Computer system, noise isolation method, and program

Abstract

To improve noise separation in measurement of one channel.SOLUTION: A computer system comprises: one or more processors; and one or more memory resources. The one or more processors acquire a signal which includes electromagnetic noise; define a plurality of partial waves which is included in the signal; estimate a generation cycle generated by the partial wave in the signal, for at least a portion of the plurality of partial waves; classify at least a portion of the plurality of partial waves into a plurality of clusters for each generation cycle; and perform filtering in which the partial wave having non-similar characteristics with the other partial wave, of the plurality of partial waves belonging to the same cluster is excluded from the cluster.SELECTED DRAWING: Figure 2

Description

Technical Field

[0001] The present invention relates to a computer system, a noise separation method, and a program.

Background Art

[0002] Electronic devices may malfunction due to various electromagnetic noises radiated from electronic devices, trains, automobiles, etc. Therefore, it is desirable to identify the electromagnetic noise sources emitting electromagnetic noises and take measures against electromagnetic noises.

[0003] As a technique for identifying an electromagnetic noise source, modulation frequency data of a plurality of electromagnetic noise sources is acquired in advance, and based on each modulation frequency data and the actually measured modulation frequency data, an electromagnetic noise source emitting electromagnetic noise is identified (Patent Document 1). However, this technique can only be applied when the modulation frequency data of the electromagnetic noise source is known, and cannot be applied when the location of the electromagnetic noise source is unknown and its modulation frequency data cannot be acquired.

[0004] On the other hand, in a technique called BSS (Blind Source Separation), the period etc. of an electromagnetic noise source can be identified without using preliminary information regarding the electromagnetic noise source.

Prior Art Documents

Patent Documents

[0005]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0006] In BSS technology, it is preferable to obtain information leading to noise source identification by measuring a single channel, in order to easily identify the noise source. However, in the technology described in Patent Document 1, there is room for improvement in noise separation when the frequency bands of multiple electromagnetic noise sources overlap in a single-channel measurement.

[0007] This invention has been made in view of these circumstances and aims to improve noise separation in single-channel measurement. [Means for solving the problem]

[0008] To solve the above problems, a computer system according to one aspect of the present invention is a computer system having one or more processors and one or more memory resources, wherein the one or more processors acquire a signal including electromagnetic noise, define a plurality of partial waves included in the signal, estimate the generation period in which the partial waves occur in the signal for at least a portion of the plurality of partial waves, classify at least a portion of the plurality of partial waves into a plurality of clusters according to the generation period, and perform filtering to exclude from the cluster a partial wave that does not have similar characteristics to other partial waves among the plurality of partial waves belonging to the same cluster. [Effects of the Invention]

[0009] According to the present invention, noise separation in single-channel measurement can be improved.

[0010] Other issues, configurations, and effects not mentioned above will be clarified by the following description of the embodiments. [Brief explanation of the drawing]

[0011] [Figure 1] Figure 1 is a configuration diagram of an example of a computer system according to the first embodiment. [Figure 2] Figure 2 is a flowchart showing an example of a process performed by the computer system in the first embodiment. [Figure 3]Figure 3(a) is a schematic diagram of an example of signal data in the first embodiment, and Figure 3(b) is a schematic diagram of an example of a partial wave in the first embodiment. [Figure 4] Figure 4 is a schematic diagram showing an example of a digitized signal generated by the processor in step S3 of the first embodiment. [Figure 5] Figure 5 is a schematic diagram showing an example of the sampling method in step S3 of the first embodiment. [Figure 6] Figure 6 is a plot of an example of the log probability in the first embodiment for each candidate period. [Figure 7] Figure 7 is a schematic diagram showing an example of the sampling method in step S4 of the first embodiment. [Figure 8] Figure 8 shows an example of a visualization of the total number k in step S4 of the first embodiment. [Figure 9] Figure 9 shows an example of a visualization of the log probability P(k) calculated from the total number k in Figure 8 in the first embodiment. [Figure 10] Figure 10 shows an example of a visualization of the negative log probability P(k) for "Cluster 2" with a different occurrence period than that shown in Figure 9, according to the first embodiment. [Figure 11] Figure 11 shows an example of a visualization of the negative log probability P(k) for "Cluster 3," which has a different occurrence period than both Figure 9 and Figure 10, in the first embodiment. [Figure 12] Figure 12 shows an example of a waveform obtained by superimposing the subwaves that belonged to "Cluster 1" before filtering in step S5 of the first embodiment. [Figure 13] Figure 13 shows an example of a waveform obtained by superimposing the individual partial waves remaining in "Cluster 1" after filtering in step S5 of the first embodiment. [Figure 14] Figure 14 shows an example of the frequency spectrum of each partial wave before filtering in step S5 of the first embodiment. [Figure 15]FIG. 15 is a diagram showing an example of the frequency spectrum of each partial wave after filtering in step S5 of the first embodiment. [Figure 16] FIG. 16 is a flowchart showing an example of the processing executed by the computer system in the second embodiment. [Figure 17] FIG. 17 is a flowchart showing an example of the processing executed by the computer system in the third embodiment. [Figure 18] FIG. 18 is a flowchart showing an example of the processing executed by the computer system in the fourth embodiment. [Figure 19] FIG. 19 is a configuration diagram of an example of the computer system according to the fifth embodiment. [Figure 20] FIG. 20 is a configuration diagram of an example of the computer system according to the sixth embodiment.

