Abnormal interference source identification method and system, terminal device, and medium

Abstract

The application discloses an abnormal interference source identification method, system, terminal equipment and computer readable storage medium. The steps of the abnormal interference source identification method comprise the following steps: determining an uplink interference cell, and extracting three-dimensional position information and signal characteristic information of the uplink interference cell; generating three-dimensional point cloud data according to the three-dimensional position information and the signal characteristic information, and segmenting the three-dimensional point cloud data to obtain a plurality of point cloud blocks; determining a position set of an abnormal interference source based on each point cloud block, so as to identify the abnormal interference source. The application can improve the positioning and identification accuracy of the abnormal interference source.

Description

Technical Field

[0001] This invention relates to the field of terminal equipment technology, and in particular to a method, system, terminal equipment, and computer-readable storage medium for identifying abnormal interference sources. Background Technology

[0002] With the widespread adoption of smart terminals and the continuous expansion of wireless networks and their user base, network interference is becoming increasingly diverse and complex. External interference, in particular, is characterized by its high intensity and difficulty in locating sources, severely impacting network quality and user experience. Currently, the network is in a period of maturing 4G networks and rapid development of 5G networks. Some weak coverage issues exist in the network to varying degrees. In such situations, some users choose to privately install inferior signal amplification equipment to improve signal strength. While this method may improve the communication quality for the individual installer, the technical specifications of these privately installed signal amplifiers are often poor, and they have not undergone professional testing. The equipment operates in a non-linear range, causing uplink noise and intermodulation interference, severely affecting the communication experience of other users in the vicinity.

[0003] Currently, 5G interference issues are quite serious and are showing a rapid growth trend, while 4G networks are also plagued by similar problems. Currently, the location of interference sources is mainly determined by aggregating the locations of the affected cells to obtain a rough two-dimensional position. Field test personnel then use frequency scanning equipment to locate the interference source within this approximate area and conduct on-site inspections.

[0004] Existing interference source location technologies have the following problems: There is currently no specific analytical method to determine whether there are privately installed 4G / 5G wireless amplifiers in a certain area, making it impossible to centrally address the aforementioned interference issues in the network; the existing location methods based on backend data have severely insufficient accuracy and are still two-dimensional; the existing location methods cannot provide an approximate number of privately installed signal amplifier devices, resulting in incomplete on-site handling.

[0005] In summary, existing interference source localization technologies suffer from a lack of localization methods, inaccurate localization, and incomplete localization information. Summary of the Invention

[0006] The main objective of this invention is to provide a method, system, terminal device, and computer-readable storage medium for identifying abnormal interference sources, aiming to improve the accuracy of locating and identifying abnormal interference sources.

[0007] To achieve the above objectives, the present invention provides a method for identifying abnormal interference sources, wherein the identification of abnormal interference sources includes:

[0008] Identify the uplink interfering cell and extract its three-dimensional location information and signal feature information;

[0009] Three-dimensional point cloud data is generated based on the three-dimensional position information and the signal feature information, and the three-dimensional point cloud data is segmented to obtain multiple point cloud blocks;

[0010] The location set of abnormal interference sources is determined based on each of the point cloud blocks in order to identify the abnormal interference sources.

[0011] Optionally, the step of extracting the three-dimensional location information and signal feature information of the uplink interfering cell includes:

[0012] By using the minimized route finding technique, the three-dimensional location information of the uplink interfering cell, including longitude, latitude and altitude, and the signal characteristic information, including uplink signal strength, uplink signal-to-noise ratio and terminal transmit power, are extracted from the preset positioning platform.

[0013] Optionally, the step of extracting the three-dimensional location information and signal feature information of the uplink interfering cell includes:

[0014] The preset OTT positioning data is associated with the network data of the uplink interfering cell to obtain three-dimensional location information including longitude, latitude and altitude, and signal characteristic information including uplink signal strength, uplink signal-to-noise ratio and terminal transmission power. The OTT positioning data is the location information data generated by the uplink interfering cell for Internet services.

[0015] Optionally, the step of generating three-dimensional point cloud data based on the three-dimensional position information and the signal feature information includes:

[0016] The longitude, latitude, and longitude of the uplink interfering cell are determined as a three-dimensional location representation;

[0017] The uplink signal strength, the uplink signal-to-noise ratio, and the terminal transmit power are mapped to a preset space in a uniform distribution manner to obtain the color RGB representation of the point cloud data;

[0018] The three-dimensional point cloud data is determined based on the three-dimensional position representation and the color RGB representation.

[0019] Optionally, the step of segmenting the three-dimensional point cloud data to obtain multiple point cloud blocks includes:

[0020] The 3D point cloud data is divided into multiple sub-regions using PointNet++, and high-dimensional cloud points are obtained by feature extraction and iteration of the sub-regions using PointNet. The high-dimensional cloud points are then subjected to inverse distance weight difference to obtain the corresponding low-dimensional cloud points.

