Noise source estimation device, noise source estimation method, and noise source estimation program

Abstract

To efficiently specify a noise source in noise failure.SOLUTION: A storage unit stores a correspondence relationship in which characteristic information including a plurality of characteristics of noise failure caused by electromagnetic noise, and noise source information indicating a noise source of the electromagnetic noise, are associated with each other. An estimation unit estimates a noise source causing noise failure to be estimated, on the basis of a comparison result of comparing the characteristic information included in the correspondence relationship with estimation object characteristic information including a plurality of characteristics of the noise failure to be estimated. An output unit outputs an estimation result indicating the noise source causing the noise failure.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention relates to noise source estimation technology.

Background Art

[0002] In the telephone service of a participating call, a failure caused by electromagnetic noise such as audible sound may occur during a call. Hereinafter, the failure caused by electromagnetic noise may be referred to as a noise failure. When a noise failure occurs, it is important to identify and remove the noise source.

[0003] Regarding noise failures, an electromagnetic noise measurement method that enables accurate identification of the noise source is known (see, for example, Patent Document 1).

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] Physical quantities such as the frequency of electromagnetic noise generated in a telephone service and the voltage level of a common mode voltage can be measured using a simple measuring instrument.

[0006] However, these days, the environment where telephones and the like are installed and the connection configuration of communication lines have become complicated, and it is difficult to identify the noise source only from the measurement results of physical quantities using a simple measuring instrument. For this reason, it has become common to measure physical quantities using more advanced measuring instruments or to have operators with more advanced experience identify the noise source, and the working time for identifying the noise source has been increasing.

[0007] Note that such a problem occurs not only in noise failures in telephone services, but also when identifying the noise sources of noise failures in various electronic devices or electrical equipment.

[0008] In one aspect, the present invention aims to efficiently identify the noise source of a noise failure.

Means for Solving the Problem

[0009] According to one embodiment, the noise source estimation device includes a storage unit, an estimation unit, and an output unit. The storage unit stores a correspondence relationship in which feature information including a plurality of features of a noise failure due to electromagnetic noise is associated with noise source information indicating the noise source of the electromagnetic noise.

[0010] The estimation unit estimates the noise source that causes the noise failure to be estimated based on the comparison result of comparing the feature information included in the correspondence relationship with the estimation target feature information including a plurality of features of the noise failure to be estimated. The output unit outputs an estimation result indicating the noise source that causes the noise failure.

Effect of the Invention

[0011] In one aspect, the noise source of the noise failure can be efficiently identified.

Brief Description of the Drawings

[0012]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Embodiment for Carrying out the Invention

[0013] Hereinafter, embodiments will be described in detail with reference to the drawings.

[0014] FIG. 1 shows a functional configuration example of a noise source estimation apparatus according to an embodiment. The noise source estimation apparatus 101 in FIG. 1 includes a storage unit 111, an estimation unit 112, and an output unit 113. The storage unit 111 stores a correspondence relationship in which feature information including a plurality of features of a noise failure due to electromagnetic noise is associated with noise source information indicating a noise source of the electromagnetic noise.

[0015] FIG. 2 is a flowchart showing an example of a first noise source estimation process performed by the noise source estimation apparatus 101 in FIG. 1. The estimation unit 112 estimates a noise source that causes a noise failure to be estimated based on a comparison result of comparing the feature information included in the correspondence relationship with estimation target feature information including a plurality of features of the noise failure to be estimated (step 201). The output unit outputs an estimation result indicating a noise source that causes the noise failure (step 202).

[0016] According to the noise source estimation apparatus 101 in FIG. 1, the noise source of the noise failure can be efficiently specified.

[0017] FIG. 3 shows a specific example of the noise source estimation apparatus 101 in FIG. 1. The noise source estimation apparatus 301 in FIG. 3 includes a reception unit 311, a calculation unit 312, a specification unit 313, an output unit 314, and a storage unit 315. The calculation unit 312 and the specification unit 313 correspond to the estimation unit 112 in FIG. 1. The output unit 314 and the storage unit 315 correspond to the output unit 113 and the storage unit 111 in FIG. 1, respectively.