MODE FOR CARRYING OUT THE INVENTION

[0012] Hereinafter, an embodiment of the present invention will be described based on the drawings. In all the drawings for explaining the embodiment, the same members are generally denoted by the same reference numerals, and the repeated explanations thereof are omitted as appropriate. Further, in the following embodiments, it is needless to say that the constituent elements (including element steps, etc.) are not necessarily essential except in cases where it is particularly明示 or considered to be clearly essential in principle. Further, when it is said that "consisting of A", "comprising A", "having A", "including A", it is needless to say that other elements are not excluded except in cases where it is particularly明示 that only that element is involved. Similarly, in the following embodiments, when referring to the shape, positional relationship, etc. of the constituent elements, etc., it is to include those that are substantially近似 or similar to the shape, etc. except in cases where it is particularly明示 and cases where it is clearly not so in principle.

[0013] <First Embodiment> FIG. 1 is a configuration diagram of an example of the computer system according to the present embodiment.

[0014] Computer system 100 performs data generation, transmission, reception, and various other processing tasks. The processor reads processing programs stored in memory resources, and the processor then executes the processing according to the program. Computer system 100 is a computer such as a personal computer, tablet terminal (computer), smartphone, server computer, blade server, or cloud server, and is a system that includes at least one of these computers. That is, computer system 100 also includes a system that includes, for example, a cloud server and a display computer (for example, a tablet terminal or smartphone). Furthermore, a controller that controls or manages some kind of device, including a processor and memory resources, is also an example of computer system 100.

[0015] Specifically, as shown in Figure 1, the computer system 100 includes one or more processors 101, one or more UI (User Interface) devices 102, one or more NI (Network Interface) devices 103, and one or more memory resources 104. The computer system 100 may also include other components. Furthermore, the processors 101, UI devices 102, NI devices 103, and memory resources 104 are interconnected via a bus 109.

[0016] The processor 101 is a computing device that reads the processing program 105 stored in the memory resource 104 and executes each process of the noise separation method according to this embodiment. The processor 101 may be a microprocessor, CPU (Central Processing Unit), GPU (Graphics Processing Unit), FPGA (Field Programmable Gate Array), quantum processor, or other computing semiconductor device.

[0017] The memory resource 104 is a storage device that stores the processing program 105 and signal data 106, and examples include non-volatile memory and / or volatile memory. Examples of volatile memory include RAM (Random Access Memory) and ROM (Read Only Memory). Examples of non-volatile memory may be rewritable storage media such as flash memory, hard disks, or SSDs (Solid State Drives), and may also be USB (Universal Serial Bus) memory, memory cards, and hard disks. In addition, RAM such as MRAM (Magnetoresistive RAM), PRAM (Phase Change RAM), and ReRAM (Resistive RAM) may be considered non-volatile memory. The processor 101 may also provide a service to distribute the processing program 105 stored in the memory resource 104 to other computers.

[0018] The UI device 102 is an input device that inputs user (or operator) instructions to the computer system 100, and an output device that outputs information generated by the computer system 100. Input devices include, for example, keyboards, touch panels, pointing devices such as mice, and voice input devices such as microphones. Output devices include, for example, displays, printers, and speech synthesizers. Unless otherwise specified below, it is assumed that information input and output between the computer system 100 and the user is performed via the UI device 102. The UI device 102 may consist solely of an input device or solely of an output device.

[0019] NI device 103 is a communication device that communicates information with external devices. NI device 103 communicates information with external devices via a predetermined communication network (not shown), such as the Internet or a LAN (Local Area Network). Unless otherwise specified below, it is assumed that information communication between the computer system 100 (or processor 101) and external devices is performed via NI device 103.

[0020] The computer system 100 acquires signal data 106 containing electromagnetic noise by executing a processing program 105 and stores it in a memory resource 104. The signal data 106 is time-series data of a signal containing electromagnetic noise. For example, the signal data 106 is the time-series data of the voltage output from an oscilloscope (not shown) after inputting the electromagnetic field strength measured by a sensor (not shown) into the oscilloscope. The sensor (not shown) that measures the electromagnetic field strength is, for example, an antenna, a current probe, and a voltage probe. Other equipment besides an oscilloscope (e.g., an amplifier, a DA converter, a control CPU, or a combination thereof) may be used as long as time-series data of a signal containing electromagnetic noise can be obtained. Furthermore, the signal data 106 may be data that has undergone various filtering and calibration processes on the time-series data.

