A power line carrier communication anomaly detection method and system

Abstract

The application provides a power line carrier communication anomaly detection method and system, and relates to the technical field of power line carrier communication, and comprises the following steps: S1, obtaining noise signal spectrum collected when a to-be-tested communication device is turned off; S2, obtaining a conducted signal spectrum collected from a power line of the to-be-tested communication device and a plurality of radiation signal spectrums under different distance information when the to-be-tested communication device is turned on, and recording an operation process of the to-be-tested communication device; S3, performing baseline calibration on the conducted signal spectrum and the plurality of radiation signal spectrums under different distance information according to the noise signal spectrum; S4, performing spectrum segment interception on the calibrated conducted signal spectrum and radiation signal spectrum according to the operation process of the to-be-tested communication device, and performing structured storage on corresponding spectrum segments; and S5, performing abnormal state judgment on a plurality of to-be-tested signal segment groups in sequence through a neural network model.

Description

Technical Field

[0001] This invention relates to the field of power line carrier communication technology, specifically to a method and system for detecting anomalies in power line carrier communication. Background Technology

[0002] Currently, in the field of power line carrier communication technology, electromagnetic information generated during the operation of information systems couples into power lines. These leaked electromagnetic signals may be wirelessly radiated into space through power lines, or they may be coupled into power line networks. Hackers can obtain and analyze these electromagnetic signals transmitted through power lines to recover confidential information processed by the information system, leading to information leakage due to power lines, which is a serious problem. Summary of the Invention

[0003] The purpose of this invention is to provide a method and system for detecting anomalies in power line carrier communication. The method uses a spectrum analyzer and an antenna array to acquire the conducted signals, radiated signals, and corresponding noise signals of the communication device under test through dual channels. At the same time, it records the operation process of the communication device under test, providing effective data for subsequent anomaly analysis of key operational events and corresponding signals generated during operation using a neural network model.

[0004] To solve the above-mentioned technical problems, the present invention adopts the following solution:

[0005] A method for detecting anomalies in power line carrier communication includes the following steps:

[0006] S1. Obtain the noise signal spectrum collected when the communication device under test is turned off;

[0007] S2. Obtain the conducted signal spectrum and the radiated signal spectrum at multiple different distances when the communication device under test is turned on, collected from the power line of the communication device under test. At the same time, record the operation process of the communication device under test. The distance information is the distance between the antenna used to collect the radiated signal spectrum and the power line.

[0008] S3. Based on the noise signal spectrum, perform baseline calibration on the conducted signal spectrum and the radiated signal spectrum under multiple different distance information to obtain the calibrated conducted signal spectrum and radiated signal spectrum.

[0009] S4. Based on the operation process of the communication device under test, the calibrated conducted signal spectrum and radiated signal spectrum are segmented into spectrum fragments, the corresponding spectrum fragments are correlated, multiple spectrum fragment groups to be detected are generated based on the correlation results, and the multiple spectrum fragment groups to be detected are stored in a structured manner.

[0010] S5. The abnormal state of several groups of signal segments to be detected is judged sequentially by a neural network model, and the corresponding abnormal signal segments and their information are obtained based on the judgment results.

[0011] Furthermore, in S1, the process of obtaining the noise signal spectrum is as follows: after turning off the communication device under test and making the power line of the communication device under test unloaded, the environment in which the communication device under test is located is sampled multiple times by a spectrum analyzer and an antenna array, that is, the environment in which the communication device under test is located is noise scanned, and the average value of the results of multiple samplings is used to generate the corresponding noise signal spectrum.

[0012] Furthermore, in S1, the process of obtaining the conducted signal spectrum is as follows: turn on the communication device under test, make the communication device under test run, and during the operation of the communication device under test, collect the conducted signal of the power line through a spectrum analyzer and perform Fourier transform processing on it to obtain the conducted signal spectrum.

