JP5435761B1 - Wiretap detector, wiretap detection system, wiretap detection method, and wiretap detection program - Google Patents

Abstract

There are provided an eavesdropper detection device, an eavesdropper detection system, an eavesdropper detection method, and an eavesdropper detection program capable of automatically finding an eavesdropper easily by using sound existing in the vicinity.
An eavesdropper detection device 2 includes a voice input unit 10 to which a voice is input, a radio wave search unit 12 that searches for a radio wave, and a signal acquisition unit 14 that acquires a voice signal from the radio wave received by the radio wave search unit 12. And a first feature signal indicating the feature of the sound input to the sound input unit 10 and a second feature signal indicating the feature of the sound indicated by the sound signal from the sound signal acquired by the signal acquisition unit 12. Based on the difference signal, the difference signal calculation unit 18 that calculates the difference signal between the feature signal extraction unit 16, the first feature signal and the second feature signal, and the similarity between the first feature signal and the second feature signal is determined. And a similarity determination unit 20 for determination.
[Selection] Figure 1

Description

  The present invention relates to a bug detector, a bug detector system, a bug detector method, and a bug detector program. In particular, the present invention relates to an eavesdropper detection device, an eavesdropper detection system, an eavesdropper detection method, and an eavesdropper detection program for detecting an eavesdropper using sound existing in the vicinity.

  Conventionally, an eavesdropper discovery device for preventing telephone wiretapping, an antenna unit that receives a transmission signal from the eavesdropper and outputs a reception signal, a broadband amplifier that amplifies the reception signal from the antenna unit, A demodulator that demodulates and outputs the amplified received signal output from the broadband amplifier to a voice signal, an input unit that inputs a voice signal from a subscriber telephone, amplifies and attenuates it as necessary, and The cross-correlation function calculation unit that calculates the cross-correlation function value between the audio signal and the demodulated audio signal is compared with the predetermined value and the cross-correlation function value calculated by the cross-correlation function calculation unit is sufficiently high 2. Description of the Related Art A wiretapping device for detecting eavesdropping of a telephone is known that includes a detecting unit that determines whether or not the device is connected, and an audiovisual display unit that notifies the presence / absence of a wiretap based on the determination result of the detecting unit (see, for example, Patent Document 1). .) According to the eavesdropper discovery device for telephone tapping described in Patent Document 1, it is possible to find an eavesdropper that eavesdrops on a telephone being talked using radio waves transmitted from the eavesdropper.

JP-A-9-275446

  However, in the bug detector disclosed in Patent Document 1, not only a bug can be found which is bugging a telephone in a call, but a cross-correlation function value is compared with a predetermined value. Therefore, an eavesdropper cannot be found accurately unless an appropriate predetermined value is calculated and set in advance. Further, in the bug detector disclosed in Patent Document 1, it is necessary to calculate a cross-correlation function value. Therefore, if the bug detector is continuously detected for a long time, a large load is imposed on the bug detector. It may take such a case.

  Accordingly, an object of the present invention is to provide an eavesdropper detection device, an eavesdropper detection system, an eavesdropper detection method, and an eavesdropper detection that can easily find an eavesdropper easily by using sound existing in the vicinity. To provide a program.

In order to achieve the above object, the present invention is a wiretap detector for detecting a wiretap device that modulates acquired sound into a radio wave and transmits it, and includes a sound input unit to which the sound is input and a radio wave to search for the wave. From the search unit, the signal acquisition unit that acquires the audio signal from the radio wave received by the radio wave search unit, the first feature signal indicating the characteristics of the audio input to the audio input unit, and the audio signal acquired by the signal acquisition unit A feature signal extraction unit that extracts a second feature signal indicating the feature of the voice indicated by the voice signal, and a first feature signal obtained at a predetermined first time from the first time to the second time The first absolute difference signal obtained by positively converting the sign of the difference signal between the second characteristic signal and the second feature signal is obtained at a predetermined second time starting from a time different from the first time. Difference between the first feature signal and the second feature signal Second absolute difference signal and the difference signal calculating unit for calculating a sum signal calculation unit for calculating a first absolute difference signal and the sum signal obtained by summing a second absolute difference signal obtained by converting the sign of the signal exactly And a wiretap detector that includes a similarity determination unit that determines the similarity between the first feature signal and the second feature signal based on the magnitude of the sum signal .

  Further, in the above-described bug detector, an average for calculating an average sum signal obtained by dividing a sum signal obtained by summing a plurality of absolute difference signals calculated by the sum signal calculation unit by the number of absolute difference signals used for summation. A summation signal calculation unit may be further provided, and the similarity judgment unit may judge the similarity based on the magnitude of the average summation signal.

  In the bug detector, the first feature signal is at least selected from the group including a volume level of the voice input to the voice input unit, a volume level in a predetermined frequency band, and a profile waveform of the volume level. The second feature signal includes at least one signal selected from the group including a volume level of the voice indicated by the voice signal, a volume level in a predetermined frequency band, and a profile waveform of the volume level. You may go out.

  Further, in the above-described bug detector, the signal acquisition unit can continuously acquire the audio signal from the radio wave every predetermined unit time.

  The bug detector further includes a continuity determination unit that determines whether or not the determination result of the similarity determination unit is continued for a predetermined period, and the similarity determination unit includes the difference signal and the difference signal. Similarity can also be determined based on the determination result of the continuity determination unit.

  Further, in the above-described bug detector, when the difference signal calculation unit does not match the input time of the sound input to the sound input unit and the acquisition time when the signal acquisition unit acquires the sound signal, the input time and the acquisition time And the difference signal can be calculated.

  The wiretapping detector further includes an information storage unit that stores similarity determination information indicating a determination result of the similarity determination unit, and the similarity determination unit is configured to input a voice to the voice input unit. When similarity is determined and similarity determination information is stored in the information storage unit, and voice input to the voice input unit is interrupted, similarity determination is interrupted and voice input to the voice input unit is resumed In this case, the similarity determination can be resumed based on the similarity determination information stored in the information storage unit.