[0021] In this specification, "measurement of one channel" means at least "using or separating one signal data 106 related to electromagnetic field strength output by one sensor." When multiple signal data 106 are used with multiple sensors at discrete positions and measurement directions, hints about the direction of the noise source or directions can be obtained by triangulation, etc. However, when "one signal data 106 output by one sensor" is used, such hints about directions cannot be obtained, so the technology of this embodiment is useful. The application of a multi-role sensor capable of sensing electromagnetic field strength and other physical quantities may also be included in the aforementioned "measurement of one channel." This is because, if we focus on electromagnetic field strength, a multi-role sensor is also a "sensor that outputs one signal data 106 related to electromagnetic field strength." It should be noted that "measurement of one channel" does not mean completely excluding cross-analysis with other channels or noise-related information after noise separation.

[0022] Figure 2 is a flowchart showing an example of a process performed by the computer system 100 in this embodiment.

[0023] First, the processor 101 acquires signal data 106 measured on one channel by an instrument such as an oscilloscope and stores it in the memory resource 104 (step S1).

[0024] Figure 3(a) is a schematic diagram of an example of signal data 106. Signal data 106 is time-series data that associates time with signal intensity.

[0025] Refer to Figure 2 again. Next, the processor 101 defines multiple subwaves included in the signal data 106 (step S2).

[0026] Figure 3(b) is a schematic diagram of an example of a partial wave 106a. A partial wave 106a is defined as the portion of the signal data 106 that exceeds a threshold set by the user. The start time of the partial wave 106a is the time when the signal strength becomes equal to or greater than the threshold, and the end time is the time when the signal strength falls below the threshold. In addition, the user may define the minimum time interval between two adjacent partial waves 106a in order to distinguish between adjacent partial waves 106a.

[0027] The processor 101 stores the partial wave 106a and an ID that identifies the partial wave 106a in the memory resource 104. Furthermore, the processor 101 stores the start time and end time of the partial wave 106a in the memory resource 104, associating them with the ID as the time period in which the partial wave 106a exists.

[0028] Refer to Figure 2 again. Next, the processor 101 estimates the generation period of each partial wave 106a based on the start or end time of the partial wave 106a (step S3). In this step, it is not necessary to estimate the generation period of all partial waves 106a. The processor 101 only needs to estimate the generation period of at least some of the multiple partial waves 106a. This embodiment is a type of BSS and is based on the premise that the generation period of electromagnetic noise and the number of electromagnetic noise sources are not known in advance. Furthermore, the estimated generation period is assumed to be specific to the source of the electromagnetic noise.

[0029] Therefore, the processor 101 classifies each of the multiple partial waves 106a into a cluster for each estimated generation period (step S4). In this step, it is not necessary to classify all of the partial waves 106a into a cluster. The processor 101 only needs to classify at least some of the multiple partial waves 106a into a cluster.

[0030] Next, the processor 101 determines whether the characteristics of each of the multiple subwave 106a belonging to the same cluster are similar, and performs filtering to exclude subwave 106a that are determined not to have similar characteristics to other subwave 106a from the cluster (step S5). As an example, the processor 101 uses a distance function that defines the distance between subwave 106a, and excludes subwave 106a from the cluster if the distance from other subwave 106a exceeds a threshold.

[0031] With the above steps, the basic processing performed by the computer system 100 in this embodiment is completed.

[0032] Next, we will explain the details of steps S3 to S5 mentioned above.

[0033] <<Step S3>> First, we will explain the method for estimating the generation period of the partial wave 106a in step S3.

[0034] Figure 4 is a schematic diagram showing an example of a digitized signal generated by the processor 101 in step S3.

[0035] As shown in Figure 4, the processor 101 generates a digitized signal 107 by digitizing the signal data 106 depending on the presence or absence of the partial wave 106a. For example, the digitized signal 107 is time-series data in which the value "1" is set during the time period in which the partial wave 106a exists, and "0" is set during the time period in which the partial wave 106a does not exist. As a result, the digitized signal 107 becomes a signal consisting of multiple rectangular waves corresponding to each partial wave 106a. Therefore, below, the rectangular waves in the digitized signal 107 will also be referred to as partial waves 106a. In addition, the width of one rectangular wave is equal to the duration during which the partial wave 106a corresponding to that rectangular wave continues.

[0036] Next, the processor 101 identifies multiple candidate periods P0 for the electromagnetic noise generation period, which are time intervals between multiple subwaves 106a in the digitized signal 107. Candidate periods P0 are identified for any two combinations of the multiple subwaves 106a. Candidate periods P0 may be the time difference between the start times of the two subwaves 106a, or the time difference between their end times.