[0013] The process of obtaining the radiation signal spectrum is as follows: an antenna array is set up along the power line of the communication device under test. The antenna array includes multiple antennas at different distances from the power line. The communication device under test is turned on and put into operation. During the operation of the communication device under test, the radiation signal of the power line is collected through multiple antennas at different distances from the power line, and Fourier transform is performed on it to obtain the radiation signal spectrum corresponding to the antenna.

[0014] Furthermore, step S3 includes the following steps:

[0015] S31. Baseline calibration is performed on the conducted signal spectrum based on the noise signal spectrum. Invalid data is marked during the calibration process to obtain the calibrated conducted signal spectrum.

[0016] S32. Obtain the antenna position information based on the distance between the antenna and the power line, obtain the noise signal spectrum corresponding to the position information, perform baseline calibration on the radiated signal spectrum based on the noise signal spectrum corresponding to the position information, and obtain the calibrated radiated signal spectrum.

[0017] Furthermore, the noise signal spectrum includes a first noise signal spectrum collected by a spectrum analyzer when the communication device under test is turned off, and a second noise signal spectrum collected by an antenna at different distances when the communication device under test is turned off. The first noise signal spectrum corresponds to the conducted signal spectrum, and the second noise signal spectrum corresponds to the radiated signal spectrum.

[0018] The conducted signal spectrum was acquired using the same spectrum analyzer when the communication device under test was turned on, and the radiated signal spectrum was acquired using the same antenna when the communication device under test was turned on.

[0019] The communication device under test, the spectrum analyzer, and the antenna are synchronized by a clock.

[0020] Furthermore, step S4 includes the following steps:

[0021] S41. Extract key operation events and their timestamps from the operation process of the communication device under test;

[0022] S42. Based on the key operation events, obtain the first threshold and the second threshold of the spectrum analyzer and the antenna respectively, and extend the timestamp with the first threshold and the second threshold respectively to obtain the first time interval and the second time interval corresponding to the key operation events.

[0023] S43. Based on the first time interval and the second time interval, spectral segments are extracted from the spectrum of the conducted signal and the spectrum of the radiated signal, respectively, to obtain the corresponding spectrum segments of the conducted signal and the spectrum segments of the radiated signal.

[0024] S44. Associate the conducted signal spectrum segment and the radiated signal spectrum segment with the corresponding key operation events and timestamps to generate a group of spectrum segments to be detected;

[0025] S45. Construct a structured database to store the conducted signal spectrum segments and radiated signal spectrum segments along with their corresponding key operational events and timestamps.

[0026] Furthermore, in S44, the process of generating a group of spectrum segments to be detected is as follows: obtaining distance information of the radiation signal spectrum, marking the radiation signal spectrum segments according to the distance information, associating the radiation signal spectrum segments under different distance information with the conduction signal spectrum segments under the same reference point and the corresponding key operation events and timestamps, and generating multiple groups of spectrum segments to be detected with distance information.

[0027] Furthermore, in S45, the process of building a structured database is as follows:

[0028] Obtain multiple groups of spectrum segments to be detected, each containing distance information;

[0029] Key operation events and timestamps are used as nodes in a tree in a structured database. Multiple groups of spectrum segments to be detected, each with distance information corresponding to the same key operation event, are used as multiple sub-nodes of the node. These multiple groups of spectrum segments to be detected are stored in isolation based on their different distance information.

[0030] Furthermore, the abnormal state judgment process is as follows: Multiple groups of spectrum segments with distance information corresponding to the same node are extracted from the structured database and input into a neural network model. The neural network model performs signal correlation analysis on the conducted signal spectrum segments and radiated signal spectrum segments in each group of spectrum segments to be detected, obtaining a first analysis result. Then, the neural network model performs signal distance correlation analysis on the radiated signal spectrum segments in the group of spectrum segments to be detected under the same node, obtaining a second analysis result. Based on a corresponding preset threshold, it is determined whether both the first and second analysis results are abnormal. If so, the corresponding radiated signal spectrum segment is determined to be an abnormal signal segment, and distance information is obtained.