In order to achieve the above object, the present invention is a wiretapping detector that detects a wiretapping device that modulates acquired sound into a radio wave and transmits the sound, and searches for a sound input unit to which the sound is input and the radio wave. From the radio wave search unit, the signal acquisition unit that acquires the audio signal from the radio wave received by the radio wave search unit, the first feature signal indicating the characteristics of the audio input to the audio input unit, and the audio signal acquired by the signal acquisition unit A feature signal extraction unit that extracts a second feature signal indicating the feature of the voice indicated by the voice signal, a difference signal calculation unit that calculates a difference signal between the first feature signal and the second feature signal, and a difference signal A similarity determination unit that determines the similarity between the first feature signal and the second feature signal, and an information storage unit that stores similarity determination information indicating a determination result of the similarity determination unit. When the judgment part receives a voice input to the voice input part The similarity is determined and the similarity determination information is stored in the information storage unit. When the voice input to the voice input unit is interrupted, the similarity determination is interrupted and the voice input to the voice input unit is input. When the computer restarts, an eavesdropper detection device is provided that restarts the determination of similarity based on the similarity determination information stored in the information storage unit.
Further, in order to achieve the above object, the present invention is a wiretap detection system for detecting a wiretap device that modulates acquired sound into a radio wave and transmits the sound, and includes a sound input unit to which peripheral sound is input, A radio wave search unit to search, a signal acquisition unit to acquire an audio signal from radio waves received by the radio wave search unit, a first feature signal indicating the characteristics of the audio input to the audio input unit, and the audio acquired by the signal acquisition unit A feature signal extracting unit that extracts a second feature signal indicating the feature of the voice indicated by the voice signal from the signal, and a first time obtained from a first time to a second time. A first absolute difference signal obtained by positively converting the sign of the difference signal between the one feature signal and the second feature signal, and a predetermined second time starting from a time different from the first time First characteristic signal and second obtained in Calculating a difference signal calculating unit for calculating a second absolute difference signal obtained by converting the sign of the difference signal between symptom signal exactly, the first absolute difference signal and the sum signal obtained by summing a second absolute difference signal A summation signal calculation unit, a similarity judgment unit that judges the similarity between the first feature signal and the second feature signal based on the magnitude of the summation signal, and a judgment result that acquires a judgment result of the similarity judgment unit When the acquisition unit and the determination result are information indicating that the first feature signal and the second feature signal are similar, an eavesdropper detection system including a notification unit that reports the presence of the eavesdropper is provided.

Further, in order to achieve the above object, the present invention is a wiretap detection method for detecting a wiretap device that modulates acquired sound into a radio wave and transmits the sound, and a sound input step in which sound is input to the sound input unit, In a radio wave search stage for searching for radio waves, a signal acquisition stage for acquiring audio signals from radio waves received in the radio wave search stage, a first feature signal indicating the characteristics of the audio input in the audio input stage, and a signal acquisition stage A feature signal extraction stage for extracting a second feature signal indicating a feature of the voice indicated by the voice signal from the acquired voice signal, and a first predetermined time from the first time to the second time The first absolute difference signal obtained by positively converting the sign of the difference signal between the first feature signal and the second feature signal obtained in step 1 and a predetermined time starting from a time different from the first time Second A difference signal calculating step of calculating a second absolute difference signal first characteristic signal obtained and positively changes the sign of the difference signal and the second characteristic signal between the first absolute difference signal and the second A summation signal calculation step of calculating a summation signal obtained by summing the absolute difference signals; and a similarity judgment step of judging the similarity between the first feature signal and the second feature signal based on the magnitude of the summation signal. An eavesdropper detection method is provided.

Further, in order to achieve the above object, the present invention is a wiretapping device detection program for detecting a wiretap device that modulates acquired sound into radio waves and transmits the sound, and a sound input function for inputting sound to a computer and a sound input unit A radio wave search function for searching for radio waves, a signal acquisition function for acquiring an audio signal from radio waves received by the radio wave search function, a first feature signal indicating the characteristics of the audio input by the audio input function, and signal acquisition A feature signal extraction function for extracting a second feature signal indicating the feature of the voice indicated by the voice signal from the voice signal acquired in the function, and a first predetermined time from the first time to the second time The first absolute difference signal obtained by positively converting the sign of the difference signal between the first feature signal and the second feature signal obtained at the time and the time starting from a time different from the first time A difference signal calculating function for calculating a first characteristic signal obtained in the second time and a second absolute difference signal positively changes the sign of the difference signal and the second characteristic signal to a predetermined, first The similarity between the first feature signal and the second feature signal is determined based on the sum signal calculation function for calculating the sum signal obtained by summing the absolute difference signal and the second absolute difference signal, and the magnitude of the sum signal. An eavesdropper detection program for realizing the similarity determination function is provided.

  According to the wiretap detection device, wiretap detection system, wiretap detection method, and wiretap detection program according to the present invention, it is possible to easily find a wiretap easily by using sound existing in the vicinity. A wiretap detection device, a wiretap detection system, a wiretap detection method, and a wiretap detection program can be provided.

It is a schematic diagram of a bug detector system according to an embodiment of the present invention. It is a functional block diagram of the bug detector device according to the embodiment of the present invention. It is a functional block diagram of a server concerning an embodiment of the invention. (A)-(c) is a schematic diagram which shows the method of calculating a difference signal in the difference signal calculation part which concerns on embodiment of this invention, respectively. It is a schematic diagram which shows the method of calculation of the total signal by the total signal calculation part which concerns on embodiment of this invention, and the calculation of the average total signal by the average total signal calculation part. It is a flowchart of a process of the bug detector system which concerns on embodiment of this invention. It is a hardware block diagram of the wiretap detection system which concerns on embodiment of this invention.

[Embodiment]
FIG. 1 shows an outline of a bug detector system according to an embodiment of the present invention.

(Overview of eavesdropper detection system 1)
The eavesdropper detection system 1 according to the present embodiment acquires a sound generated in the vicinity and receives a radio wave 900. The eavesdropper detection system 1 automatically determines in real time whether the eavesdropper 7 is installed in the detection area 8 based on whether the acquired audio signal including the characteristics of the audio is modulated by the received radio wave 900. System that automatically detects and reports. Here, for example, the wiretapping device 7 has a predetermined sound acquisition unit that acquires external sound, a conversion unit that converts the sound acquired by the sound acquisition unit into an audio signal, and an audio signal converted by the conversion unit. A radio wave transmission unit that modulates the radio wave in the frequency band and transmits the radio wave to an external receiver.

  Specifically, the eavesdropper detection device 2 included in the eavesdropper detection system 1 according to the present embodiment has a sound that naturally occurs in the detection area 8 that is an area where the eavesdropper 7 may be installed (for example, To get life sounds). The bug detector 2 extracts a first feature signal indicating the feature of the sound. Further, when the eavesdropper detection device 2 receives a radio wave in a predetermined frequency band, the eavesdropper detection device 2 determines whether or not the radio signal includes an audio signal. Then, when the radio wave includes an audio signal, the wiretap detection device 2 extracts a second feature signal indicating the audio feature indicated by the audio signal. Next, the bug detector 2 determines whether the first feature signal and the second feature signal are similar based on a difference signal that is a difference between the first feature signal and the second feature signal. Then, the bug detector 2 determines whether the bug 7 is installed in the detection area 8 based on the similarity between the first feature signal and the second feature signal.

  When it is determined that the wiretap 7 is installed in the detection area 8, the wiretap detector 2 transmits information indicating the determination result to the server 3 via the communication network 6. In addition, when it is judged that the bug detector 7 is installed in the detection area 8, the bug detector 2 can also issue an alarm from the bug detector 2 itself. When the server 3 acquires information indicating the determination result, the server 3 reports the information to the user terminal 4 and the user terminal 5 that are determined in advance via the communication network 6. Note that the server 3 can also report information indicating the determination result to a plurality of other user terminals via the communication network 6.

  Here, the user terminal 4 and the user terminal 5 are, for example, communication terminals or information terminals held by a user who manages the detection area 8. The users include individuals who manage the detection area 8, private businesses, administrative organizations such as municipalities, national government offices, and the like. The user who has acquired the determination result indicating that the bugging device 7 is installed from the user terminal, for example, removes the bugging device 7 by dispatching a person in charge for handling the wireless device to the detection area 8. Can do.