[0037] Next, the processor 101 performs sampling on the digitized signal 107. Figure 5 is a schematic diagram showing an example of the sampling method in step S3.

[0038] As shown in Figure 5, the processor 101 samples a partial wave 106a from the digitized signal 107 at a candidate period P0. The processor 101 performs this sampling for all of the multiple candidate periods P0. Figure 5 illustrates the case where sampling is performed using one of the multiple candidate periods P0.

[0039] Processor 101 sets "1" for the time when the partial wave 106a was sampled and "0" for the time when it was not sampled. Furthermore, processor 101 calculates the total number of "1"s k. Processor 101 calculates the total number k for all candidate periods P0.

[0040] The closer the candidate period P0 is to the true generation period, the easier it is to sample the partial wave 106a generated at that period, and therefore the total number k is expected to be larger. However, in reality, it is not this simple, and the total number k also depends on the duration and generation period of the partial wave 106a. For example, partial waves 106a with longer durations are easier to sample, and if the generation period is short, more partial waves 106a will be sampled, resulting in a larger total number k.

[0041] Therefore, in this embodiment, the total number k is evaluated by considering the probability that a wave is randomly sampled from a random signal in which multiple waves are randomly distributed. As an example, in this embodiment, the probability that the number of waves sampled from the random signal at a candidate period P0 is greater than the total number k is considered.

[0042] If we denote the cumulative distribution function for sampling k or fewer waves in a random signal as F(k), then the negative log probability P(k) is defined by the following equation (1).

[0043]

number

[0044] In equation (1), 1-F(k) is the complementary cumulative distribution function, which represents the probability of sampling more than k waves in a random signal. The cumulative distribution function F(k) is not particularly limited and can be a binomial distribution or a Poisson distribution. The negative logarithmic probability P(k) is an indicator that increases as the probability of sampling more than k waves in a random signal (1-F(k)) decreases. Therefore, candidate periods P0 with a large value of P(k) are less likely to have a total number of k by chance and are more likely to be equal to the true generation period of electromagnetic noise.

[0045] Figure 6 shows an example of the log probability calculated according to equation (1), plotted for each candidate period P0.

[0046] In this embodiment, the processor 101 estimates the candidate period Pt as the true occurrence period Pt, where the negative log probability P(k) is greater than or equal to a predetermined threshold. This allows for the exclusion of candidate periods P0 where the total number happens to be k, and the extraction of candidate periods P0 that are close to the true occurrence period of electromagnetic noise.

[0047] In this case, there may be multiple estimated true occurrence periods Pt. When one occurrence period Pt is an integer multiple of another occurrence period Pt, these occurrence periods Pt are equal to the occurrence periods of the same noise source. Therefore, the processor 101 excludes occurrence periods that are integer multiples of other occurrence periods. This eliminates occurrence periods related to the same noise source.

[0048] The generation period Pt corresponds to each individual noise source. Therefore, if there are multiple noise sources, the number of generation period Pt values ​​ultimately obtained in this step will also be multiple.

[0049] The user may set the maximum and minimum values ​​for the occurrence period Pt. In this case, the processor 101 discards occurrence period Pt that is smaller than the minimum value or larger than the maximum value and does not use them in subsequent processing.

[0050] <<Step S4>> Next, we will explain how to classify into clusters in step S4.

[0051] First, the processor 101 samples the digitized signal 107 at the generation period Pt estimated above. Figure 7 is a schematic diagram showing an example of this sampling method.

[0052] As shown in Figure 7, the processor 101 samples a partial wave 106a from the digitized signal 107 with a generation period Pt. In this example, the processor 101 sets "1" at the time the partial wave 106a was sampled and "0" at the time it was not sampled. Furthermore, the processor 101 calculates the total number of "1"s k and identifies the sampled partial wave 106a and its duration.

[0053] The time periods in which each subwave 106a exists are stored in the memory resource 104. Therefore, the processor 101 can identify the time period containing the time when "1" is set as the duration of the subwave 106a. Furthermore, since the ID of the subwave 106a is stored in the memory resource 104 in association with the time period, the processor 101 can also identify the subwave 106a at the time when "1" is set.

[0054] Furthermore, the processor 101 performs the above sampling for each of multiple generation periods Pt while changing the shift amount Δ that shifts the start time of sampling. As a result, the processor 101 can obtain analysis data in which the total number k, the sampled partial wave 106a, the duration of the sampled partial wave 106a, the generation period Pt, and the shift amount Δ are correlated with each other.

[0055] Figure 8 shows an example of a visualization of the total number k associated with a certain occurrence period Pt in the analysis data.