[0031] A power line carrier communication anomaly detection system, applying the aforementioned power line carrier communication anomaly detection method, includes:

[0032] Noise signal acquisition module: Obtains the noise signal spectrum when the communication device under test is turned off;

[0033] Power line signal acquisition module: acquires the conducted signal spectrum and the radiated signal spectrum at multiple different distances when the communication device under test is turned on, and records the operation process of the communication device under test; the distance information is the distance between the antenna used to acquire the radiated signal spectrum and the power line.

[0034] Baseline calibration module: Based on the noise signal spectrum, the conducted signal spectrum and the radiated signal spectrum under multiple different distance information are baseline calibrated to obtain the calibrated conducted signal spectrum and radiated signal spectrum;

[0035] Structured storage module: Based on the operation process of the communication device under test, the calibrated conducted signal spectrum and radiated signal spectrum are extracted into spectrum segments, the corresponding spectrum segments are correlated, multiple groups of spectrum segments to be detected are generated based on the correlation results, and the multiple groups of spectrum segments to be detected are stored in a structured manner.

[0036] Anomaly detection module: It sequentially performs anomaly detection on several groups of signal segments to be detected using a neural network model, and obtains the corresponding abnormal signal segments and their information based on the detection results.

[0037] The beneficial effects of this invention are:

[0038] This invention provides a method and system for detecting anomalies in power line carrier communication. The same spectrum analyzer and antenna array are used for dual-channel acquisition when the communication device under test is turned off and on, respectively. The environmental noise of the communication device under test when it is turned off can be recorded and common interference sources can be marked. During signal analysis, interference from these sources can be effectively eliminated, achieving rapid detection.

[0039] At the same time, when the communication device under test is turned on, the operation process of the communication device under test also needs to be recorded. By analyzing the key operation events and their timestamps, the amount of data processed by the neural network model can be reduced and the detection rate can be improved.

[0040] Furthermore, in this invention, antennas with different distance information are used to collect signals from power lines. Based on the distance information and key operation events, the collected signal segments are stored in a structured manner, which avoids data clutter and accumulation. At the same time, the corresponding signal segments can be quickly located through the node index, which facilitates the subsequent signal analysis of the signal segments by the neural network model. Attached Figure Description

[0041] Figure 1 This is a flowchart illustrating the power line carrier communication anomaly detection method in Embodiment 1 of the present invention. Detailed Implementation

[0042] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The following description of at least one exemplary embodiment is merely illustrative and is in no way intended to limit the present invention or its application or use. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0043] Unless otherwise specifically stated, the relative arrangement, numerical expressions, and values ​​of the components and steps described in these embodiments do not limit the scope of the invention.

[0044] At the same time, it should be understood that, for ease of description, the dimensions of the various parts shown in the accompanying drawings are not drawn according to actual scale.

[0045] Furthermore, for clarity and brevity, descriptions of well-known structures, functions, and configurations may have been omitted. Those skilled in the art will recognize that various changes and modifications can be made to the examples described herein without departing from the spirit and scope of this disclosure.

[0046] Techniques, methods, and equipment known to those skilled in the art may not be discussed in detail, but where appropriate, such techniques, methods, and equipment should be considered part of the specification.

[0047] In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limitations. Therefore, other examples of exemplary embodiments may have different values.

[0048] The present invention will now be described in detail with reference to the accompanying drawings and embodiments:

[0049] Example 1

[0050] In this embodiment, to address the problem of power line information leakage, a method for detecting anomalies in power line carrier communication is provided. This method is based on dual-channel signal acquisition and a structured database, employing a neural network model to analyze the signals, thereby achieving real-time detection of power lines. Furthermore, when an abnormal signal is detected, an early warning is issued, and the current communication device under test can be interrupted to prevent power line information leakage.