[0018] The noise source estimation device 301 is used to estimate the noise source of a noise failure in various electronic devices or electrical equipment. For example, in the case of communication equipment used in communication services such as telephone services and Internet connection services of a subscribed telephone, noise failures may occur in electronic devices such as telephones and modems. Hereinafter, electronic devices such as telephones and modems may be simply referred to as devices.

[0019] The storage unit 315 stores the teacher information 321. The teacher information 321 includes the feature information of each of N (N is an integer of 2 or more) noise failures that have occurred in the past. The feature information of each noise failure includes K (K is an integer of 2 or more) features. In the feature information of each noise failure, noise source information indicating the noise source that is the source of the electromagnetic noise is associated as a label. The teacher information 321 corresponds to the correspondence relationship stored in the storage unit 111 of FIG. 1.

[0020] FIG. 4 shows an example of features included in the feature information of a noise failure in communication equipment. In this example, K = 10. The item represents the name of the feature, and each item has a plurality of categories. The score is a numerical value indicating each category of each item.

[0021] For example, "environment" represents the environment where the device is installed and has three categories: "indoors at home", "outside the place", and "inside the place". The scores for "indoors at home", "outside the place", and "inside the place" are 1, 2, and 3, respectively.

[0022] "Type of failure" represents the type of noise failure in the communication service and has five categories: "noise", "link down", "speed reduction", "malfunction", and "alarm". The scores for "noise", "link down", "speed reduction", "malfunction", and "alarm" are 1, 2, 3, 4, and 5, respectively.

[0023] The "faulty device" represents the device where a noise fault has occurred, and has five categories: "telephone", "modem", "aggregation device", "RSBM (Remote Subscriber Module)", and "CSM (Clock Supply Module)". The scores for "telephone", "modem", "aggregation device", "RSBM", and "CSM" are 1, 2, 3, 4, and 5 respectively.

[0024] The "type of audible sound" represents the type of audible sound that has occurred, and has five categories: "none", "boon", "bee", "buzz", and "poo". "None" indicates that no audible sound is occurring. The scores for "none", "boon", "bee", "buzz", and "poo" are 1, 2, 3, 4, and 5 respectively.

[0025] The "line" represents the communication line used in the communication service, and has five categories. The five categories are "subscribed telephone", "ADSL (Asymmetric Digital Subscriber Line)", "ISDN (Integrated Services Digital Network)", "dedicated line", and "VDSL (Very high-bit-rate Digital Subscriber Line)". The scores for "subscribed telephone", "ADSL", "ISDN", "dedicated line", and "VDSL" are 1, 2, 3, 4, and 5 respectively.

[0026] The "connection device" represents the connection device used in the communication service, and has five categories: "telephone", "modem", "aggregation device", "RSBM", and "CSM". The scores for "telephone", "modem", "aggregation device", "RSBM", and "CSM" are 1, 2, 3, 4, and 5 respectively.

[0027] "Environmental change" represents the change in the environment where the device is installed and has five categories: "New connection", "Device modification", "Purchase of electrical appliances", "Neighboring construction", and "Others". The scores for "New connection", "Device modification", "Purchase of electrical appliances", "Neighboring construction", and "Others" are 1, 2, 3, 4, and 5 respectively.

[0028] "Frequency of failure" represents the occurrence frequency of noise failures and has five categories: "Only once", "Constantly", "More than once a day", "More than once a week", and "More than once a month". The scores for "Only once", "Constantly", "More than once a day", "More than once a week", and "More than once a month" are 1, 2, 3, 4, and 5 respectively.

[0029] "Frequency of noise" represents the frequency of electromagnetic noise that caused the noise failure and has five categories: "500 Hz - 3 kHz", "7.5 - 22.5 kHz", "35 - 105 kHz", "125 - 375 kHz", and "0.5 - 1.5 MHz". The scores for "500 Hz - 3 kHz", "7.5 - 22.5 kHz", "35 - 105 kHz", "125 - 375 kHz", and "0.5 - 1.5 MHz" are 1, 2, 3, 4, and 5 respectively.