[0056] In Figure 8, the cluster of partial wave 106a corresponding to the generation period Pt will be referred to as "Cluster 1" below. The horizontal and vertical axes in Figure 8 represent the shift amount and duration, respectively. The intensity of the color indicates the total number k, with darker colors indicating a larger total number k.

[0057] Furthermore, in the following, the partial wave 106a associated with the generation period Pt of "Cluster 1" will be referred to as partial wave 106a belonging to "Cluster 1".

[0058] The grayscale patterns in Figure 8 characterize each subwave 106a belonging to "Cluster 1". However, at this stage, a light gray area extends around the black areas, and the grayscale patterns are not clearly defined. This is because there is a shift amount where the total value k happens to be large.

[0059] Therefore, the processor 101 clarifies the intensity pattern by calculating the negative log probability P(k) in equation (1). The negative log probability P(k) is an index that increases as the probability that the number of waves sampled with generation period Pt from a random signal in which multiple waves are randomly distributed becomes smaller than the total number k.

[0060] Figure 9 shows an example of a visualization of the negative log probability P(k) calculated from the total number k in Figure 8. Similar to Figure 8, the horizontal and vertical axes in Figure 9 represent the shift amount and duration, respectively. The binomial distribution was used as the cumulative distribution function F(k) in equation (1).

[0061] As shown in Figure 9, the area around the black part becomes closer to white, and the shading pattern becomes clearer compared to Figure 8.

[0062] Next, as shown in Figure 7, when the digitized signal 107 is sampled at a certain generation period Pt, the processor 101 classifies the sampled partial wave 106a, which is sampled with a shift amount such that the negative log probability P(k) is greater than or equal to a predetermined threshold, into a cluster corresponding to its generation period Pt. This prevents partial waves 106a corresponding to a shift amount where the total value k is large by chance from being classified into that cluster, thereby improving the accuracy of cluster classification.

[0063] Figure 10 shows an example of a visualization of the negative log probability P(k) for "Cluster 2" with a different occurrence period Pt than that shown in Figure 9.

[0064] Figure 11 also shows an example of a visualization of the negative log probability P(k) for "Cluster 3," which has an occurrence period Pt different from both Figure 9 and Figure 10.

[0065] As shown in Figures 9 to 11, the intensity patterns characterize each cluster.

[0066] <<Step S5>> Next, we will explain the filtering in step S5.

[0067] In step S5, the processor 101 determines whether each of the multiple subwaves 106a belonging to the same cluster has similar characteristics to the subwave 106a that represents that cluster. Here, the processor 101 determines the similarity of features based on the distance of each subwave 106a. For example, the processor 101 removes any subwaves 106a whose distance from the representative subwave 106a of a cluster is greater than or equal to a predetermined threshold, and leaves the other subwaves 106a in the cluster. The distance can be calculated from the cross-correlation value of each subwave 106a in the signal data 106 before digitization. An example of the distance can be expressed as shown in equation (2) below.

[0068]

number

[0069] In equation (2), x and y represent the signal strengths of the two partial waves 106a, respectively. The bars above x and y represent the average signal strength over the duration of the partial wave 106a. "mac xcorr" represents the maximum value of the cross-correlation between the two arguments.

[0070] One method for determining the partial wave 106a that represents the cluster is the N / PC (Noise / Possibilistic Clustering) algorithm disclosed in the following references [1] and [2]. [1] RN Dave and R. Krishnapuram, “Robust Clustering Methods: A Unified View,” IEEE Trans. Fuzzy Systems, vol. 5, no. 2, pp. 270-293, 1997. [2] R. Krishnapuram, JM Keller: “A Possibilistic Approach to Clustering,” IEEE Transactions on Fuzzy Systems, vol. 1, no. 2, pp. 98-110, May 1993.

[0071] Figure 12 shows an example of a waveform obtained by superimposing each subwave 106a that belonged to "Cluster 1" before filtering in step S5. Figure 13 shows an example of a waveform obtained by superimposing each subwave 106a that remained in "Cluster 1" after filtering in step S5.

[0072] As is clear from Figures 12 and 13, filtering based on distance reduces the number of partial waves 106a belonging to the same cluster, allowing only partial waves 106a with similar waveforms to remain in the same cluster. The effect of filtering becomes clearer when using the frequency spectra of the partial waves 106a belonging to each cluster.

[0073] Figure 14 shows an example of the frequency spectrum of each partial wave 106a before filtering in step S5. Figure 15 shows an example of the frequency spectrum of each partial wave 106a after filtering.

[0074] In Figures 14 and 15, "Signal Data" is the frequency spectrum of the original signal data 106. "Cluster 1," "Cluster 2," and "Cluster 3" are the frequency spectra of the partial waves 106a belonging to each respective cluster.

[0075] As shown in Figure 14, even before filtering, it is possible to determine to some extent which cluster contains the main component of electromagnetic noise at a given frequency. For example, at a frequency of around 70 MHz, the spectral intensity of "Cluster 1" is about the same as that of the "Signal Data," so it can be determined that "Cluster 1" is the main component of electromagnetic noise.