[0021] Feature fusion and feature extraction are performed on the low-dimensional cloud points, and the weights of each cloud point after feature fusion and feature extraction are globally normalized to obtain multiple point cloud blocks.

[0022] Optionally, the step of determining the location set of the abnormal interference sources based on each of the point cloud blocks to complete the identification of the abnormal interference sources includes:

[0023] Obtain the center position and uplink feature value of each point cloud block;

[0024] The point cloud blocks are classified based on the feature values, and abnormal point cloud blocks are identified from the classified point cloud blocks according to the center position.

[0025] The location set of abnormal interference sources is determined based on the three-dimensional center position of the abnormal point cloud block, so as to complete the identification of the abnormal interference sources.

[0026] Optionally, the center location includes: longitude center location, latitude center location, and altitude center location, and the uplink characteristic values ​​include: uplink average signal strength, uplink average signal-to-noise ratio, and uplink terminal average transmit power.

[0027] The step of classifying the point cloud blocks based on the feature values ​​and determining abnormal point cloud blocks from the classified point cloud blocks according to the center position includes:

[0028] Based on the uplink average signal strength, uplink average signal-to-noise ratio, and uplink terminal average transmit power of each point cloud block, the point cloud blocks are grouped according to a preset classification label to obtain multiple target point cloud block groups. The target point cloud block groups include: a first target point cloud block group whose uplink average signal strength and uplink average signal-to-noise ratio satisfy a preset first classification threshold, and a second target point cloud block group whose uplink average signal strength and uplink average signal-to-noise ratio satisfy a preset second classification threshold.

[0029] The longitude center position, latitude center position, and altitude center position are transformed into approximately Cartesian equivalent points.

[0030] Based on the longitude center position, latitude center position, and altitude center position after approximate Cartesian equivalent point transformation, the KNN algorithm is used to determine the proportion of point cloud blocks belonging to the second target point cloud block group among the preset number of point cloud blocks closest to the point cloud block to be detected in the first target point cloud block group. When the proportion exceeds a preset proportion threshold, the point cloud block to be detected is identified as an abnormal point cloud block.

[0031] Optionally, the step of determining the uplink interfering cell includes:

[0032] The interference level indicators of each cell are collected and imported into a preset database. The average interference level within a preset time period is then selected from the preset database, and the cell corresponding to the average interference level that meets the uplink interference cell screening criteria is identified as the uplink interference cell.

[0033] To achieve the above objectives, the present invention also provides a system for identifying abnormal interference sources, the system comprising:

[0034] An information extraction module is used to identify uplink interfering cells and extract their three-dimensional location information and signal characteristic information.

[0035] The segmentation module is used to generate three-dimensional point cloud data based on the three-dimensional position information and the signal feature information, and to segment the three-dimensional point cloud data to obtain multiple point cloud blocks.

[0036] An abnormal interference source identification module is used to determine the location set of the abnormal interference sources based on each point cloud block, so as to identify the abnormal interference sources.

[0037] In this invention, each functional module of the abnormal interference source identification system implements the steps of the abnormal interference source identification method described above during operation.

[0038] To achieve the above objectives, the present invention also provides a terminal device, the terminal device comprising: a memory, a processor, and an identification program for abnormal interference sources stored in the memory and executable on the processor, wherein the identification program for abnormal interference sources, when executed by the processor, implements the steps of the abnormal interference source identification method as described above.

[0039] Furthermore, to achieve the above objectives, the present invention also proposes a computer-readable storage medium storing an identification program for abnormal interference sources, wherein the identification program for abnormal interference sources, when executed by a processor, implements the steps of the abnormal interference source identification method as described above.

[0040] In addition, to achieve the above objectives, the present invention also provides a computer program product, which includes a computer program that, when executed by a processor, implements the steps of the method for identifying abnormal interference sources as described above.

[0041] This invention provides a method, system, terminal device, computer-readable storage medium, and computer program product for identifying abnormal interference sources. The method for identifying abnormal interference sources includes the following steps: determining an uplink interfering cell and extracting its three-dimensional location information and signal feature information; generating three-dimensional point cloud data based on the three-dimensional location information and the signal feature information, and segmenting the three-dimensional point cloud data to obtain multiple point cloud blocks; determining the location set of abnormal interference sources based on each point cloud block to identify the abnormal interference source.