[0030] "Frequency of noise" represents the occurrence frequency of electromagnetic noise that caused the noise failure and has five categories: "Only once", "Constantly", "More than once a day", "More than once a week", and "More than once a month". The scores for "Only once", "Constantly", "More than once a day", "More than once a week", and "More than once a month" are 1, 2, 3, 4, and 5 respectively.

[0031] "Frequency of noise" and "Frequency of noise" are measured using, for example, a simple measuring instrument. Each category of "Frequency of noise" and "Frequency of noise" is an example of a feature indicating the measurement result of electromagnetic noise. The measurement result of electromagnetic noise may be other physical quantities such as the voltage level of electromagnetic noise and the voltage level of common mode voltage.

[0032] Each category of "environment", "type of failure", "failed device", "type of audible sound", "line", "connecting device", "environmental change", and "frequency of failure" is an example of a feature indicating the attributes of a noise failure due to electromagnetic noise. The number of categories for each item may be 4 or less, or may be 6 or more.

[0033] Figure 5 shows an example of characteristic information of a noise failure in communication equipment. The horizontal axis represents items, and the vertical axis represents scores. Aj (j = 1 to 10) represents the j-th item. A1 represents "environment", A2 represents "type of failure", A3 represents "failed device", A4 represents "type of audible sound", A5 represents "line", A6 represents "connecting device", A7 represents "environmental change", A8 represents "frequency of failure", A9 represents "frequency of noise", and A10 represents "intensity of noise".

[0034] The broken line 501 represents the scores of each of the items A1 to A10, corresponding to the characteristic information of a specific noise failure in communication equipment. The scores of each of the items A1 to A10 are an example of a plurality of characteristics of a noise failure due to electromagnetic noise.

[0035] Noise source information indicating the noise source of a specific noise failure is associated with the broken line 501 as a label. The noise source of a specific noise failure is, for example, a ventilation fan, a fluorescent lamp, an electric fence, a solar power generation system, a UPS (Uninterruptible Power Systems), an inverter, a power line, a rapid charger, etc.

[0036] By using characteristic information such as that in Figure 5, when a new noise failure occurs, it is possible to easily compare the characteristic information of the new noise failure with the characteristic information of noise failures that occurred in the past, and to efficiently analyze the characteristics of the new noise failure.

[0037] When a new noise failure occurs, the user inputs K features of the occurred noise failure to the noise source estimation device 301. As the K features of the occurred noise failure, for example, the scores of each item in FIG. 4 are input. The occurred noise failure is an example of the noise failure to be estimated.

[0038] The reception unit 311 receives the input K features, generates estimation target feature information 322 including the received K features, and stores it in the storage unit 315. The estimation target feature information 322 has, for example, the same data format as the feature information shown in FIG. 5. The scores of each of items A1 to A10 included in the estimation target feature information 322 are an example of a plurality of features of the noise failure to be estimated.

[0039] The calculation unit 312 calculates the similarity D(i) between the estimation target feature information 322 and the i-th (i = 1 to N) feature information included in the teacher information 321. As the similarity D(i), for example, a weighted Euclidean distance is used.

[0040] When using the weighted Euclidean distance, the calculation unit 312 calculates the difference p(j) - q(i,j) between the j-th (j = 1 to 10) feature p(j) of the estimation target feature information 322 and the j-th feature q(i,j) of the i-th feature information included in the teacher information 321. Then, the calculation unit 312 calculates the similarity D(i) by the following formula using the differences p(j) - q(i,j) calculated for each of the K features.

[0041] D(i)=(Σw(j)(p(j)-q(i,j)) 2 ) 1 / 2 (1)

[0042] w(j) represents the weight coefficient for the feature p(j), and Σ represents the sum with respect to j = 1 to 10. w(j) may be a real number in the range of, for example, 0.1 to 1. For example, w(j) for features such as "noise frequency" and "type of audible sound" with high correlation with the noise source may be set to 0.1 to increase the priority. Also, w(j) for other features may be set to 0.5 or the like according to the priority.