[0051] Specifically, the method is as follows: Figure 1 As shown, it includes the following steps:

[0052] S1. Obtain the noise signal spectrum collected when the communication device under test is turned off;

[0053] S2. Obtain the conducted signal spectrum and the radiated signal spectrum at multiple different distances when the communication device under test is turned on, collected from the power line of the communication device under test. At the same time, record the operation process of the communication device under test. The distance information is the distance between the antenna used to collect the radiated signal spectrum and the power line.

[0054] S3. Based on the noise signal spectrum, perform baseline calibration on the conducted signal spectrum and the radiated signal spectrum under multiple different distance information to obtain the calibrated conducted signal spectrum and radiated signal spectrum.

[0055] S4. Based on the operation process of the communication device under test, the calibrated conducted signal spectrum and radiated signal spectrum are segmented into spectrum fragments, the corresponding spectrum fragments are correlated, multiple spectrum fragment groups to be detected are generated based on the correlation results, and the multiple spectrum fragment groups to be detected are stored in a structured manner.

[0056] S5. The abnormal state of several groups of signal segments to be detected is judged sequentially by a neural network model, and the corresponding abnormal signal segments and their information are obtained based on the judgment results.

[0057] The process of obtaining the noise signal spectrum is as follows: After turning off the communication device under test (DUT) and ensuring its power line is unloaded, the environment in which the DUT is located is sampled multiple times using a spectrum analyzer and an antenna array. This involves scanning the environment for noise, and the average of the multiple samples is used to generate the corresponding noise signal spectrum. Multiple sampling reduces random errors and helps identify stable noise characteristics. For example, when certain noises are prominent, the DUT can be investigated to eliminate the influence of external interference sources.

[0058] Ensuring the power line is unloaded is crucial to guarantee a complete power-off of the equipment and that there are no other loads on the power line, thus preventing conducted noise from external current from interfering with the test results. Even better, sufficient time should be allowed after the equipment is completely powered off to eliminate the effects of residual charge.

[0059] The process of obtaining the conducted signal spectrum is as follows: The communication device under test (DUT) is turned on and put into operation. During the operation of the DUT, the conducted signal of the power line is acquired using a spectrum analyzer, and then subjected to Fourier transform processing to obtain the conducted signal spectrum. Acquiring the conducted signal of the power line using a spectrum analyzer and performing Fourier transform processing is a conventional method for analyzing power line carrier communication, and will not be elaborated further here.

[0060] The process of obtaining the radiation signal spectrum is as follows: an antenna array is set up along the power line of the communication device under test. The antenna array includes multiple antennas at different distances from the power line. The communication device under test is turned on and put into operation. During the operation of the communication device under test, the radiation signal of the power line is collected through multiple antennas at different distances from the power line, and Fourier transform is performed on it to obtain the radiation signal spectrum corresponding to the antenna.

[0061] In one embodiment, step S3 includes the following steps:

[0062] S31. Baseline calibration is performed on the conducted signal spectrum based on the noise signal spectrum. Invalid data is marked during the calibration process to obtain the calibrated conducted signal spectrum.

[0063] S32. Obtain the antenna position information based on the distance between the antenna and the power line, obtain the noise signal spectrum corresponding to the position information, perform baseline calibration on the radiated signal spectrum based on the noise signal spectrum corresponding to the position information, and obtain the calibrated radiated signal spectrum.

[0064] In one embodiment, the noise signal spectrum includes a first noise signal spectrum collected by a spectrum analyzer when the communication device under test is turned off, and a second noise signal spectrum collected by an antenna at different distances when the communication device under test is turned off. Furthermore, the first noise signal spectrum corresponds to the conducted signal spectrum, and the second noise signal spectrum corresponds to the radiated signal spectrum.

[0065] The conducted signal spectrum was acquired using the same spectrum analyzer when the communication device under test was turned on, and the radiated signal spectrum was acquired using the same antenna when the communication device under test was turned on.