[0076] However, at frequencies around 180MHz, 330MHz, and 370MHz, the spectral intensities of "Cluster 1" and "Cluster 2" are similar, making it impossible to definitively determine which cluster contains the main component of the electromagnetic noise.

[0077] On the other hand, as shown in Figure 15, filtering allows us to determine that the main component of electromagnetic noise around 180 MHz, 330 MHz, and 370 MHz is "Cluster 2".

[0078] According to the embodiment described above, the generation period of the partial wave 106a in the signal data 106 obtained from a single channel measurement is estimated (step S3). Then, at least a portion of the multiple partial waves 106a are classified into multiple clusters according to their generation period (step S4), and partial waves 106a that do not have similar characteristics to other partial waves 106a in the same cluster are excluded from that cluster (step S5). This improves the noise separation of the signal data 106 obtained from a single channel measurement.

[0079] <Second Embodiment> Figure 16 is a flowchart showing an example of the processing performed by the computer system 100 in this embodiment. In Figure 16, the same reference numerals are used for the same steps as in Figure 2, and their explanations are omitted below.

[0080] In this embodiment, the partial waves 106a that were excluded by filtering in step S5 are reclassified into clusters as follows.

[0081] First, the processor 101 executes each of the processes in steps S1 to S5 according to the first embodiment.

[0082] Next, the processor 101 reclassifies the partial wave 106a that was excluded in the filtering of step S5 into a different cluster from the cluster from which it was excluded (step S6). For example, the processor 101 classifies a certain partial wave 106a into the cluster with the frequency closest to the cluster from which it was excluded among several clusters.

[0083] Next, the processor 101 performs filtering again in the same manner as in step S5 (step S7).

[0084] With the above steps, the basic processing performed by the computer system 100 in this embodiment is completed.

[0085] According to the embodiment described above, it is possible to increase the number of partial waves 106a that are correctly classified into each cluster, and to decrease the number of partial waves 106a that are excluded without being classified into a cluster.

[0086] <Third Embodiment> Figure 17 is a flowchart showing an example of a process performed by the computer system 100 in this embodiment. In Figure 17, the same reference numerals are used for the same steps as in Figure 2, and their descriptions are omitted below.

[0087] In this embodiment, the processor 101 calculates the frequency spectrum shown in Figure 15 of the first embodiment.

[0088] First, the processor 101 executes each of the processes in steps S1 to S5 according to the first embodiment.

[0089] Next, the processor 101 calculates the frequency spectrum of each partial wave 106a belonging to each cluster (step S8). The processor 101 may also instruct the UI device 102 to display the calculated frequency spectrum.

[0090] With the above steps, the basic processing performed by the computer system 100 in this embodiment is completed.

[0091] To avoid communication interference at a certain frequency, electromagnetic noise levels are often evaluated in the frequency domain. In this embodiment, since the spectrum is calculated in the frequency domain, when the electromagnetic noise level is high at a particular frequency, the user can identify the main components of the electromagnetic noise at that frequency.

[0092] Furthermore, since each cluster corresponds to a single generation cycle, it corresponds to the switching frequency of a device that repeatedly switches on and off during that cycle. Once the switching frequency is known based on the frequency spectrum, the device that switches on and off at that frequency can be identified from its specifications. In this way, this embodiment can help users identify devices that are sources of electromagnetic noise.

[0093] <Fourth Embodiment> Figure 18 is a flowchart showing an example of a process performed by the computer system 100 in this embodiment. In Figure 18, the same reference numerals are used for the same steps as in Figure 2, and their descriptions are omitted below.

[0094] As shown in Figure 18, in this embodiment, the processor 101 receives input for the parameters required for each process in steps S2 to S5, and the processor 101 instructs the UI device 102 to display the results of each process.

[0095] For example, in step S2, the processor 101 accepts input for a first parameter required to define the partial wave 106a. The first parameter may be, for example, a threshold value for the signal intensity used to define the partial wave 106a in the signal data 106. Another example of a first parameter is the minimum time interval between two adjacent partial waves 106a.

[0096] The processor 101 then instructs the UI device 102 to display the partial wave 106a defined based on the first parameter, and the user confirms it.

[0097] Meanwhile, in steps S3 to S5, the processor 101 accepts input for the second parameter required in these steps.

[0098] For example, an example of a second parameter in step S3 is the maximum and minimum values ​​of the occurrence period Pt. After executing step S3, the processor 101 instructs the UI device 102 to display the occurrence period Pt between the maximum and minimum values.

[0099] An example of the second parameter in step S4 is a threshold value for the negative log probability P(k) shown in Figure 7. After executing step S4, the processor 101 instructs the UI device 102 to display the clusters classified based on the second parameter, and the user confirms this. Examples of cluster displays include the grayscale patterns shown in Figures 9 to 11.