[0043] D(i) in Equation (1) becomes smaller as the i-th feature information included in the estimated target feature information 322 and the teacher information 321 is more similar, and becomes larger as the i-th feature information included in the estimated target feature information 322 and the teacher information 321 is more different. D(i) represents the comparison result of comparing the i-th feature information included in the estimated target feature information 322 and the teacher information 321.

[0044] Based on the calculated similarities D(1) to D(N), the specifying unit 313 specifies the feature information similar to the estimated target feature information 322 from among the feature information included in the teacher information 321. For example, the specifying unit 313 specifies, as the feature information similar to the estimated target feature information 322, the feature information having the smallest similarity D(i) among the similarities D(1) to D(N).

[0045] Next, the specifying unit 313 estimates, as the noise source that causes the generated noise failure, the noise source indicated by the noise source information associated with the specified feature information. The output unit 314 outputs an estimation result indicating the estimated noise source.

[0046] Next, a specific example of an experiment using the noise source estimation device 301 in FIG. 3 will be described. In this experiment, N = 50 and K = 10, and feature information in the same data format as in FIG. 5 is used.

[0047] w(j) for "type of audible sound" is 0.1, w(j) for "faulty device" is 0.3, and w(j) for "noise frequency" is 0.5. w(j) for other features is 1.

[0048] The generated noise failure is the link - down of ADSL at the customer's home, and the noise source is the electric fence.

[0049] The category of the "environment" of the generated noise failure is "indoors", the category of the "type of failure" is "link - down", the category of the "failed device" is "modem", and the category of the "type of audible sound" is "none". The category of the "line" is "ADSL", the category of the "connection device" is "modem", the category of the "environmental change" is "new activation", and the category of the "frequency of failure" is "more than 1 time / day".

[0050] The category of the "noise frequency" is "7.5 - 22.5 kHz", and the category of the "noise frequency" is "more than 1 time / day".

[0051] According to the results of this experiment, among the similarity degrees D(i) of the 50 pieces of feature information included in the teacher information 321, the smallest similarity degree D(i) is 0.71, and the second - smallest similarity degree D(i) is 1.22. There is only 1 piece of feature information with D(i)=0.71, and its label indicates an electric fence. There are 2 pieces of feature information with D(i)=1.22, and their labels indicate a ventilation fan and a solar power generation system.

[0052] Therefore, the electric fence indicated by the label of the feature information with D(i)=0.71 is estimated to be the noise source causing the link - down of ADSL. The estimated noise source is consistent with the actual noise source of the generated ADSL link - down.

[0053] According to the noise source estimation device 301 in FIG. 3, by using the teacher information 321, it is possible to efficiently identify the noise source of the noise failure in a short time without the need for highly - sophisticated measuring equipment or an operator with highly - sophisticated experience.

[0054] The calculation unit 312 may calculate the similarity degree D(i) by using other distances such as the Manhattan distance and the Mahalanobis distance instead of the weighted Euclidean distance.

[0055] The calculation unit 312 may calculate the similarity D(i) using, for example, the cosine similarity of vectors instead of the distance. When using the cosine similarity, D(i) increases as the i-th feature information included in the estimation target feature information 322 and the teacher information 321 becomes more similar, and decreases as the i-th feature information included in the estimation target feature information 322 and the teacher information 321 becomes more different. In this case, the specifying unit 313 specifies the feature information having the largest similarity D(i) as the feature information similar to the estimation target feature information 322.

[0056] FIG. 6 is a flowchart showing an example of the second noise source estimation process performed by the noise source estimation device 301 of FIG. 3. First, the reception unit 311 receives K features input by the user, and generates estimation target feature information 322 including the received K features (step 601).

[0057] Next, the calculation unit 312 sets 1 to the control variable i (step 602), and calculates the similarity D(i) between the i-th feature information included in the estimation target feature information 322 and the teacher information 321 (step 603).