  In the present embodiment, the server 3 is one or more servers, and can be configured to include a plurality of servers. For example, the server 3 includes a server that stores digital information such as a mail server or a file server. The communication network 6 is a mobile phone network and / or a communication network such as the Internet. The communication network 6 can also include a communication network such as a wired LAN and a wireless LAN. Furthermore, the user terminal 4 and the user terminal 5 include, for example, a personal computer, a notebook computer, a tablet PC, a smartphone, or a mobile phone.

  FIG. 2 shows an example of the functional configuration of the wiretap detector according to the embodiment of the present invention. FIG. 3 shows an example of the functional configuration of the server according to the embodiment of the present invention.

  An eavesdropper detection system 1 according to the present embodiment includes an eavesdropper detection device 2 that detects an eavesdropper 7 that emits a radio wave modulated from a sound signal of a sound generated in the detection area 8 and an eavesdropper detection device 2 And a server 3 that receives information indicating that the wiretap 7 has been detected and notifies the user and / or the user terminal that the wiretap 7 has been detected. The wiretap detector 2 can have a part or all of the functions and configurations of the server 3. The server 3 may have a part or all of the functions and configurations of the bug detector 2.

(Details of the configuration of the wiretap detector 2)
An eavesdropper detection device 2 that detects an eavesdropper 7 that modulates acquired sound into radio waves and transmits the sound, includes an audio input unit 10 to which audio is input, a radio wave search unit 12 that searches for radio waves, and an audio signal from the radio waves. A signal acquisition unit 14 to acquire, a feature signal extraction unit 16 to extract a feature signal indicating a feature of speech, a difference signal calculation unit 18 to calculate a difference signal between the first feature signal and the second feature signal, and a difference signal A similarity determination unit 20 that determines the similarity between the first feature signal and the second feature signal, and an output unit 28 that outputs the determination result of the similarity determination unit 20.

  Further, the bug detector 2 includes a sum signal calculation unit 22 that sums a plurality of absolute difference signals obtained by converting the signs of the plurality of difference signals to a positive value, and an average sum signal calculation unit that calculates an average value of the absolute difference signals. 24, a continuity determination unit 26 that determines whether the determination result of the similarity determination unit 20 continues for a predetermined period, an information storage unit 30 that stores the determination result of the similarity determination unit 20, and a determination result A transmission unit 32 configured to transmit to the outside;

(Voice input unit 10)
The voice input unit 10 receives a voice generated around the bug detector 2. The voice input unit 10 is, for example, a microphone. Moreover, the sound generated in the vicinity of the bug detector 2 includes sounds generated in real time such as the user's voice and living sounds in the vicinity of the bug detector 2. The bug detector 2 includes a single voice input unit 10 or a plurality of voice input units 10. The voice input unit 10 supplies the voice signal of the input voice to the feature signal extraction unit 16. Further, the voice input unit 10 supplies information indicating that the voice is being input to the similarity determination unit 20 while the voice is being input. The voice input unit 10 can also supply a voice signal of the input voice to the feature signal extraction unit 16 in association with the time when the voice is input (hereinafter referred to as “input time”). In addition, the voice input unit 10 receives external voice input continuously or intermittently.

(Radio wave search unit 12)
The radio wave search unit 12 searches for radio waves in a predetermined frequency band. The radio wave search unit 12 searches for radio waves continuously, intermittently, or in response to a search instruction from the outside. Specifically, the radio wave search unit 12 searches for radio waves in a plurality of predetermined frequency bands that are considered to be used by the wiretapping device 7. When the radio wave search unit 12 detects a radio wave in a predetermined frequency band, the radio wave search unit 12 supplies the detected radio wave to the signal acquisition unit 14. The radio wave search unit 12 supplies information indicating that no radio wave has been received to the output unit 28 when no radio wave is received in a predetermined frequency band after searching for the radio wave for a predetermined time. You can also

(Signal acquisition unit 14)
When the radio wave search unit 12 detects a radio wave, the signal acquisition unit 14 acquires an audio signal modulated by the detected radio wave. That is, the signal acquisition unit 14 demodulates the audio signal modulated by the radio wave. The signal acquisition unit 14 supplies the demodulated audio signal to the feature signal extraction unit 16.

  The signal acquisition unit 14 can acquire an audio signal according to the search by the radio wave search unit 12. That is, when the radio wave search unit 12 searches for radio waves continuously, intermittently, or in response to a search instruction from the outside, the signal acquisition unit 14 is also continuous, intermittent, or while the search instruction is issued, Acquire an audio signal. Further, when the audio signal is not modulated with the radio wave, the signal acquisition unit 14 can also supply the output unit 28 with information indicating that the audio signal is not modulated with the radio wave. The signal acquisition unit 14 can also acquire the audio signal in association with the time when the radio signal is detected and the audio signal is acquired (hereinafter referred to as “acquisition time”).

(Feature signal extraction unit 16)
The feature signal extraction unit 16 extracts a feature signal indicating the feature of the audio signal from the audio signal of the audio input to the audio input unit 10 and the audio signal acquired by the signal acquisition unit 14. The feature signal extraction unit 16 can extract a feature signal according to the operations of the voice input unit 10 and the signal acquisition unit 14. That is, the feature signal extraction unit 16 extracts feature signals while there are continuous, intermittent, or external search instructions. Specifically, the feature signal extraction unit 16 determines the feature of the sound indicated by the sound signal from the first feature signal indicating the feature of the sound input to the sound input unit 10 and the sound signal acquired by the signal acquisition unit 14. The second feature signal shown is extracted.

  The feature signal is a signal that is extracted from the audio signal and that indicates the characteristics of the audio indicated by the audio signal. Specifically, the first feature signal includes a volume level of the voice input to the voice input unit 10, a volume level in a predetermined frequency band, and a profile waveform of the volume level (that is, a change in volume level along a time series). At least one signal selected from the group including: Further, the first characteristic signal may be a signal including all of the volume level of the voice input to the voice input unit 10, the volume level in a predetermined frequency band, and the profile waveform of the volume level. Further, the second feature signal is at least one selected from the group including a volume level of a voice indicated by the voice signal acquired by the signal acquisition unit 14, a volume level in a predetermined frequency band, and a profile waveform of the volume level. Includes signal. Further, the second feature signal may be a signal including all of the sound volume level indicated by the sound signal, the sound volume level in a predetermined frequency band, and the profile waveform of the sound volume level. As an example, the first feature signal is a volume level of a voice input to the voice input unit 10, a volume level in a 400 Hz band, a volume level in a 1 KHz band, and a volume level in a 2 KHz band. Similarly, the second feature signal includes a sound volume level indicated by the sound signal acquired by the signal acquisition unit 14, a sound volume level in the 400 Hz band, a sound volume level in the 1 KHz band, and a sound volume level in the 2 KHz band. The feature signal extraction unit 16 supplies the extracted first feature signal and second feature signal to the difference signal calculation unit 18. Note that the feature signal extraction unit 16 can associate information indicating the input time with the first feature signal and also associate information indicating the acquisition time with the second feature signal.

