JP2005292036A - Noise source identification system and method - Google Patents

Abstract

There is provided a noise source identification system capable of performing analysis for identifying a noise source of noise sound generated from a device or the like at a remote place and notifying the analysis result.
A first mobile phone terminal 1 and a second mobile phone terminal 2 are put into a call state, and the first mobile phone terminal 1 uses a built-in microphone to collect noise sounds collected from a telephone D by a mobile phone network (not shown). 2) to the second mobile phone terminal 2. The second mobile phone terminal 2 reproduces the transmitted noise sound from the built-in speaker. The analysis device 3 collects and processes the reproduced noise sound with the microphone M, extracts the frequency latent in the noise sound based on the linear prediction method, and specifies the extracted frequency and / or the frequency. The type and / or characteristic of the noise source is notified to the first mobile phone terminal 1 via the second mobile phone terminal 2 and the mobile phone network.
[Selection] Figure 1

Description

  The present invention relates to a noise source identification system and method for identifying a noise source from a noise sound.

  Conventionally, there is a time analysis method using a linear prediction method as a method for identifying a noise source from a noise sound generated from a device. In this method, first, the noise sound time-series data is generated by sampling the waveform of the noise sound. Then, the time series data is subjected to time frequency analysis, thereby extracting the frequency component of the time series data and identifying the noise source.

  As a method for extracting the frequency of time series data, there is a Fourier transform method, and recently, a linear prediction method and a wavelet transform method have been proposed.

  The Fourier transform method examines a correlation coefficient between time series data and a sine wave having a predetermined frequency. On the other hand, the linear prediction method obtains a prediction coefficient of time series data that is a measurement signal and, as a result, extracts a characteristic frequency using the prediction coefficient. The wavelet transform method examines the correlation between a pulse having a certain pulse width and a certain time delay and time-series data.

  In these methods, there is a problem of S / N between white noise and signals mixed in time-series data, and there is a problem that the formant frequency extraction accuracy is remarkably deteriorated when the noise increases.

  In addition, measurement signals measured in an actual environment are often affected by electromagnetic induction from a commercial power supply, and there is a problem that the analysis result fluctuates due to the influence.

  Furthermore, in these methods, noise is characterized by two-dimensionally displaying a frequency obtained from time series data and its level or correlation value.

  Below, an example of the frequency analysis method by the linear prediction method is shown.

Formant analysis, which is one of the frequency analysis methods based on the linear prediction method, is known as an analysis method for obtaining characteristic or periodic frequency components present in speech time-series data (not necessarily limited to speech). ing. There are several methods for frequency analysis. Here, a linear prediction analysis method will be described. Specifically, in order to extract frequency information indicating the characteristics of time series data {x (n): n = 0, 1, 2,... N, T: sample interval}, prediction for the following time series data is performed. Think about the value.

Note that a _i (i = 1, 2,..., P) is also called a prediction coefficient. The concept used in equation (1) is similar to the regression analysis method performed in multivariate analysis or the like, and predicts time-series data at a certain time from past time-series data. Note that p is the number of terms in the prediction coefficient. The prediction order depends on the number of frequency components included in the time-series data, but usually the prediction order p is often selected to be about 8-20.

At this time, in order to obtain an unknown prediction coefficient, a norm I which is a difference between the time series data and the predicted value is defined as follows.

Here, N is the total number of time series data.

Note that simultaneous linear equations relating to the prediction coefficient can be obtained by performing partial differential variation on the expression (2) with the prediction coefficient. The prediction coefficient can be obtained by solving the simultaneous linear equations. In addition, the frequency which time series data has can be obtained by calculating | requiring the solution of the high-order algebraic equation regarding the following prediction coefficients from the obtained prediction coefficient.

However,

Note that ω is an angular frequency, and the solution of equation (3) provides p frequencies (sine wave frequency and attenuation frequency).
JP 2000-000000 A

  In the time analysis method using the linear prediction method as described above, an analysis apparatus is brought into a place where the equipment is located, and noise sound generated from the equipment is measured and analyzed on the spot. Therefore, there has been a problem that the measurement system becomes large and noise noise cannot be identified quickly.