[0058] Next, the calculation unit 312 compares i with N (step 604). If i is less than N (step 604, NO), the calculation unit 312 increments i by 1 (step 608). Then, the noise source estimation device 301 repeats the processes after step 603.

[0059] If i = N (step 604, YES), the specifying unit 313 specifies, based on the calculated similarities D(1) to D(N), the feature information similar to the estimation target feature information 322 from among the feature information included in the teacher information 321 (step 605).

[0060] Next, the specifying unit 313 estimates the noise source indicated by the noise source information associated with the specified feature information as the noise source that causes the generated noise failure (step 606). Then, the output unit 314 outputs an estimation result indicating the estimated noise source (step 607).

[0061] The configurations of the noise source estimation device 101 in FIG. 1 and the noise source estimation device 301 in FIG. 3 are merely examples, and some of the components may be omitted or changed according to the use or conditions of the noise source estimation device.

[0062] The flowcharts shown in FIGS. 2 and 6 are merely examples, and some of the processes may be omitted or changed according to the configuration or conditions of the noise source estimation device.

[0063] The features of the noise failure shown in FIGS. 4 and 5 are merely examples, and feature information may be generated using other features.

[0064] Equation (1) is merely an example, and the calculation unit 312 may calculate the similarity D(i) using another calculation formula.

[0065] FIG. 7 shows a hardware configuration example of an information processing apparatus (computer) used as the noise source estimation device 101 in FIG. 1 and the noise source estimation device 301 in FIG. 3. The information processing apparatus in FIG. 7 includes a CPU (Central Processing Unit) 701, a memory 702, an input device 703, an output device 704, an auxiliary storage device 705, a media drive device 706, and a network connection device 707. These components are hardware and are connected to each other by a bus 708.

[0066] The memory 702 is, for example, a semiconductor memory such as a ROM (Read Only Memory) or a RAM (Random Access Memory), and stores programs and data used for processing. The memory 702 may operate as the storage unit 111 in FIG. 1 or the storage unit 315 in FIG. 3.

[0067] The CPU 701 (processor) operates as the estimation unit 112 in FIG. 1 by executing a program using, for example, the memory 702. The CPU 701 also operates as the reception unit 311, the calculation unit 312, and the specification unit 313 in FIG. 3 by executing a program using the memory 702.

[0068] The input device 703 is, for example, a keyboard, a pointing device, etc., and is used for inputting instructions or information from a user or an operator. The output device 704 is, for example, a display device, a printer, a speaker, etc., and is used for outputting inquiries or instructions to the user or operator and the processing results. The output device 704 may operate as the output unit 113 in FIG. 1 or the output unit 314 in FIG. 3. The processing result may be an estimation result indicating a noise source.

[0069] The auxiliary storage device 705 is, for example, a magnetic disk device, an optical disk device, a magneto-optical disk device, a tape device, etc. The auxiliary storage device 705 may be a hard disk drive or an SSD (Solid State Drive). The information processing apparatus can store programs and data in the auxiliary storage device 705 and load them into the memory 702 for use. The auxiliary storage device 705 may operate as the storage unit 111 in FIG. 1 or the storage unit 315 in FIG. 3.

[0070] The medium drive device 706 drives the portable recording medium 709 and accesses the recorded content. The portable recording medium 709 is a memory device, a flexible disk, an optical disk, a magneto-optical disk, etc. The portable recording medium 709 may be a CD-ROM (Compact Disk Read Only Memory), a DVD (Digital Versatile Disk), a USB (Universal Serial Bus) memory, etc. A user or an operator can store programs and data in the portable recording medium 709 and load them into the memory 702 for use.

[0071] Thus, a computer-readable recording medium for storing a program and data used in processing is a physical (non-transitory) recording medium such as a memory 702, an auxiliary storage device 705, or a portable recording medium 709.

[0072] The network connection device 707 is a communication circuit that is connected to a communication network such as a WAN (Wide Area Network) or a LAN (Local Area Network) and performs data conversion associated with communication. The information processing device can receive a program and data from an external device via the network connection device 707 and load them into the memory 702 for use. The network connection device 707 may operate as the output unit 113 in FIG. 1 or the output unit 314 in FIG. 3.