  The sound volume level input to the sound input unit 10 and the sound volume level indicated by the sound signal indicate, for example, the accent of the sound. As an example, when the voice input to the voice input unit 10 is a human voice, information indicating the accent of the voice and information indicating a break in the vocabulary are included in the volume characteristic and the voice feature signal indicated by the voice signal. included. When the voice input to the voice input unit 10 is music, information indicating a rhythm having strong and weak regularity is included in the voice feature signal. Furthermore, the volume level in the predetermined frequency band can include volume levels in a plurality of frequency bands. Thereby, when the voice input to the voice input unit 10 is a human voice, the characteristic signal is a signal indicating a gender-specific pitch, a signal indicating individual differences in voice, and a plurality of volume levels for each band. It can include information indicating that When the voice input to the voice input unit 10 is music, the change in the melody of the music corresponds to the time change in the volume level of each of the plurality of frequency bands. Information indicating changes may be included. The profile waveform indicates a transition that changes along the predetermined time axis of the volume level or the volume level in each of a plurality of predetermined frequency bands.

(Difference signal calculation unit 18)
The difference signal calculation unit 18 calculates a difference signal between the first feature signal and the second feature signal received from the feature signal extraction unit 16 as a deviation signal. The difference signal calculation unit 18 calculates the difference signal while there are continuous, intermittent, or external search instructions. As an example, the first feature signal is a time-series change in the volume level (hereinafter referred to as “first volume level”) of the voice input to the voice input unit 10, and the second feature signal is the signal acquisition unit 14. In the case of the time series change of the sound volume level (hereinafter, referred to as “second volume level”) indicated by the sound signal acquired in step S1, the difference signal calculation unit 18 sets the first volume level and the first volume level along the time series. The difference signal is calculated by calculating the difference between the two sound volume levels. Moreover, the difference signal calculation part 18 can calculate a difference signal for every predetermined unit time. Here, the “predetermined unit time” is a time determined from the time that the pronunciation continues when the morpheme, which is the smallest meaningful language unit, is pronounced. As an example, the predetermined unit time is set to a time of about several milliseconds to several hundred milliseconds. Furthermore, the difference signal calculation unit 18 can also calculate an absolute difference signal obtained by converting the sign of the calculated difference signal to a positive value.

  Here, in the bug detector 2, since the similarity between the first feature signal and the second feature signal is determined without using the cross-correlation function, the delay of the sound signal that occurs based on the sound speed of the sound is taken into consideration. It may be necessary. Therefore, the difference signal calculation unit 18 matches the input time and the acquisition time when the input time of the audio input to the audio input unit 10 and the acquisition time when the signal acquisition unit 14 acquires the audio signal do not match. The difference signal can also be calculated. That is, the difference signal calculation unit 18 compares the information indicating the input time associated with the first feature signal and the information indicating the acquisition time associated with the second feature signal. Then, when the input time and the acquisition time are different, the difference signal calculation unit 18 calculates the difference signal by matching the start point of either the first feature signal or the second feature signal with the start point of the other signal. be able to. The difference signal calculation unit 18 supplies the calculated difference signal or absolute difference signal to the similarity determination unit 20. Further, the difference signal calculation unit 18 supplies the calculated absolute difference signal to the total signal calculation unit 22.

  When the feature indicated by the first feature signal or the second feature signal includes a sudden change, the difference signal calculation unit 18 uses the first feature signal or the second feature signal as a reference when the sudden change occurs. The time axis of the feature signal can also be shifted. For example, it is assumed that the first feature signal and the second feature signal are volume level profile waveforms, and the first feature signal includes a sudden change in volume level. In this case, the first feature signal is extracted based on the voice input to the voice input unit 10, and the second feature signal is extracted based on the voice signal included in the radio wave detected by the radio wave search unit 12. Strictly speaking, there is a difference between the time when the sudden volume level change occurs in the one feature signal and the time when the sudden volume level change occurs in the second feature signal depending on the sound speed. Accordingly, the difference signal calculation unit 18 shifts the time axis of the second feature signal in a direction matching the time axis of the first feature signal, starting from the time when the sudden volume level change in the first feature signal occurs. . Thereby, the difference signal calculation part 18 can calculate a difference signal irrespective of the difference which arises according to the sound speed.

(Similarity judgment unit 20)
The similarity determination unit 20 determines the similarity between the first feature signal and the second feature signal based on the difference signal calculated by the difference signal calculation unit 18. For example, when the first feature signal and the second feature signal are volume level profile waveforms, the similarity determination unit 20 determines the first feature signal when the magnitude of the difference signal is equal to or less than a predetermined threshold. And the second feature signal are the same or similar. On the other hand, when the magnitude of the difference signal exceeds a predetermined threshold, the similarity determination unit 20 determines that the first feature signal and the second feature signal are not similar.

  The similarity determination unit 20 can also determine the similarity between the first feature signal and the second feature signal based on the absolute difference signal. For example, the similarity determination unit 20 determines that the first feature signal and the second feature signal are the same or similar when the magnitude of the absolute difference signal is equal to or less than a predetermined threshold. Then, the similarity determination unit 20 determines that the first feature signal and the second feature signal are not similar when the magnitude of the absolute difference signal exceeds a predetermined threshold. The similarity determination unit 20 supplies determination result information, which is information indicating the determination result, to the continuity determination unit 26, the output unit 28, and / or the information storage unit 30.

(Total signal calculation unit 22)
First, the difference signal calculation unit 18 calculates an absolute difference signal along a time series for each predetermined unit time. That is, the difference signal calculation unit 18 continuously calculates a plurality of absolute difference signals along a time series. Then, the sum signal calculation unit 22 sums the plurality of absolute difference signals calculated by the difference signal calculation unit 18 to calculate a sum signal. Specifically, the sum signal calculation unit 22 obtains the first absolute difference signal obtained at a predetermined first time (that is, the time from the first time to the second time after the first time). And a sum signal obtained by summing the second absolute difference signal obtained at a predetermined second time starting from a time different from the first time. That is, the sum signal calculation unit 22 sums the nth absolute difference signal obtained at the nth time and the n + 1th absolute difference signal obtained at the (n + 1) th time (where n is an integer equal to or greater than 1). .) The sum signal calculation unit 22 can also calculate a sum signal obtained by summing the nth absolute difference signal to the n + mth absolute difference signal (where n and m are integers equal to or greater than 1, and n ≦ m.)

  The total signal calculation unit 22 supplies the calculated total signal to the similarity determination unit 20 and the average total signal calculation unit 24. The similarity determination unit 20 determines similarity based on the magnitude of the sum signal. For example, the similarity determination unit 20 determines that the first feature signal and the second feature signal are not similar when the sum signal exceeds a predetermined threshold, and the first feature when the sum signal is equal to or less than the threshold. It is determined that the signal and the second feature signal are similar or coincident. Therefore, since the sum signal calculation unit 22 sums the absolute difference signals, it is possible to prevent cancellation of the difference signals that may occur when the plurality of difference signals are simply summed and the signs of the difference signals are different from each other.