  It is an object of the present invention to provide a noise source identification system and method capable of performing analysis for identifying a noise source of noise sound generated from a device or the like at a remote place and notifying the analysis result.

  In order to solve the above problems, the present invention of claim 1 is directed to a first mobile phone terminal that collects and transmits a noise sound, a second mobile phone terminal that receives the noise sound from the first mobile phone terminal, and the reception A noise source identification system using an analysis device for analyzing the generated noise sound, wherein the first mobile phone terminal transmits the collected noise sound to the second mobile phone terminal via a mobile phone network, The second mobile phone terminal reproduces the received noise sound, and the analysis device collects the reproduced noise sound with a microphone and extracts a frequency latent in the noise sound based on a linear prediction method. The extracted frequency and / or the type of noise source specified by the frequency and / or its characteristics are notified to the first mobile phone terminal via the second mobile phone terminal and the mobile phone network. The solutions have a noise source identification system, characterized and.

  The present invention of claim 2 is characterized in that the analysis device notifies a predetermined computer via the Internet instead of notifying the first mobile phone terminal. The system is the solution.

  The present invention of claim 3 is characterized in that, instead of the first and second mobile phone terminals, a first wireless communication device and a second wireless communication device capable of performing wireless communication directly are used. The noise source identification system according to claim 1 or 2 is used as a solving means.

  The present invention of claim 4 is a noise source identification system using a mobile phone terminal that collects and transmits a noise sound and an analysis device that receives and analyzes the noise sound. The sound is transmitted to the analysis device via the Internet, and the analysis device extracts a frequency latent in the received noise sound based on a linear prediction method and is specified by the extracted frequency and / or the frequency. The noise source identification system is characterized in that the type and / or characteristics of the noise source is notified to the mobile phone terminal via the Internet.

  According to a fifth aspect of the present invention, in the noise source identification system according to the fourth aspect, the analysis device notifies a predetermined computer via the Internet instead of notifying the mobile phone terminal. Let it be a solution.

  The present invention of claim 6 is a noise source identification system using a computer that collects and transmits a noise sound and an analysis device that receives and analyzes the noise sound, and the computer collects the noise sound with a microphone. And the collected noise sound is transmitted to the analysis device via the Internet, and the analysis device extracts a frequency latent in the received noise sound based on a linear prediction method, and the extracted frequency and / or The noise source identification system that notifies the computer of the type and / or characteristics of the noise source specified by the frequency via the Internet is used as the solving means.

  According to a seventh aspect of the present invention, there is provided the noise source identification system according to the sixth aspect, wherein the communication line between the computer and the Internet is constituted by a wireless LAN device.

  The present invention according to claim 8 includes a database in which types of known noise sources and / or characteristics thereof and frequencies extracted by noise sounds from the noise sources are associated in advance. 8. The method according to any one of claims 1 to 7, further comprising: extracting a predetermined frequency based on a linear prediction method and comparing the extracted frequency with a frequency in the database to specify a type and / or characteristic of a noise source. The noise source identification system described above is used as a solution.

  The present invention of claim 9 is a first mobile phone terminal that collects and transmits a noise sound, a second mobile phone terminal that receives the noise sound from the first mobile phone terminal, and an analysis that analyzes the received noise sound. A noise source identification method performed by a noise source identification system using a device, wherein the first mobile phone terminal transmits a collected noise sound to the second mobile phone terminal via a mobile phone network, and 2 The mobile phone terminal reproduces the received noise sound, and the analysis device collects the reproduced noise sound with a microphone and extracts a frequency latent in the noise sound based on a linear prediction method. Notifying the first mobile phone terminal of the extracted frequency and / or the type and / or characteristics of the noise source specified by the frequency via the second mobile phone terminal and the mobile phone network. The solutions have a noise source identification method characterized.

  According to a tenth aspect of the present invention, in the noise source identification according to the ninth aspect, the analysis device notifies a predetermined computer via the Internet instead of notifying the first mobile phone terminal. Let the method be the solution.

  The present invention of claim 11 is characterized in that, instead of the first and second mobile phone terminals, a first wireless communication device and a second wireless communication device capable of directly performing wireless communication are used. The noise source identifying method according to claim 9 or 10 is used as a solving means.