[0066] Specifically, the communication device under test (DUT), the spectrum analyzer, and the antenna are synchronized by a clock. Clock synchronization ensures that the conducted signal spectrum acquired by the spectrum analyzer is associated with key operational events generated during the operation of the DUT. Similarly, clock synchronization ensures that the radiated signal spectrum acquired by the antenna array is associated with key operational events generated during the operation of the DUT. Therefore, in this embodiment, signals can be associated with key operational events based on timestamps and stored in a structured manner, facilitating subsequent signal analysis using a neural network model.

[0067] In one embodiment, step S4 includes the following steps S41-S44.

[0068] Step S41: Extract key operation events and their timestamps from the operation of the communication device under test.

[0069] Step S42: Obtain the first threshold and the second threshold of the spectrum analyzer and the antenna respectively based on the key operation event, and perform time extension on the timestamp with the first threshold and the second threshold respectively to obtain the first time interval and the second time interval corresponding to the key operation event.

[0070] Step S43: Based on the first time interval and the second time interval, spectral segments are extracted from the conduction signal spectrum and the radiation signal spectrum respectively to obtain the corresponding conduction signal spectrum segment and radiation signal spectrum segment.

[0071] Step S44: Associate the conducted signal spectrum segment and the radiated signal spectrum segment with the corresponding key operation events and timestamps to generate a group of spectrum segments to be detected;

[0072] Specifically, the process of generating a group of spectrum segments to be detected is as follows: obtaining distance information of the radiation signal spectrum, marking the radiation signal spectrum segments according to the distance information, associating the radiation signal spectrum segments under different distance information with the conduction signal spectrum segments under the same reference point and the corresponding key operation events and timestamps, and generating multiple groups of spectrum segments to be detected with distance information.

[0073] Step S45: Construct a structured database, storing the conducted signal spectrum segments and radiated signal spectrum segments along with their corresponding key operational events and timestamps into the structured database;

[0074] In S45, the process of building a structured database is as follows:

[0075] Obtain multiple groups of spectrum segments to be detected, each containing distance information;

[0076] Key operation events and timestamps are used as nodes in a tree in a structured database. Multiple groups of spectrum segments to be detected, each with distance information corresponding to the same key operation event, are used as multiple sub-nodes of the node. These multiple groups of spectrum segments to be detected are stored in isolation based on their different distance information.

[0077] In one embodiment, the abnormal state judgment process is as follows: extract multiple groups of spectrum segments to be detected with distance information corresponding to the same node from the structured database and input them into a neural network model. Perform signal correlation analysis on the conducted signal spectrum segment and the radiated signal spectrum segment in each group of spectrum segments to be detected through the neural network model to obtain a first analysis result; then perform signal distance correlation analysis on the radiated signal spectrum segment in the group of spectrum segments to be detected under the same node through the neural network model to obtain a second analysis result; determine whether the first analysis result and the second analysis result are both abnormal according to the corresponding preset threshold. If so, determine that the corresponding radiated signal spectrum segment is an abnormal signal segment and obtain the distance information.

[0078] Example 2

[0079] A power line carrier communication anomaly detection system, applying the aforementioned power line carrier communication anomaly detection method, includes:

[0080] Noise signal acquisition module: Obtains the noise signal spectrum when the communication device under test is turned off;

[0081] Power line signal acquisition module: acquires the conducted signal spectrum and the radiated signal spectrum at multiple different distances when the communication device under test is turned on, and records the operation process of the communication device under test; the distance information is the distance between the antenna used to acquire the radiated signal spectrum and the power line.

[0082] Baseline calibration module: Based on the noise signal spectrum, the conducted signal spectrum and the radiated signal spectrum under multiple different distance information are baseline calibrated to obtain the calibrated conducted signal spectrum and radiated signal spectrum;

[0083] Structured storage module: Based on the operation process of the communication device under test, the calibrated conducted signal spectrum and radiated signal spectrum are extracted into spectrum segments, the corresponding spectrum segments are correlated, multiple groups of spectrum segments to be detected are generated based on the correlation results, and the multiple groups of spectrum segments to be detected are stored in a structured manner.