(Average total signal calculation unit 24)
The average sum signal calculation unit 24 divides the sum signal obtained by summing a plurality of absolute difference signals calculated by the sum signal calculation unit 22 by the number of absolute difference signals used for summation of the sum signal, and is obtained as an average sum signal Is calculated. For example, when the sum signal calculation unit 22 calculates the sum signal by summing the first absolute difference signal and the second absolute difference signal, the average sum signal calculation unit 24 calculates the sum signal by the number of absolute difference signals. The average sum signal is calculated by dividing by “2”. That is, the average sum signal calculation unit 24 continuously calculates the moving average of the sum signal as an average sum signal by dividing the sum signal obtained by summing the x absolute difference signals by “x” (however, x Is an integer greater than or equal to 1.)

  The average sum signal calculation unit 24 supplies the calculated average sum signal to the similarity determination unit 20. The similarity determination unit 20 determines similarity based on the magnitude of the average sum signal. For example, the similarity determination unit 20 determines that the first feature signal and the second feature signal are the same or similar when the magnitude of the average sum signal is equal to or less than a predetermined threshold. And the similarity judgment part 20 judges that a 1st feature signal and a 2nd feature signal are not similar, when the magnitude | size of an average sum signal exceeds a predetermined threshold value.

(Continuity determination unit 26)
The continuity determination unit 26 determines whether or not the determination result of the similarity determination unit 20 continues as the substantially same determination result for a predetermined period. That is, information indicating that the first feature signal and the second feature signal are similar is used as the first determination result information, and information indicating that the first feature signal and the second feature signal are not similar is the second determination result information. In the case of the result information, the continuity determination unit 26 indicates that the first determination result information is continued when the first determination result information is continuously acquired for a predetermined period. Information is supplied to the similarity determination unit 20. On the other hand, the continuity determination unit 26 receives the first determination result information and the second determination when the second determination result information is continuously acquired for a predetermined period, and within the predetermined period. When the result information is mixed, information indicating that the first determination result information is not continued is supplied to the similarity determination unit 20.

  The similarity determination unit 20 determines the similarity between the first feature signal and the second feature signal based on the difference signal, the absolute difference signal, or the sum signal, and the determination result of the continuity determination unit 26. Specifically, when the similarity determination unit 20 receives information indicating that the first determination result information is continued from the continuity determination unit 26, the first feature signal and the second feature signal are similar. Judge that. On the other hand, when the similarity determination unit 20 receives information indicating that the first determination result information is not continued from the continuity determination unit 26, the first feature signal and the second feature signal are not similar. Judge. Accordingly, the similarity determination unit 20 can determine that the first feature signal and the second feature signal are similar when the first feature signal and the second feature signal are similar for a predetermined period. When the first feature signal and the second feature signal are accidentally similar in a time that is less than a predetermined period, it is possible to prevent erroneous detection of the wiretapping device 7.

(Output unit 28)
The output unit 28 outputs the determination result information of the similarity determination unit 20. When the output unit 28 receives information indicating that no radio wave has been received from the radio wave search unit 12 (hereinafter referred to as “radio wave non-reception information”), the output unit 28 can also output radio wave non-reception information. Further, when the output unit 28 receives information indicating that the audio signal is not modulated into a radio wave (hereinafter referred to as “audio signal non-existence information”) from the signal acquisition unit 14, the output unit 28 outputs the audio signal non-existence information. You can also

  The output unit 28 can also have a function of emitting a predetermined sound as an alarm sound. For example, the output unit 28 can emit an alarm sound that alerts the user. As an example, when the output unit 28 receives determination result information indicating that the first feature signal and the second feature signal are similar from the similarity determination unit 20, the output unit 28 emits an alarm sound.

  The output unit 28 can also include a plurality of lamp units that emit different colors. When the output unit 28 receives determination result information indicating that the first feature signal and the second feature signal are similar from the similarity determination unit 20, the output unit 28 causes the lamp unit that emits the first color to emit light. On the other hand, when receiving the determination result information indicating that the first feature signal and the second feature signal are not similar, the output unit 28 emits a lamp unit that emits a second color different from the first color. Let Similarly, when the output unit 28 receives information indicating that the radio wave has not been received, the output unit 28 may cause the lamp unit that emits a third color different from the first color and the second color to emit light. it can. The output unit 28 supplies the determination result information to the transmission unit 32.

(Information storage unit 30)
The information storage unit 30 stores similarity determination information indicating the determination result of the similarity determination unit 20. The information storage unit 30 stores the similarity determination information in association with the time when the similarity determination unit 20 outputs the determination result. Then, the similarity determination unit 20 determines the similarity and continues to store the similarity determination information in the information storage unit 30 while the voice is input to the voice input unit 10. The similarity determination unit 20 interrupts the similarity determination when the voice input to the voice input unit 10 is interrupted. Subsequently, when the input of voice to the voice input unit 10 is resumed, the similarity judgment unit 20 may resume the judgment of similarity based on the similarity judgment information stored in the information storage unit 30. it can.

(Transmitter 32)
The transmission unit 32 transmits the determination result information received from the output unit 28. Specifically, the transmission unit 32 transmits the determination result information to the server 3 via the communication network 6 such as a mobile phone network or the Internet. The transmission unit 32 can also transmit the radio wave non-reception information and / or the audio signal non-existence information acquired by the output unit 28 to the server 3.

(Judgment result acquisition unit 36)
The determination result acquisition unit 36 acquires the determination result information from the similarity determination unit 20. The determination result acquisition unit 36 can also associate a detection device identifier that uniquely identifies the bug detector 2 that has transmitted the determination result information with the determination result information. The determination result acquisition unit 36 supplies the acquired determination result information to the determination result storage unit 38 in association with the detection device identifier. The determination result acquisition unit 36 can also acquire radio wave non-reception information and audio signal non-existence information. In this case, the determination result acquisition unit 36 can also supply radio wave non-reception information and audio signal non-existence information to the determination result storage unit 38.

(Judgment result storage unit 38)
The determination result storage unit 38 stores the determination result information received from the determination result acquisition unit 36 in association with the detection device identifier. The determination result storage unit 38 supplies the detection device identifier and the determination result information associated with the detection device identifier to the notification unit 40. Also, the determination result storage unit 38 can store the radio wave non-reception information and the audio signal non-existence information in association with the detection device identifier. The determination result storage unit 38 can also store radio wave non-reception information and audio signal non-existence information in association with time.

(Reporting unit 40)
When the determination result information indicates that the first feature signal and the second feature signal are similar, that is, when the notification unit 40 indicates that the bug 7 exists in the detection area 8, Report the presence information to the user terminal. The reporting unit 40 reports to the user terminal substantially in real time via the communication network 6. The reporting unit 40 can also report by transmitting text data, audio data, and / or image data including information indicating the presence of the bug 7 to the user terminal. Further, the reporting unit 40 uses the determination result information stored in the determination result storage unit 38 to report to the user terminal at a predetermined time, or the determination result acquisition unit 36 acquires the determination result information. It is also possible to notify the user terminal after a predetermined time has elapsed from the time.