[0100] An example of a second parameter in step S5 is a threshold value for the distance between a partial wave 106a included in a cluster and a partial wave 106a that represents that cluster. After executing step S5, the processor 101 instructs the UI device 102 to display the partial waves 106a included in each cluster.

[0101] According to the embodiment described above, the user can check the results of processing according to the first and second parameters, and can identify the noise source while adjusting the first and second parameters.

[0102] <Fifth Embodiment> Figure 19 is a diagram showing an example of a computer system according to this embodiment. In Figure 19, the same reference numerals are used for the same elements as those described in Figure 1, and their descriptions are omitted below.

[0103] As shown in Figure 19, the computer system 100 according to this embodiment includes at least one sensor 200 that acquires signal data 106. The sensor 200 is a physical device such as an antenna, current probe, and voltage probe installed in a space where electromagnetic noise is present, and is wired to the NI device 103. This allows the processor 101 to acquire signal data 106 containing electromagnetic noise from the sensor 200.

[0104] <Sixth Embodiment> Figure 20 is a diagram showing an example of a computer system according to this embodiment. In Figure 20, the same elements as those described in Figure 1 are denoted by the same reference numerals, and their descriptions are omitted below.

[0105] As shown in Figure 20, the computer system 100 according to this embodiment includes at least one sensor unit 202 that acquires signal data 106. The sensor unit 202 has a sensor 200 according to the fifth embodiment and a transmitter 201. The transmitter 201 is a device that wirelessly transmits the signal data 106 acquired by the sensor 200.

[0106] According to this, the NI device 103 receives the signal data 106 from the transmitter 201, allowing the processor 101 to acquire the signal data 106. Furthermore, since the transmitter 201 wirelessly transmits the signal data 106, it also increases the options for where the sensor unit 202 can be installed.

[0107] The effects described herein are merely illustrative and not limited to those described herein; other effects may also occur.

[0108] The present invention is not limited to the embodiments described above, and various modifications are included. For example, each of the embodiments described above is described in detail for the purpose of clearly illustrating the present invention, and the present invention is not necessarily limited to having all of the described components. Furthermore, it is possible to replace a part of the configuration of one embodiment with the configuration of another embodiment, and it is also possible to add the configuration of another embodiment to the configuration of one embodiment. In addition, it is possible to add, delete, or replace parts of the configuration of each embodiment with other configurations.

[0109] Furthermore, each of the above configurations, functions, processing units, and processing means may be implemented in hardware, either partially or entirely, by designing them as integrated circuits, for example. Alternatively, each of the above configurations and functions may be implemented in software by a processor interpreting and executing programs that implement each function. Information such as programs, decision tables, and files that implement each function can be stored in memory, storage devices such as HDDs and SSDs, or recording media such as IC (Integrated Circuit) cards, SD (Secure Digital) cards, and DVDs (Digital Versatile Discs). Also, control lines and information lines are shown only if deemed necessary for explanation, and not all control lines and information lines are necessarily shown in the actual product. In practice, almost all configurations can be considered interconnected.

[0110] This computer system 100 may be implemented by having a user (operator) perform some or all of the functions and processes realized by the processing program 105.

[0111] In some cases, the computer system 100 may not have a UI device 102, and instead entrust some of the output processing to the user and some of the input processing from the user to an external processor system such as a smartphone or tablet terminal (referred to as an external processor system). In such cases, the computer system 100 (or processor 101, processing program 105) may do the following in order to execute the processing and other parts of the program as described above.

[0112] *As an alternative to outputting to the user using the UI device 102 described above, data necessary for outputting to the user is sent to an external processor system via the NI device 103. Examples of such data include the output data itself, data for generating the output data on the separate processor system, but it may also be a program or web data that describes the process of performing user output on the external processor system.

[0113] *Instead of receiving user input or operations using the UI device 102 described above, data indicating user input or operations is received from an external processor system via the NI device 103. From another perspective, the meaning of data output to the user may include not only output by the computer system 100 itself, but also causing (serving) another entity other than the computer system 100 to output such data. Furthermore, the meaning of receiving user input or operations may include not only direct output or reception to the user by the UI device 102 of the computer system 100, but also indirect reception by the computer system 100. [Explanation of Symbols]

[0114] 100...Computer system, 101...Processor, 102...UI device, 103...NI device, 104...Memory resource, 105...Processing program, 106...Signal data, 106a...Partial wave, 107...Digitized signal, 109...Bus, 200...Sensor, 201...Transmitter, 202...Sensor unit.