[0042] Compared to existing methods for identifying abnormal interference sources, this invention uses the extracted three-dimensional location information and signal feature information of the uplink interfering cell as a sample analysis system for abnormal interference sources, significantly reducing analysis costs. The uplink point cloud feature scene segmentation method improves interference localization from two-dimensional to three-dimensional, enhancing the accuracy of abnormal interference source location. By identifying the location points of suspected privately installed abnormal interference sources within the area, centralized localization of problems within the area is achieved, and the number of abnormal interference sources is further determined, improving the efficiency of investigating abnormal interference sources. Attached Figure Description

[0043] Figure 1 This is a schematic diagram of the hardware operating environment involved in the embodiments of the present invention;

[0044] Figure 2 This is a flowchart illustrating an embodiment of the method for identifying abnormal interference sources according to the present invention.

[0045] Figure 3 This is a schematic diagram of the overall process of an abnormal interference source identification system according to an embodiment of the abnormal interference source identification method of the present invention;

[0046] Figure 4 This is a schematic diagram of three-dimensional point cloud data segmentation, representing an embodiment of the method for identifying abnormal interference sources according to the present invention.

[0047] Figure 5 This is a functional module diagram of an embodiment of the abnormal interference source identification system of the present invention.

[0048] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0049] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0050] like Figure 1 As shown, Figure 1 This is a schematic diagram of the device structure of the hardware operating environment involved in the embodiments of the present invention.

[0051] It should be noted that the terminal device in the embodiments of the present invention can be a terminal device used to identify abnormal interference sources. Specifically, the terminal device can be a mobile phone, computer, server, or network device, etc.

[0052] like Figure 1 As shown, the device may include: a processor 1001, such as a CPU; a network interface 1004; a user interface 1003; a memory 1005; and a communication bus 1002. The communication bus 1002 is used to enable communication between these components. The user interface 1003 may include a display screen or an input unit such as a keyboard; optionally, the user interface 1003 may also include a standard wired interface or a wireless interface. The network interface 1004 may optionally include a standard wired interface or a wireless interface (such as a Wi-Fi interface). The memory 1005 may be high-speed RAM or non-volatile memory, such as a disk drive. Optionally, the memory 1005 may also be a storage device independent of the aforementioned processor 1001.

[0053] Those skilled in the art will understand that Figure 1 The device structure shown does not constitute a limitation on the device and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0054] like Figure 1 As shown, the memory 1005, as a computer storage medium, may include an operating system, a network communication module, a user interface module, and a program for identifying abnormal interference sources. The operating system is a program that manages and controls the hardware and software resources of the device, supporting the operation of the program for identifying abnormal interference sources and other software or programs. Figure 1 In the device shown, the user interface 1003 is mainly used for data communication with the client; the network interface 1004 is mainly used for establishing a communication connection with the server; and the processor 1001 can be used to call the abnormal interference source identification program stored in the memory 1005 and perform the following operations:

[0055] Identify the uplink interfering cell and extract its three-dimensional location information and signal feature information;

[0056] Three-dimensional point cloud data is generated based on the three-dimensional position information and the signal feature information, and the three-dimensional point cloud data is segmented to obtain multiple point cloud blocks;

[0057] The location set of abnormal interference sources is determined based on each of the point cloud blocks in order to identify the abnormal interference sources.

[0058] Furthermore, the processor 1001 can also be used to call the identification program for abnormal interference sources stored in the memory 1005, and also perform the following operations:

[0059] By using the minimized route finding technique, the three-dimensional location information of the uplink interfering cell, including longitude, latitude and altitude, and the signal characteristic information, including uplink signal strength, uplink signal-to-noise ratio and terminal transmit power, are extracted from the preset positioning platform.

[0060] Furthermore, the processor 1001 can also be used to call the identification program for abnormal interference sources stored in the memory 1005, and also perform the following operations:

[0061] The preset OTT positioning data is associated with the network data of the uplink interfering cell to obtain three-dimensional location information including longitude, latitude and altitude, and signal characteristic information including uplink signal strength, uplink signal-to-noise ratio and terminal transmission power. The OTT positioning data is the location information data generated by the uplink interfering cell for Internet services.

[0062] Furthermore, the processor 1001 can also be used to call the identification program for abnormal interference sources stored in the memory 1005, and also perform the following operations:

[0063] The longitude, latitude, and longitude of the uplink interfering cell are determined as a three-dimensional location representation;

[0064] The uplink signal strength, the uplink signal-to-noise ratio, and the terminal transmit power are mapped to a preset space in a uniform distribution manner to obtain the color RGB representation of the point cloud data;

[0065] The three-dimensional point cloud data is determined based on the three-dimensional position representation and the color RGB representation.

[0066] Furthermore, the processor 1001 can also be used to call the identification program for abnormal interference sources stored in the memory 1005, and also perform the following operations:

[0067] The 3D point cloud data is divided into multiple sub-regions using PointNet++, and high-dimensional cloud points are obtained by feature extraction and iteration of the sub-regions using PointNet. The high-dimensional cloud points are then subjected to inverse distance weight difference to obtain the corresponding low-dimensional cloud points.