  The detection device identifier can be associated with a user terminal identifier that uniquely identifies the user terminal to be notified. One or a plurality of user terminal identifiers can be associated with the detection device identifier. The reporting unit 40 can report information indicating the presence of the wiretapping device 7 to the user terminal identified by the user terminal identifier associated with the detection device identifier received from the determination result storage unit 38.

  The wiretap detection device 2 can further include a sound generation unit that generates sound having a predetermined output pattern. In this case, the wiretap detector 2 generates the sound from the sound generation unit and causes the sound input unit 10 to input the sound. Then, in the same manner as described above, the bug detector 2 may detect the presence of the bug 7 based on the similarity between the first feature signal and the second feature signal.

  4 (a) to 4 (c) schematically show a method for calculating a difference signal in the difference signal calculation unit according to the embodiment of the present invention. 4A to 4C schematically illustrate a method in which the difference signal calculation unit 18 calculates the difference signal in a predetermined unit time. However, the difference signal calculation unit 18 may continuously or The difference signal can be calculated intermittently.

  The difference signal calculation unit 18 calculates a difference signal between the first feature signal and the second feature signal received from the feature signal extraction unit 16 as a deviation signal. For example, refer to FIG. In the upper graph of FIG. 4A, the X axis is time, and the Y axis is the volume level. The same applies to FIGS. 4B and 4C. In FIG. 4A, the first feature signal 160 and the second feature signal 162 have profiles of substantially the same volume level within a predetermined time. In this case, the difference signal 180 calculated by the difference signal calculation unit 18 indicates a signal profile that is substantially zero in magnitude within a predetermined time, as shown in the lower part of FIG.

  Reference is also made to FIG. In this case, the first feature signal 160 and the second feature signal 164 have a profile in which the volume level of the first feature signal 160 is larger than that of the second feature signal 164 within a predetermined time. In this case, the difference signal 182 calculated by the difference signal calculation unit 18 shows a signal profile larger than at least the difference signal 180 within a predetermined time, as shown in the lower part of FIG.

  Further, refer to FIG. In this case, the first feature signal 168 and the second feature signal 166 have a profile in which the volume level of the first feature signal 168 is smaller than that of the second feature signal 166 within a predetermined time. In this case, the difference signal 184 calculated by the difference signal calculation unit 18 shows a negative signal profile within a predetermined time, as shown in the lower part of FIG.

  The difference signal calculated by the difference signal calculation unit 18 is added by the sum signal calculation unit 22 and calculated as a sum signal. In this case, if a positive difference signal and a negative difference signal are mixed in the difference signal used by the sum signal calculation unit 22 for summation, a correct sum signal may not be calculated. Therefore, it is preferable that the difference signal calculation unit 18 calculates the absolute difference signal by converting the sign of the calculated difference signal to positive.

  FIG. 5 schematically shows a method of calculating a summed signal by the summed signal calculating unit and a method of calculating an average summed signal by the average summed signal calculating unit according to the embodiment of the present invention.

In the example of FIG. 5, the case where the difference signal calculation unit 18 continuously calculates the difference signal will be described. In FIG. 5, the horizontal axis indicates the passage of time, and the time from time t _n to time t _{n + 1} (that is, the time calculated from the difference between time t _{n + 1} and time t _n ) is the same. (However, n is an integer of 1 or more.) In this case, the difference signal calculation unit 18 calculates a difference signal every predetermined unit time. For example, if the unit time determined in advance is the time 200 the same length and from time 0 to time t _3, the difference signal calculating unit 18, first between the time 200 from time 0 to time t ₃ 1 A difference signal between the feature signal and the second feature signal is calculated. It should be noted that the predetermined unit time can be changed as appropriate according to the pronunciation time of the morpheme.

Subsequently, the difference signal calculating unit 18 calculates a difference signal in the time 202 from the time _{t 1} to time _{t 4.} Similarly, the differential signal calculation unit 18 sequentially calculates differential signals at time 204, time 206, time 208, time 210, time 212, and time 214. Then, the difference signal calculation unit 18 converts the sign of each of the calculated plurality of difference signals to a positive value to calculate an absolute difference signal.

  Subsequently, the combined signal calculation unit 22 calculates a combined signal by adding a plurality of absolute difference signals. The number of absolute difference signals added by the combined signal calculation unit 22 can be determined as appropriate according to the time required for the calculation process. As an example, the sum signal calculation unit 22 sums three absolute difference signals from time 200 to time 204 to calculate a sum signal. Similarly, the sum signal calculation unit 22 calculates a plurality of sum signals by summing the absolute difference signals of three consecutive times including the time before and after the one time, with one time as a reference. To do. Then, the average sum signal calculation unit 24 continuously calculates the moving average of the sum signal as an average sum signal by dividing the sum signal by the number of absolute difference signals used to calculate the sum signal.

(Effect of embodiment)
The eavesdropper detection system 1 according to the present embodiment acquires a sound existing around the eavesdropper detection device 2, for example, a living sound including conversation, music, noise, and the like, and obtains a first characteristic signal of the living sound. In addition to extraction, a second feature signal of an audio signal modulated to radio waves in a predetermined frequency band is extracted. Then, the eavesdropper detection system 1 determines whether or not the first feature signal and the second feature signal are similar based on the difference signal between the first feature signal and the second feature signal. Is installed in the detection area 8. Therefore, according to the eavesdropper detection system 1, the eavesdropper 7 can be automatically found only by using the sound existing around the eavesdropper detection device 2. It is not necessary to intentionally emit a loud signal. Thereby, according to the wiretapping device detection system 1, when the wiretap device 7 is detected, the wiretap device 7 can be easily detected without having to endure the signal sound or going out of the detection region 8. Can be executed automatically.

  In addition, the wiretap detector 2 according to the present embodiment can move only the wiretap detector 2 alone to various places. Thereby, according to the eavesdropper detection device 2, an eavesdropper can be detected even in a moving vehicle, a vehicle interior space such as public transportation, or various rooms in various facilities.

(Outline of the processing flow of the bug detector system 1)
FIG. 6 shows an example of the processing flow of the bug detector system according to the embodiment of the present invention.

  First, the voice input unit 10 inputs ambient voice (step 10; hereinafter, step is represented as “S”). The voice input unit 10 supplies the acquired voice to the feature signal extraction unit 16. Further, the radio wave search unit 12 searches for radio waves in a predetermined frequency band (S15). When the radio wave search unit 12 detects a radio wave (S20: Yes), the signal acquisition unit 14 acquires an audio signal from the radio wave (S25). On the other hand, when the radio wave search unit 12 does not detect the radio wave (S20: No), the signal acquisition unit 14 supplies information indicating that the radio wave is not detected to the output unit 28. The output unit 28 supplies the information to the transmission unit 32. The transmission unit 32 transmits the information to the server 3 (S75).