  The present invention of claim 12 is a noise source identification method performed by a noise source identification system using a mobile phone terminal that collects and transmits a noise sound and an analysis device that receives and analyzes the noise sound. The terminal transmits the collected noise sound to the analysis device via the Internet, and the analysis device extracts a frequency latent in the received noise sound based on a linear prediction method, and the extracted frequency and / or Alternatively, the noise source identification method is characterized in that the type and / or characteristic of the noise source specified by the frequency is notified to the mobile phone terminal via the Internet.

  According to a thirteenth aspect of the present invention, in the noise source identification method according to the thirteenth aspect, the analysis device notifies a predetermined computer via the Internet instead of notifying the mobile phone terminal. Let it be a solution.

  The present invention of claim 14 is a noise source identification method performed by a noise source identification system using a computer that collects and transmits a noise sound and an analysis device that receives and analyzes the noise sound. The sound is collected by a microphone and the collected noise sound is transmitted to the analysis device via the Internet. The analysis device extracts a frequency latent in the received noise sound based on a linear prediction method, and the extraction The noise source identification method is characterized by notifying the computer of the frequency and / or the type and / or characteristics of the noise source specified by the frequency via the Internet.

  According to a fifteenth aspect of the present invention, there is provided the noise source identification method according to the fourteenth aspect, wherein the communication line between the computer and the Internet is constituted by a wireless LAN device.

  According to the present invention of claim 16, the analysis device is provided when a database in which types of known noise sources and / or characteristics thereof and frequencies extracted by noise sounds from the noise sources are associated in advance is provided. The frequency of noise noise is extracted based on a linear prediction method, and the extracted frequency is compared with the frequency of the database to identify the type of noise source and / or its characteristics. The noise source identification method according to any one of 9 to 15 is used as a solving means.

  The present invention of claim 17 is a solving means comprising a computer program for causing a noise source identification system to execute the noise source identification method according to any of claims 9 to 16.

  The eighteenth aspect of the present invention provides a recording medium storing a computer program for causing a noise source identification system to execute the noise source identification method according to any of the ninth to sixteenth aspects.

  ADVANTAGE OF THE INVENTION According to this invention, the noise source identification system which can perform the analysis for identifying the noise source of the noise sound which generate | occur | produces from an apparatus etc. in a remote place, and can notify the analysis result can be provided. As a result, a noise source can be identified from a remote location. Therefore, the noise source can be quickly identified without making the measurement system large. In addition, it is possible to support the cause of the noise failure from a remote location.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings.

  FIG. 1 is a diagram showing a configuration of a noise source identification system according to the first embodiment of the present invention.

  The system includes a first mobile phone terminal 1 that collects and transmits a noise sound generated from a telephone D by a noise source N, a second mobile phone terminal 2 that receives the noise sound from the first mobile phone terminal 1, and the reception. It is the structure used with the analysis apparatus 3 which analyzes the noise noise produced. The analysis device 3 is connected to a microphone M that collects a noise sound reproduced by the second mobile phone terminal 2. The noise source N is, for example, an electronic device, and noise noise is generated from the telephone set D by superimposing noise on the telephone line connected to the telephone set D.

  In this system, first, the first mobile phone terminal 1 and the second mobile phone terminal 2 are put into a call state, and the first mobile phone terminal 1 carries the noise sound collected from the telephone D with the built-in microphone. The data is transmitted to the second mobile phone terminal 2 via a telephone network (not shown). The second mobile phone terminal 2 reproduces the transmitted noise sound from the built-in speaker. The analysis device 3 collects and processes the reproduced noise sound with the microphone M, and a frequency (two or more frequencies or multiple frequencies (frequency group) may be used.) The same applies hereinafter. Extraction based on the linear prediction method and the extracted frequency and / or the type of noise source specified by the frequency and / or its characteristics are first transmitted via the second mobile phone terminal 2 and the mobile phone network. The mobile phone terminal 1 is notified. At this time, the analysis apparatus 3 can perform frequency extraction based on the linear prediction method by the method described in the background art. The same applies to other embodiments.

  As shown in FIG. 2, the analysis device 3 may notify the computer 4 via the Internet, for example, by e-mail instead of notifying the first mobile phone terminal 1. Of course, the notification to the computer 4 may be performed in another embodiment.