[0073] Note that the information processing device does not necessarily need to include all the components in FIG. 7, and it is also possible to omit some components according to the application or conditions. For example, if an interface with a user or an operator is not required, the input device 703 and the output device 704 may be omitted. If a portable recording medium 709 or a communication network is not used, the medium drive device 706 or the network connection device 707 may be omitted.

[0074] Although the disclosed embodiments and their advantages have been described in detail, those skilled in the art will be able to make various changes, additions, and omissions without departing from the scope of the invention clearly described in the claims.

Explanation of Reference Numerals

[0075] 101, 301 Noise source estimation device 111, 315 Storage unit 112 Estimation unit 113, 314 Output unit 311 Reception unit 312 Calculation unit 313 Identification unit 321 Teacher information 322 Estimation target feature information 501 Broken line 701 CPU 702 Memory 703 Input device 704 Output device 705 Auxiliary storage device 706 Media drive 707 Network connection device 708 Bus 709 Portable recording medium

Claims

1. A storage unit that stores a correspondence relationship in which feature information including a plurality of features of a noise failure due to electromagnetic noise and noise source information indicating a noise source of the electromagnetic noise are associated with each other; An estimation unit that estimates a noise source that causes the noise failure of the estimation target based on a comparison result of comparing the feature information included in the correspondence relationship with estimation target feature information including a plurality of features of the noise failure of the estimation target; An output unit that outputs an estimation result indicating a noise source that causes the noise failure; A noise source estimation device, characterized by comprising the above.

2. The estimation unit uses the plurality of features of the estimation target feature information and the plurality of features of the feature information included in the correspondence relationship to calculate a similarity between the estimation target feature information and the feature information included in the correspondence relationship, and based on the similarity, identify feature information similar to the estimation target feature information from among the feature information included in the correspondence relationship, and estimate the noise source indicated by the noise source information associated with the identified feature information as the noise source that causes the noise failure. The noise source estimation device according to Claim 1.

3. The estimation unit calculates a difference between each of the plurality of features of the estimation target feature information and a feature corresponding to each of the plurality of features among the plurality of features of the feature information included in the correspondence relationship, and calculates the similarity using the differences calculated for each of the plurality of features. The noise source estimation device according to Claim 2.

4. The plurality of features of the noise failure due to electromagnetic noise include a feature indicating a measurement result of the electromagnetic noise and a feature indicating an attribute of the noise failure due to electromagnetic noise. The noise source estimation device according to any one of Claims 1 to 3.

5. The noise failure due to electromagnetic noise is a failure related to communication equipment, and the features indicating the attributes of the noise failure are the environment in which the electronic equipment of the communication equipment is installed, changes in the environment in which the electronic equipment of the communication equipment is installed, the type of the noise failure, the failed device of the noise failure, the type of audible sound of the noise failure, the type of communication line used in the communication, the connection device used in the communication, or the occurrence frequency of the noise failure. The noise source estimation device according to Claim 4.

6. Based on the comparison result of comparing the feature information including a plurality of features of the noise failure due to electromagnetic noise with the feature information included in the correspondence relationship in which the noise source information indicating the noise source of the electromagnetic noise is associated, and the estimated target feature information including a plurality of features of the noise failure to be estimated, estimate the noise source that causes the noise failure to be estimated, Output an estimation result indicating the noise source that causes the noise failure, A noise source estimation method characterized in that a computer executes the process.

7. Based on the comparison result of comparing the feature information including a plurality of features of the noise failure due to electromagnetic noise with the feature information included in the correspondence relationship in which the noise source information indicating the noise source of the electromagnetic noise is associated, and the estimated target feature information including a plurality of features of the noise failure to be estimated, estimate the noise source that causes the noise failure to be estimated, Output an estimation result indicating the noise source that causes the noise failure, A noise source estimation program for causing a computer to execute the process.

Images