle repetitions, memory at least times the number of repetitions (for example, 32 times) is required, and a large buffer size must be allocated. Furthermore, in the FPGA of the embodiment used as the digital signal processing circuit 83, the number of FFT stages (number of time data) is 512 as standard, and by analyzing only 7.7 MHz to 8.7 MHz, the memory usage can be compressed to about 1 / 100.

[0034] 7(A) to 7(I) show the transmission frequency f 12 The measurement conditions are that the resonant tag TG is 5 cm square, the resonant frequency of the resonant tag TG is 8.2 MHz, the distance from the gate of the resonant tag TG, i.e., the detection unit 1a, is 150 cm, and the transmission frequency is f0 to f24 = 7.8 MHz to 8.52 MHz (in 30 kHz increments). 12The results for Hz = 8.16 MHz were picked up. The synchronization frequency was 90 Hz, and 1 cycle = 11.1 ms. It can be seen that the position and magnitude of the FFT peaks for each cycle in Figures 7(A) to 7(I) vary greatly due to the effects of noise, etc.

[0035] Figure 8(A) is the same FFT intensity distribution as Figure 7(G) at cycle c=A-1, Figure 8(B) is the moving average spectral distribution up to cycle c=A-1 (memory is allocated for each cycle and the spectral distributions obtained in each cycle are averaged), and Figure 8(C) is the spectral distribution obtained by repeating the smoothing process up to cycle c=A-1. Figure 8(D) is a chart comparing the spectral distribution after smoothing shown in Figure 8(C) with the moving average spectral distribution shown in Figure 8(B). Both spectral distributions are nearly identical, indicating almost no difference in accuracy.

[0036] Figure 9(A) is the same as the FFT intensity distribution at c = A cycles in Figure 7(H), Figure 9(B) is the moving average spectral distribution up to c = A cycles, and Figure 9(C) is the spectral distribution after the smoothing process has been repeated up to c = A cycles. Figure 9(D) is a chart comparing the spectral distribution after smoothing shown in Figure 9(C) with the moving average spectral distribution shown in Figure 9(B).

[0037] Figure 10(A) is the same as the FFT intensity distribution at cycle c=A+1 in Figure 7(I), Figure 10(B) is the moving average spectral distribution up to cycle c=A+1, and Figure 10(C) is the spectral distribution after the smoothing process has been repeated up to cycle c=A+1. Figure 10(D) is a chart comparing the spectral distribution after smoothing shown in Figure 10(C) with the moving average spectral distribution shown in Figure 1(B).

[0038] According to the presence detection system 100 of the first embodiment described above, the signal processing unit 80 performs fast Fourier transform on a series of detection data by digital processing, and then performs smoothing on the obtained spectral distribution by weighted addition taking into account the degree of inclusion. Therefore, not only can the frequency characteristics of detection be made highly sensitive by digital processing and data aggregation, but also processing speed can be increased while reducing the amount of memory required for processing.

[0039] [Second embodiment] A presence detection system according to a second embodiment of the present invention will be described below. The presence detection system according to the second embodiment is a partial modification of the presence detection system according to the first embodiment, and a description of common features will be omitted.

[0040] The operation of the presence detection system 100 of the second embodiment will be described with reference to Fig. 11. In this case, before analyzing the averaged spectral distribution in step S20, a series of averaged spectral distributions corresponding to the transmitted wave are added for each frequency while changing the frequency (step S31). In this case, even if the frequency changes, an intensity peak appears at the position of the resonant frequency of the resonant tag TG. The height of the intensity peak increases as the frequency of the transmitted wave and the resonant frequency of the resonant tag TG become closer. However, by adding the averaged spectral distributions for the number of stages swept by the transmitted wave, the intensity peak is emphasized and noise is reduced.

[0041] In the above, we have described adding up the averaged spectral distribution for each frequency, but the intensity peaks are also emphasized if we add up the spectral distributions before averaging (for example, difference spectral distributions) for each frequency. It is also possible to obtain multiple such added spectral distributions over multiple detection cycles and average the multiple added spectral distributions over the number of detection cycles.

[0042] When adding the averaged spectral distribution or the differential spectral distribution for each frequency, it is possible to obtain an averaged value by dividing the added value by a countable number in addition to simply adding them.

[0043] Figure 12 is a diagram conceptually explaining the addition of spectral distributions. As shown in Figure 12(A), even if the intensity peaks in the individual spectral distributions are low, if the spectral distributions are added for the number of stages swept by the transmitted wave, the intensity peaks become higher as shown in Figure 12(B), and not only are the peaks emphasized, but noise is also reduced. The addition of the spectral distributions, that is, the added distribution y' of the FFT result, is TIFF0007764184000002.tif16170. Note that the y i (f, c) is not limited to the differential spectral distribution as described above, but may also be an averaged spectral distribution.