  Further, instead of the first and second mobile phone terminals, a first radio communication device and a second radio communication device capable of directly performing wireless communication may be used.

  FIG. 3 is a diagram showing a configuration of a noise source identification system according to the second exemplary embodiment of the present invention. Here, the same code | symbol is provided with respect to the element used by 1st Embodiment. In addition, redundant description regarding these elements is omitted.

  This system is configured using a mobile phone terminal 1 that collects and transmits a noise sound and an analysis device 3 that receives and analyzes the noise sound.

  In this system, the mobile phone terminal 1 transmits the noise sound collected from the telephone D with the built-in microphone to the analysis device 3 via the Internet. The analysis device 3 receives and processes a noise sound, extracts a potential frequency based on a linear prediction method, and also extracts the extracted frequency and / or the type of noise source specified by the frequency and / or the noise source. The feature is notified to the mobile phone terminal 1 via the Internet.

  FIG. 4 is a diagram showing a configuration of a noise source identification system according to the third exemplary embodiment of the present invention. Here, the same code | symbol is provided with respect to the element used by 1st or 2nd embodiment. In addition, redundant description regarding these elements is omitted.

  This system is configured using a computer 5 that collects and transmits a noise sound and an analysis device 3 that receives and analyzes the noise sound. The computer 5 is connected to a microphone M that collects noise sound from the telephone D.

  In this system, the computer 5 collects the noise sound from the telephone D with the microphone M and transmits the collected noise sound to the analysis device 3 via the Internet. The analysis device 3 receives and processes a noise sound, extracts a potential frequency based on a linear prediction method, and also extracts the extracted frequency and / or the type of noise source specified by the frequency and / or the noise source. The feature is notified to the computer 5 via the Internet.

  Note that a communication line between the computer 5 and the Internet may be configured by a wireless LAN device.

  In the first to third embodiments, the following configuration can be adopted. In other words, a database in which the type and / or characteristics of a known noise source and the frequency extracted from the noise sound from the noise source are associated in advance is provided in the fixed magnetic storage device (hard disk) on the analysis device 3 side. The analysis apparatus 3 extracts the frequency latent in the transmitted noise sound based on the linear prediction method, and compares the extracted frequency with the frequency of the database to identify the type of noise source and / or its characteristics. You can

  The computer program that causes the noise source identification system of each embodiment to execute the above-described processing is stored in a computer-readable recording medium such as a semiconductor memory, a magnetic disk, an optical disk, a magneto-optical disk, or a magnetic tape, It can be widely distributed by being transmitted via a communication network such as the Internet.

It is a figure which shows the structure of the noise source identification system which concerns on the 1st Embodiment of this invention. It is a figure which shows the other structure of the noise source identification system which concerns on the 1st Embodiment of this invention. It is a figure which shows the structure of the noise source identification system which concerns on the 2nd Embodiment of this invention. It is a figure which shows the structure of the noise source identification system which concerns on the 3rd Embodiment of this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 1st mobile telephone terminal 2 2nd mobile telephone terminal 3 Analysis apparatus 4,5 Computer D Telephone M Microphone N Noise source

Claims (18)