Claims

1. A computer system having one or more processors and one or more memory resources, The one or more processors mentioned above are: Acquire a signal containing electromagnetic noise, Define a plurality of partial waves included in the aforementioned signal, The generation period at which the partial wave occurs in the signal is estimated for at least a portion of the plurality of partial waves. At least a portion of the multiple partial waves are classified into multiple clusters according to their generation period, Filtering is performed to exclude from the cluster any subwaves belonging to the same cluster that do not have similar characteristics to other subwaves. Computer system.

2. The computer system according to claim 1, The estimation of the occurrence period is as follows: A digitized signal is generated by digitizing the signal according to the presence or absence of the aforementioned partial wave. Multiple time intervals between the multiple partial waves are identified as candidate periods for the generation period. The candidate period is estimated as the generation period, where an index increases as the probability decreases that the number of waves sampled at the candidate period from a random signal in which multiple waves are randomly distributed is greater than the total number of partial waves sampled from the digitized signal at the candidate period. A computer system that includes processing.

3. The computer system according to claim 2, The estimation of the occurrence period involves excluding the occurrence period that is an integer multiple of the other occurrence periods if multiple occurrence periods are estimated. A computer system that includes processing.

4. The computer system according to claim 1, The classification into the aforementioned clusters is, A digitized signal is generated by digitizing the signal according to the presence or absence of the aforementioned partial wave. The total number of partial waves sampled from the digitized signal in the aforementioned generation cycle and the partial waves themselves are determined for each generation cycle by changing the amount by which the sampling start time is shifted. The sampled partial waves are classified into the cluster of the generation period when sampling is performed with a shift amount such that the index, which increases as the probability of the number of waves sampled in the generation period from a random signal in which multiple waves are randomly distributed becomes smaller than the total number, is greater than or equal to a threshold. A computer system that includes processing.

5. The computer system according to claim 1, The determination of whether the aforementioned features are similar is made based on the cross-correlation values ​​between the partial waves. Computer system.

6. The computer system according to claim 5, The one or more processors mentioned above are: The excluded partial wave is reclassified into a different cluster from the cluster from which it was excluded. After the aforementioned reclassification, the filtering is performed again. Computer system.

7. The computer system according to claim 1, After the filtering, the frequency spectrum of each of the partial waves belonging to each of the clusters is calculated. Computer system.

8. The computer system according to claim 1, The one or more processors mentioned above are: The system accepts input of parameters required for any of the following processes: defining the partial wave, estimating the generation period, classifying into clusters, and filtering. The command to display the result of the above process, Computer system.

9. The computer system according to claim 1, The system further includes a sensor that acquires the aforementioned signal. Computer system.

10. The computer system according to claim 1, A sensor that acquires the aforementioned signal, The system further comprises a transmitter that transmits the signal acquired by the sensor, The one or more processors acquire the signal transmitted by the transmitter. Computer system.

11. A noise isolation method performed by a computer system having one or more processors and one or more memory resources, Steps include acquiring a signal containing electromagnetic noise, The steps include defining a plurality of partial waves included in the signal, The steps include: estimating the generation period at which the partial wave occurs in the signal for at least a portion of the plurality of partial waves; The steps include classifying at least a portion of the multiple partial waves into multiple clusters according to their generation period, A step of filtering out of multiple partial waves belonging to the same cluster that do not have similar characteristics to other partial waves from the cluster, A noise separation method that includes [details omitted].

12. A noise separation method according to claim 11, The step of estimating the occurrence period is: A step of generating a digitized signal by digitizing the signal according to the presence or absence of the aforementioned partial wave, The steps include: identifying multiple time intervals between multiple partial waves as candidate periods for the generation period; Steps include: estimating the candidate period as the generation period, where an index that increases as the probability of the number of waves sampled at the candidate period from a random signal in which multiple waves are randomly distributed is greater than the total number of partial waves sampled from the digitized signal at the candidate period is smaller, and the index that increases as the index increases above a threshold; A noise separation method having the following characteristics.

13. A noise separation method according to claim 12, The step of estimating the occurrence period includes, if there are multiple estimated occurrence periods, excluding the occurrence period that is an integer multiple of the other occurrence periods. A noise separation method having the following characteristics.

14. A noise separation method according to claim 11, The step of classifying the multiple partial waves into multiple clusters according to their generation period is: A step of generating a digitized signal by digitizing the signal according to the presence or absence of the aforementioned partial wave, The total number of partial waves sampled from the digitized signal at the aforementioned generation cycle and the part The steps involve determining the wave and the starting time of sampling by changing the amount of the shift, for each generation period, A step of classifying the sampled partial waves into the cluster of the generation period when sampling is performed with a shift amount such that the index, which increases as the probability of the number of waves sampled in the generation period from a random signal in which multiple waves are randomly distributed becomes smaller than the total number, is greater than or equal to a threshold, A noise separation method having the following characteristics.

15. A program that causes a computer system to execute the noise separation method described in any one of claims 11 to 14.

Images