  Next, the feature signal extraction unit 16 extracts the first feature signal from the voice received from the voice input unit 10 and extracts the second feature signal from the voice signal received from the signal acquisition unit 14 (S30). The feature signal extraction unit 16 supplies the extracted first feature signal and second feature signal to the difference signal calculation unit 18. The difference signal calculation unit 18 calculates a difference signal between the first feature signal and the second feature signal (S35). Further, the difference signal calculation unit 18 converts the difference signal into an absolute difference signal by converting the sign of the calculated difference signal to positive (S40). The difference signal calculation unit 18 supplies the absolute difference signal to the sum signal calculation unit 22. The sum signal calculation unit 22 calculates a sum signal using the absolute difference signal (S45). The total signal calculation unit 22 supplies the calculated total signal to the similarity determination unit 20.

  The similarity determination unit 20 determines whether or not the first feature signal and the second feature signal are similar based on the sum signal (S50). When the similarity determination unit 20 determines that the first feature signal and the second feature signal are similar or identical (S50: Yes), the similarity determination unit 20 indicates that the bug 7 is in the detection area 8. It is determined that it exists (S55). The similarity determination unit 20 supplies determination result information indicating that the bug 7 exists to the output unit 28. The output unit 28 outputs the determination result information received from the similarity determination unit 20 and supplies the determination result information to the transmission unit 32 (S60). The transmission unit 32 transmits the determination result information received from the output unit 28 to the server 3 via the communication network 6.

  On the other hand, when the similarity determination unit 20 determines that the first feature signal and the second feature signal are not similar (S50: No), the similarity determination unit 20 indicates that the bug 7 is not present. Is supplied to the output unit 28. The output unit 28 supplies information indicating that the bug 7 is not present to the transmission unit 32, and the transmission unit 32 transmits the information to the server 3 (S75).

  The determination result acquisition unit 36 acquires determination result information from the bug detector 2 (S65). The determination result acquisition unit 36 supplies the acquired determination result information to the determination result storage unit 38. The determination result storage unit 38 stores determination result information. Then, the reporting unit 40 uses the determination result information stored in the determination result storage unit 38 to notify the user terminal of information indicating that the bug 7 is present (S70).

  The difference signal calculation unit 18 can also supply the calculated difference signal or absolute difference signal to the similarity determination unit 20. The total signal calculation unit 22 can also supply the calculated total signal to the average total signal calculation unit 24. The average sum signal calculation unit 24 calculates a moving average of the sum signal as an average sum signal from the plurality of sum signals. Then, the average sum signal calculation unit 24 supplies the calculated average sum signal to the similarity determination unit 20. The similarity determination unit 20 can also determine similarity based on a difference signal, an absolute difference signal, or an average sum signal (that is, a moving average of the sum signal).

  FIG. 7 shows an example of the hardware configuration of the bug detector system according to the embodiment of the present invention.

  An eavesdropper detection system 1 according to an embodiment of the present invention includes a CPU 1500, a graphic controller 1520, a memory 1530 such as a random access memory (RAM), a read-only memory (ROM), and / or a flash ROM, and a storage for storing data. A device 1540, a read / write device 1545 for reading data from and / or writing data to a recording medium, an input device 1560 for inputting data, a communication interface 1550 for transmitting / receiving data to / from an external communication device, and a CPU 1500 A chip set that connects the graphic controller 1520, the memory 1530, the storage device 1540, the read / write device 1545, the input device 1560, and the communication interface 1550 so that they can communicate with each other. And a 1510. Note that the constituent elements such as the CPU 1500 are one or more CPUs, and may include a plurality of CPUs.

  The chip set 1510 includes a memory 1530, a CPU 1500 that accesses the memory 1530 and executes predetermined processing, and a graphic controller 1520 that controls display on an external display device. Perform data passing. The CPU 1500 operates based on a program stored in the memory 1530 and controls each component. The graphic controller 1520 displays an image on a predetermined display device based on the image data temporarily stored on the buffer provided in the memory 1530.

  The chip set 1510 connects a storage device 1540, a read / write device 1545, and a communication interface 1550. The storage device 1540 stores programs and data used by the CPU 1500 of the bug detector system. The storage device 1540 is, for example, a flash memory. The read / write device 1545 reads the program and / or data from the storage medium storing the program and / or data, and stores the read program and / or data in the storage device 1540. For example, the reading / writing device 1545 acquires a predetermined program from a server on the Internet via the communication interface 1550, and stores the acquired program in the storage device 1540.

  The communication interface 1550 executes data transmission / reception with an external device via a communication network. Further, when the communication network is disconnected, the communication interface 1550 can execute data transmission / reception with an external device without going through the communication network. An input device 1560 such as a tablet or a microphone is connected to the chipset 1510 via a predetermined interface.

  The eavesdropper detection program stored in the storage device 1540 is provided to the storage device 1540 via a communication network such as the Internet or a recording medium such as a magnetic recording medium or an optical recording medium. The eavesdropper detection program stored in the storage device 1540 is executed by the CPU 1500.

  An eavesdropper detection program executed by the eavesdropper detection system 1 works on the CPU 1500 to change the eavesdropper detection system 1 into the voice input unit 10, the radio wave search unit 12, the signal acquisition unit 14 described with reference to FIGS. 1 to 6, Feature signal extraction unit 16, difference signal calculation unit 18, similarity determination unit 20, total signal calculation unit 22, average total signal calculation unit 24, continuity determination unit 26, output unit 28, information storage unit 30, transmission unit 32, It is made to function as the judgment result acquisition part 36, the judgment result storage part 38, and the report part 40.

  While the embodiments of the present invention have been described above, the embodiments described above do not limit the invention according to the claims. In addition, it should be noted that not all combinations of features described in the embodiments are necessarily essential to the means for solving the problems of the invention. Furthermore, the technical elements of the above-described embodiments may be applied independently, or may be applied by being divided into a plurality of parts such as program parts and hardware parts.

DESCRIPTION OF SYMBOLS 1 Wiretap detection system 2 Wiretap detection device 3 Server 4, 5 User terminal 6 Communication network 7 Wiretap 8 Detection area 10 Voice input part 12 Radio wave search part 14 Signal acquisition part 16 Feature signal extraction part 18 Difference signal calculation part 20 Similarity Sex determination unit 22 Total signal calculation unit 24 Average total signal calculation unit 26 Continuity determination unit 28 Output unit 30 Information storage unit 32 Transmission unit 36 Determination result acquisition unit 38 Determination result storage unit 40 Notification unit 160, 168 First characteristic signal 162 164, 166 Second feature signal 180, 182, 184 Differential signal 200, 202, 204, 206 Time 208, 210, 212, 214 Time 900 Radio wave 1500 CPU
1510 chip set 1520 graphic controller 1530 memory 1540 storage device 1545 read / write device 1550 communication interface 1560 input device

Claims (11)