[0084] Anomaly detection module: It sequentially performs anomaly detection on several groups of signal segments to be detected using a neural network model, and obtains the corresponding abnormal signal segments and their information based on the detection results.

[0085] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Based on the technical essence of the present invention, any simple modifications, equivalent substitutions, and improvements made to the above embodiments within the spirit and principles of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A method for detecting anomalies in power line carrier communication, characterized in that, Includes the following steps: S1. Obtain the noise signal spectrum collected when the communication device under test is turned off; The noise signal spectrum includes a first noise signal spectrum collected by a spectrum analyzer when the communication device under test is turned off, and a second noise signal spectrum collected by an antenna at different distances when the communication device under test is turned off. The first noise signal spectrum corresponds to the conducted signal spectrum, and the second noise signal spectrum corresponds to the radiated signal spectrum. The conducted signal spectrum was acquired using the same spectrum analyzer when the communication device under test was turned on, and the radiated signal spectrum was acquired using the same antenna when the communication device under test was turned on. The communication device under test, the spectrum analyzer, and the antenna are synchronized by a clock. S2. Obtain the conducted signal spectrum and the radiated signal spectrum at multiple different distances when the communication device under test is turned on, collected from the power line of the communication device under test. At the same time, record the operation process of the communication device under test. The distance information is the distance between the antenna used to collect the radiated signal spectrum and the power line. S3. Based on the noise signal spectrum, perform baseline calibration on the conducted signal spectrum and the radiated signal spectrum under multiple different distance information to obtain the calibrated conducted signal spectrum and radiated signal spectrum. Step S3 includes the following steps: S31. Baseline calibration is performed on the conducted signal spectrum based on the noise signal spectrum. Invalid data is marked during the calibration process to obtain the calibrated conducted signal spectrum. S32. Obtain the antenna position information based on the distance between the antenna and the power line, obtain the noise signal spectrum corresponding to the position information, perform baseline calibration on the radiated signal spectrum based on the noise signal spectrum corresponding to the position information, and obtain the calibrated radiated signal spectrum. S4. Based on the operation process of the communication device under test, the calibrated conducted signal spectrum and radiated signal spectrum are segmented into spectrum fragments, the corresponding spectrum fragments are correlated, multiple spectrum fragment groups to be detected are generated based on the correlation results, and the multiple spectrum fragment groups to be detected are stored in a structured manner. S5. The abnormal state of several groups of signal segments to be detected is judged sequentially by a neural network model, and the corresponding abnormal signal segments and their information are obtained based on the judgment results.

2. The method for detecting anomalies in power line carrier communication according to claim 1, characterized in that, In S1, the process of obtaining the noise signal spectrum is as follows: after turning off the communication device under test and making the power line of the communication device under test unloaded, the environment in which the communication device under test is located is sampled multiple times by a spectrum analyzer and an antenna array, that is, the environment in which the communication device under test is located is noise scanned, and the average value of the results of multiple samplings is used to generate the corresponding noise signal spectrum.

3. The method for detecting anomalies in power line carrier communication according to claim 1, characterized in that, In S1, the process of obtaining the conducted signal spectrum is as follows: turn on the communication device under test, make the communication device under test run, and during the operation of the communication device under test, collect the conducted signal of the power line through a spectrum analyzer and perform Fourier transform processing on it to obtain the conducted signal spectrum. The process of obtaining the radiation signal spectrum is as follows: an antenna array is set up along the power line of the communication device under test. The antenna array includes multiple antennas at different distances from the power line. The communication device under test is turned on and put into operation. During the operation of the communication device under test, the radiation signal of the power line is collected through multiple antennas at different distances from the power line, and Fourier transform is performed on it to obtain the radiation signal spectrum corresponding to the antenna.