[0044] A specific measurement example will be described with reference to Figures 13(A) to 13(F). The measurement conditions were as follows: the resonant tag TG was 5 cm square, the resonant frequency of the resonant tag TG was 8.2 MHz, the distance from the gate of the resonant tag TG (i.e., the detection unit 1a) was 150 cm, the transmission frequency was f0 to f24 = 7.8 MHz to 8.52 MHz (30 kHz intervals), and the synchronization frequency was 90 Hz (1 cycle = 11.1 ms). In reality, transmission and reception were performed at 25 different frequencies, but only frequencies f10 to f15 with relatively high FFT intensities are shown, excluding those with low FFT intensities. Looking at the FFT results for each frequency, only frequency f12 has a discernible peak. When the resonant tag TG receives radio waves, it returns radio waves at its own resonant frequency, regardless of the frequency of the received radio waves.

[0045] Figure 14(A) shows the sum of the FFT results shown in Figures 13(A) to 13(F). It can be seen that the FFT intensity is emphasized at 8.2 MHz compared to Figures 13(A) to 13(F). In other words, if a resonant tag TG is present, adding the FFT results of each frequency will greatly emphasize the FFT intensity at the frequency position of the tag's resonant frequency.

[0046] Figure 14(B) shows data from a comparative example where the above-mentioned addition was not performed, and is the FFT result at f12 when the same tag was detected 110 cm away from the gate, i.e., detection unit 1a. The FFT intensity distribution in Figure 14(B) shows almost the same peak value as the added FFT intensity distribution in Figure 14(A). From this result, it can be said that under the above measurement conditions, peak detection performance improved by just under 40%.

[0047] According to the second embodiment of the presence detection system 100 described above, the signal processing unit 80 performs a fast Fourier transform on a series of detection data by digital processing, and adds up a series of spectral distributions obtained from the series of detection data for each frequency.This not only makes it possible to increase the sensitivity of the detection frequency characteristics by digital processing, but also makes it possible to achieve improved sensitivity by adding up, since the frequency characteristics of the response from a specific resonant tag attached to an object remain the same even if the frequency of the transmitted wave changes.

[0048] Although the present invention has been described above based on the embodiments, the present invention is not limited to the above embodiments. For example, although the difference spectral distribution is obtained by subtracting the prior spectral distribution from the measured spectral distribution, the measured spectral distribution can be used as is in the same way as the difference spectral distribution, or the prior spectral distribution can be obtained at the beginning of a detection cycle and the same prior spectral distribution can be used repeatedly during the same detection cycle.

[0049] The number of steps or stages for frequency sweep is not limited to 25 and can be set appropriately depending on the conditions of use and the required accuracy. The number of cycles for repeating the sweep can also be set appropriately depending on the conditions of use and the required accuracy. [Explanation of symbols]

[0050] 1a...detection unit, 10...antenna, 11a, 11b, 11c, 12a, 12b...antenna loop, 30...transmitter, 32...switch, 33...amplifier circuit, 40...receiver, 41...switch, 42...filter, 43...AD conversion circuit, 50...alarm device, 80...signal processing section, 81...computer, 83...digital signal processing circuit, 100...presence detection system, 100a...circuit body, OB...target, PW...passage, TG...resonant tag, W1...transmitted wave, W2...response wave

Claims

1. a transmitting unit that transmits a transmission wave that is a radio wave; a receiving unit that detects a response wave output from the resonant tag; a signal processing unit that determines whether or not a target is present based on the response wave; The signal processing unit A series of detection data corresponding to the response waves in response to the transmission waves transmitted while gradually changing the frequency in a stepwise manner is acquired multiple times; A fast Fourier transform is performed on the series of detection data acquired each time by digital processing; Smoothing is performed sequentially for each additional spectral distribution obtained by fast Fourier transform. During smoothing, a weighted sum is performed on the smoothing results up to that point and the added spectral distribution using a smoothing coefficient corresponding to the number of smoothing operations up to that point and a smoothing coefficient corresponding to the number of smoothing operations including the current smoothing operation. If the intensity of the spectrum after addition exceeds a predetermined threshold, it is determined that the object has been detected. Presence detection system.

2. The presence detection system according to claim 1 , wherein the signal processing unit performs the smoothing on a group of data after fast Fourier transform that has the same frequency as the transmission wave, among the series of detection data acquired multiple times.

3. The presence detection system according to claim 1 or 2, wherein the signal processing unit adds, for each frequency, a series of spectral distributions obtained from the series of detection data corresponding to the transmission waves transmitted while changing the frequency.

4. 4. The presence detection system according to claim 1, wherein the signal processing unit excludes components of the spectral distribution that are out of the frequency range of the transmission wave.

5. a transmitting unit that transmits a transmission wave that is a radio wave; a receiving unit that detects a response wave output from the resonant tag; a signal processing unit that determines whether or not a target is present based on the response wave; The signal processing unit A series of detection data corresponding to the response waves in response to the transmission waves transmitted while gradually changing the frequency in a stepwise manner is acquired multiple times; A fast Fourier transform is performed on the series of detection data acquired each time by digital processing; A series of spectral distributions obtained by fast Fourier transform are added for each frequency, If the intensity of the spectrum after addition exceeds a predetermined threshold, it is determined that the object has been detected. Presence detection system.

Images