An eavesdropper detection device that detects an eavesdropper that modulates the acquired sound into a radio wave and transmits it,
An audio input unit for inputting audio;
A radio wave search unit for searching for the radio wave;
A signal acquisition unit for acquiring an audio signal from the radio wave received by the radio wave search unit;
A first feature signal indicating the feature of the sound input to the sound input unit and a second feature signal indicating the feature of the sound indicated by the sound signal are extracted from the sound signal acquired by the signal acquisition unit. A feature signal extraction unit;
A first absolute difference signal obtained by positively converting a sign of a difference signal between the first feature signal and the second feature signal obtained at a first time determined in advance from a first time to a second time. And the sign of the difference signal between the first feature signal and the second feature signal obtained at a predetermined second time starting from a time different from the first time is converted to positive A difference signal calculation unit for calculating the second absolute difference signal,
A combined signal calculation unit for calculating a combined signal obtained by adding the first absolute difference signal and the second absolute difference signal;
An eavesdropper detection device comprising: a similarity determination unit that determines the similarity between the first feature signal and the second feature signal based on the magnitude of the sum signal. An average sum signal calculation unit that calculates an average sum signal obtained by dividing a sum signal obtained by summing a plurality of absolute difference signals calculated by the sum signal calculation unit by the number of the absolute difference signals used for summation is further provided. ,
The wiretap detection device according to claim 1, wherein the similarity determination unit determines the similarity based on a magnitude of the average sum signal. The first characteristic signal includes at least one signal selected from a group including a volume level of the voice input to the voice input unit, a volume level in a predetermined frequency band, and a profile waveform of the volume level;
The said 2nd characteristic signal contains at least 1 signal selected from the group containing the volume level of the audio | voice shown by the said audio | voice signal, the volume level in a predetermined frequency band, and the profile waveform of a volume level. 2. An eavesdropping detector according to 2.   The wiretap detector according to any one of claims 1 to 3, wherein the signal acquisition unit continuously acquires the audio signal from the radio wave at predetermined unit times. A continuity determination unit that determines whether the determination result of the similarity determination unit continues for a predetermined period;
The wiretap detection device according to claim 1, wherein the similarity determination unit determines the similarity based on the difference signal and a determination result of the continuity determination unit.   When the difference signal calculation unit does not match the input time of the sound input to the sound input unit and the acquisition time when the signal acquisition unit acquires the sound signal, the input time and the acquisition time are The eavesdropper detection device according to claim 1, wherein the difference signal is calculated by making them coincide with each other. An information storage unit for storing similarity determination information indicating a determination result of the similarity determination unit;
The similarity determination unit
While the voice is being input to the voice input unit, the similarity is determined and the similarity determination information is stored in the information storage unit,
When the input of the voice to the voice input unit is interrupted, the determination of the similarity is interrupted,
The determination of the similarity is restarted based on the similarity determination information stored in the information storage unit when the input of the voice to the voice input unit is restarted. The wiretap detector according to the item. An eavesdropper detection device that detects an eavesdropper that modulates the acquired sound into a radio wave and transmits it,
An audio input unit for inputting audio;
A radio wave search unit for searching for the radio wave;
A signal acquisition unit for acquiring an audio signal from the radio wave received by the radio wave search unit;
A first feature signal indicating the feature of the sound input to the sound input unit and a second feature signal indicating the feature of the sound indicated by the sound signal are extracted from the sound signal acquired by the signal acquisition unit. A feature signal extraction unit;
A difference signal calculation unit for calculating a difference signal between the first feature signal and the second feature signal;
A similarity determination unit that determines the similarity between the first feature signal and the second feature signal based on the difference signal;
An information storage unit for storing similarity determination information indicating a determination result of the similarity determination unit;
The similarity determination unit
While the voice is being input to the voice input unit, the similarity is determined and the similarity determination information is stored in the information storage unit,
When the input of the voice to the voice input unit is interrupted, the determination of the similarity is interrupted,
An eavesdropper detection device that resumes the similarity determination based on the similarity determination information stored in the information storage unit when the input of the audio to the audio input unit is resumed. An eavesdropper detection system that detects an eavesdropper that modulates acquired sound into radio waves and transmits it,
An audio input unit for inputting peripheral audio;
A radio wave search unit for searching for the radio wave;
A signal acquisition unit for acquiring an audio signal from the radio wave received by the radio wave search unit;
A first feature signal indicating the feature of the sound input to the sound input unit and a second feature signal indicating the feature of the sound indicated by the sound signal are extracted from the sound signal acquired by the signal acquisition unit. A feature signal extraction unit;
A first absolute difference signal obtained by positively converting a sign of a difference signal between the first feature signal and the second feature signal obtained at a first time determined in advance from a first time to a second time. And the sign of the difference signal between the first feature signal and the second feature signal obtained at a predetermined second time starting from a time different from the first time is converted to positive A difference signal calculation unit for calculating the second absolute difference signal,
A combined signal calculation unit for calculating a combined signal obtained by adding the first absolute difference signal and the second absolute difference signal;
A similarity determination unit that determines similarity between the first feature signal and the second feature signal based on a magnitude of the sum signal;
A determination result acquisition unit for acquiring a determination result of the similarity determination unit;
An eavesdropper detection system comprising: a notification unit that reports the presence of the eavesdropper when the determination result is information indicating that the first characteristic signal and the second characteristic signal are similar. An eavesdropper detection method for detecting an eavesdropper that modulates the acquired sound into a radio wave and transmits it,
A voice input stage in which voice is input to the voice input section;
A radio wave search stage for searching for the radio wave;
A signal acquisition step of acquiring an audio signal from the radio wave received in the radio wave search step;
Extracting a first feature signal indicating the feature of the sound input in the sound input step and a second feature signal indicating a feature of the sound indicated by the sound signal from the sound signal acquired in the signal acquisition step. A feature signal extraction stage,
A first absolute difference signal obtained by positively converting a sign of a difference signal between the first feature signal and the second feature signal obtained at a first time determined in advance from a first time to a second time. And the sign of the difference signal between the first feature signal and the second feature signal obtained at a predetermined second time starting from a time different from the first time is converted to positive A difference signal calculating step for calculating the second absolute difference signal,
A combined signal calculating step of calculating a combined signal obtained by adding the first absolute difference signal and the second absolute difference signal;
An eavesdropper detection method comprising a similarity determination step of determining the similarity between the first feature signal and the second feature signal based on the magnitude of the sum signal. An eavesdropper detection program for detecting an eavesdropper that modulates the acquired sound into radio waves and transmits it,
On the computer,
A voice input function for inputting voice to the voice input section;
A radio wave search function for searching for the radio wave;
A signal acquisition function for acquiring an audio signal from the radio wave received in the radio wave search function;
Extracting the first feature signal indicating the feature of the voice input in the voice input function and the second feature signal indicating the feature of the voice indicated by the voice signal from the voice signal acquired in the signal acquisition function A feature signal extraction function,
A first absolute difference signal obtained by positively converting a sign of a difference signal between the first feature signal and the second feature signal obtained at a first time determined in advance from a first time to a second time. And the sign of the difference signal between the first feature signal and the second feature signal obtained at a predetermined second time starting from a time different from the first time is converted to positive A difference signal calculation function for calculating the second absolute difference signal,
A combined signal calculating function for calculating a combined signal obtained by adding the first absolute difference signal and the second absolute difference signal;
An eavesdropper detection program for realizing a similarity determination function for determining the similarity between the first feature signal and the second feature signal based on the magnitude of the sum signal.