Noise source identification using a first mobile phone terminal that collects and transmits a noise sound, a second mobile phone terminal that receives the noise sound from the first mobile phone terminal, and an analysis device that analyzes the received noise sound A system,
The first mobile phone terminal transmits the collected noise sound to the second mobile phone terminal via a mobile phone network;
The second mobile phone terminal reproduces the received noise sound,
The analysis device collects the reproduced noise sound with a microphone, extracts a frequency latent in the noise sound based on a linear prediction method, and extracts the extracted frequency and / or noise specified by the frequency. A noise source identification system for notifying the first mobile phone terminal of the type and / or characteristics of the source via the second mobile phone terminal and a mobile phone network.   2. The noise source identification system according to claim 1, wherein the analysis device notifies a predetermined computer via the Internet instead of notifying the first mobile phone terminal.   3. The noise according to claim 1, wherein a first radio communication device and a second radio communication device capable of performing direct wireless communication are used instead of the first and second mobile phone terminals. Source identification system. A noise source identification system using a mobile phone terminal that collects and transmits noise sound and an analysis device that receives and analyzes the noise sound,
The mobile phone terminal transmits the collected noise sound to the analysis device via the Internet,
The analysis device extracts a frequency latent in the received noise sound based on a linear prediction method, and extracts the extracted frequency and / or the type of noise source specified by the frequency and / or its characteristics on the Internet. A noise source identification system characterized by notifying the mobile phone terminal via   5. The noise source identification system according to claim 4, wherein the analysis device notifies a predetermined computer via the Internet instead of notifying the mobile phone terminal. A noise source identification system using a computer that collects and transmits noise sound and an analysis device that receives and analyzes the noise sound,
The computer collects noise sound with a microphone and transmits the collected noise sound to the analysis device via the Internet.
The analysis device extracts a frequency latent in the received noise sound based on a linear prediction method, and extracts the extracted frequency and / or the type of noise source specified by the frequency and / or its characteristics on the Internet. A noise source identification system for notifying the computer via   7. The noise source identification system according to claim 6, wherein a communication line between the computer and the Internet is constituted by a wireless LAN device. A database in which types of known noise sources and / or characteristics thereof and frequencies extracted by noise sounds from the noise sources are associated in advance;
The analysis apparatus extracts a frequency of a noise sound based on a linear prediction method and compares the extracted frequency with a frequency of the database to identify a type of noise source and / or a feature thereof. The noise source identification system according to any one of claims 1 to 7. Noise source identification using a first mobile phone terminal that collects and transmits a noise sound, a second mobile phone terminal that receives the noise sound from the first mobile phone terminal, and an analysis device that analyzes the received noise sound A noise source identification method performed by the system,
The first mobile phone terminal transmits the collected noise sound to the second mobile phone terminal via a mobile phone network;
The second mobile phone terminal reproduces the received noise sound,
The analysis device collects the reproduced noise sound with a microphone, extracts a frequency latent in the noise sound based on a linear prediction method, and extracts the extracted frequency and / or noise specified by the frequency. A noise source identification method comprising: notifying the first mobile phone terminal of a source type and / or characteristics thereof via the second mobile phone terminal and a mobile phone network.   The noise source identification method according to claim 9, wherein the analysis apparatus notifies a predetermined computer via the Internet instead of notifying the first mobile phone terminal.   11. The noise according to claim 9, wherein a first wireless communication device and a second wireless communication device capable of performing wireless communication directly are used instead of the first and second mobile phone terminals. Source identification method. A noise source identification method performed by a noise source identification system using a mobile phone terminal that collects and transmits noise sound and an analysis device that receives and analyzes the noise sound,
The mobile phone terminal transmits the collected noise sound to the analysis device via the Internet,
The analysis device extracts a frequency latent in the received noise sound based on a linear prediction method, and extracts the extracted frequency and / or the type of noise source specified by the frequency and / or its characteristics on the Internet. A noise source identification method comprising: notifying the mobile phone terminal via a mobile phone terminal.   13. The noise source identification method according to claim 12, wherein the analysis device notifies a predetermined computer via the Internet instead of notifying the mobile phone terminal. A noise source identification method performed by a noise source identification system using a computer that collects and transmits noise sound and an analysis device that receives and analyzes the noise sound,
The computer collects noise sound with a microphone and transmits the collected noise sound to the analysis device via the Internet.
The analysis device extracts a frequency latent in the received noise sound based on a linear prediction method, and extracts the extracted frequency and / or the type of noise source specified by the frequency and / or its characteristics on the Internet. A noise source identifying method comprising: notifying the computer via a computer.   15. The noise source identification method according to claim 14, wherein a communication line between the computer and the Internet is constituted by a wireless LAN device. When a database in which the types of known noise sources and / or their characteristics are associated with the frequencies extracted from the noise sound from the noise sources in advance is provided,
The analysis apparatus extracts a frequency of a noise sound based on a linear prediction method and compares the extracted frequency with a frequency of the database to identify a type of noise source and / or a feature thereof. The noise source identification method according to claim 9.   A computer program for causing a noise source identification system to execute the noise source identification method according to any one of claims 9 to 16. A recording medium storing a computer program for causing a noise source identification system to execute the noise source identification method according to claim 9.

Images