4. The method for detecting anomalies in power line carrier communication according to claim 1, characterized in that, Step S4 includes the following steps: S41. Extract key operation events and their timestamps from the operation process of the communication device under test; S42. Based on the key operation events, obtain the first threshold and the second threshold of the spectrum analyzer and the antenna respectively, and extend the timestamp with the first threshold and the second threshold respectively to obtain the first time interval and the second time interval corresponding to the key operation events. S43. Based on the first time interval and the second time interval, spectral segments are extracted from the spectrum of the conducted signal and the spectrum of the radiated signal, respectively, to obtain the corresponding spectrum segments of the conducted signal and the spectrum segments of the radiated signal. S44. Associate the conducted signal spectrum segment and the radiated signal spectrum segment with the corresponding key operation events and timestamps to generate a group of spectrum segments to be detected; S45. Construct a structured database to store the conducted signal spectrum segments and radiated signal spectrum segments along with their corresponding key operational events and timestamps.

5. The method for detecting anomalies in power line carrier communication according to claim 4, characterized in that, In S44, the process of generating a group of spectrum segments to be detected is as follows: obtaining distance information of the radiation signal spectrum, marking the radiation signal spectrum segments according to the distance information, associating the radiation signal spectrum segments under different distance information with the conduction signal spectrum segments under the same reference point and the corresponding key operation events and timestamps, and generating multiple groups of spectrum segments to be detected with distance information.

6. The method for detecting anomalies in power line carrier communication according to claim 5, characterized in that, In S45, the process of building a structured database is as follows: Obtain multiple groups of spectrum segments to be detected, each containing distance information; Key operation events and timestamps are used as nodes in a tree in a structured database. Multiple groups of spectrum segments to be detected, each with distance information corresponding to the same key operation event, are used as multiple sub-nodes of the node. These multiple groups of spectrum segments to be detected are stored in isolation based on their different distance information.

7. The method for detecting anomalies in power line carrier communication according to claim 6, characterized in that, The abnormal state judgment process is as follows: Multiple groups of spectrum segments with distance information corresponding to the same node are extracted from the structured database and input into a neural network model. The neural network model performs signal correlation analysis on the conducted signal spectrum segments and radiated signal spectrum segments in each group of spectrum segments to be detected, obtaining a first analysis result. Then, the neural network model performs signal distance correlation analysis on the radiated signal spectrum segments in the group of spectrum segments to be detected under the same node, obtaining a second analysis result. Based on a corresponding preset threshold, it is determined whether both the first and second analysis results are abnormal. If so, the corresponding radiated signal spectrum segment is determined to be an abnormal signal segment, and distance information is obtained.

8. A power line carrier communication anomaly detection system, characterized in that, The application of the power line carrier communication anomaly detection method as described in any one of claims 1-7 includes: Noise signal acquisition module: Obtains the noise signal spectrum when the communication device under test is turned off; Power line signal acquisition module: acquires the conducted signal spectrum and the radiated signal spectrum at multiple different distances when the communication device under test is turned on, and records the operation process of the communication device under test; the distance information is the distance between the antenna used to acquire the radiated signal spectrum and the power line. Baseline calibration module: Based on the noise signal spectrum, the conducted signal spectrum and the radiated signal spectrum under multiple different distance information are baseline calibrated to obtain the calibrated conducted signal spectrum and radiated signal spectrum; Structured storage module: Based on the operation process of the communication device under test, the calibrated conducted signal spectrum and radiated signal spectrum are extracted into spectrum segments, the corresponding spectrum segments are correlated, multiple groups of spectrum segments to be detected are generated based on the correlation results, and the multiple groups of spectrum segments to be detected are stored in a structured manner. Anomaly detection module: It sequentially performs anomaly detection on several groups of signal segments to be detected using a neural network model, and obtains the corresponding abnormal signal segments and their information based on the detection results.

Images