[0068] Feature fusion and feature extraction are performed on the low-dimensional cloud points, and the weights of each cloud point after feature fusion and feature extraction are globally normalized to obtain multiple point cloud blocks.

[0069] Furthermore, the processor 1001 can also be used to call the identification program for abnormal interference sources stored in the memory 1005, and also perform the following operations:

[0070] Obtain the center position and uplink feature value of each point cloud block;

[0071] The point cloud blocks are classified based on the feature values, and abnormal point cloud blocks are identified from the classified point cloud blocks according to the center position.

[0072] The location set of abnormal interference sources is determined based on the three-dimensional center position of the abnormal point cloud block, so as to complete the identification of the abnormal interference sources.

[0073] Furthermore, the center location includes: longitude center location, latitude center location, and altitude center location, and the uplink characteristic values ​​include: uplink average signal strength, uplink average signal-to-noise ratio, and uplink terminal average transmit power;

[0074] Processor 1001 can also be used to call the identification program for abnormal interference sources stored in memory 1005, and also perform the following operations:

[0075] Based on the uplink average signal strength, uplink average signal-to-noise ratio, and uplink terminal average transmit power of each point cloud block, the point cloud blocks are grouped according to a preset classification label to obtain multiple target point cloud block groups. The target point cloud block groups include: a first target point cloud block group whose uplink average signal strength and uplink average signal-to-noise ratio satisfy a preset first classification threshold, and a second target point cloud block group whose uplink average signal strength and uplink average signal-to-noise ratio satisfy a preset second classification threshold.

[0076] The longitude center position, latitude center position, and altitude center position are transformed into approximately Cartesian equivalent points.

[0077] Based on the longitude center position, latitude center position, and altitude center position after approximate Cartesian equivalent point transformation, the KNN algorithm is used to determine the proportion of point cloud blocks belonging to the second target point cloud block group among the preset number of point cloud blocks closest to the point cloud block to be detected in the first target point cloud block group. When the proportion exceeds a preset proportion threshold, the point cloud block to be detected is identified as an abnormal point cloud block.

[0078] Furthermore, the processor 1001 can also be used to call the identification program for abnormal interference sources stored in the memory 1005, and also perform the following operations:

[0079] The interference level indicators of each cell are collected and imported into a preset database. The average interference level within a preset time period is then selected from the preset database, and the cell corresponding to the average interference level that meets the uplink interference cell screening criteria is identified as the uplink interference cell.

[0080] Reference Figure 2 , Figure 2 This is a flowchart illustrating the first embodiment of the method for identifying abnormal interference sources according to the present invention.

[0081] This embodiment provides an example of a method for identifying abnormal interference sources. It should be noted that although the logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than that shown here.

[0082] This embodiment uses PointNet++ to perform 3D localization of abnormal interference sources in 4G / 5G signals for identification. Deep learning has become a powerful tool in the field of computer vision, especially in image processing, where convolutional neural network-based deep learning methods have dominated most problems. However, research on deep learning methods for unordered point cloud data has progressed relatively slowly. This is mainly because point clouds have three characteristics: disorder, sparsity, and limited information content. In the past, when processing point clouds using deep learning methods, they were often converted into more regular formats such as depth images or voxels from a specific viewpoint to facilitate the definition of weight-sharing convolution operations. PointNet is a deep learning method that directly processes point clouds, and it is a pioneer in the application of deep learning to 3D point sets. However, PointNet cannot capture the local structure generated by points in metric space, thus limiting its ability to identify and classify fine-grained patterns and its versatility in complex scenes. PointNet++ improves upon this by introducing a multi-layered neural network model. This model iteratively applies PointNet to the embedded grouping of the input point set. By measuring spatial distance, the deep network can learn local features by increasing the relevant scale. During the learning process, point sets are typically sampled at varying densities, which significantly reduces the network's performance when training on point clouds with uniform density. PointNet++, however, effectively learns features from deep point sets. PointNet++ also significantly outperforms other techniques in challenging 3D point cloud benchmarks. Currently, PointNet++ is primarily applied in 3D object recognition, object classification, scene semantic segmentation, and point cloud data segmentation.

[0083] In this embodiment, as Figure 3 As shown, firstly, cells within the analyzed area are filtered based on the average uplink interference level during busy hours. For cells whose uplink interference levels reach the threshold, their 3D positioning data and associated signal feature data are extracted from the positioning platform and stored in the database. Then, the 3D location information data and interference feature data of users in the interfering cells are sorted to obtain 3D point cloud data. Next, the PointNet++ network is used to segment the 3D point cloud data of the user's location to generate point cloud blocks for different scenarios. Finally, the nearest neighbor algorithm is used to detect the point cloud blocks and identify the location of abnormal interference sources.

[0084] Step S10: Identify the uplink interfering cell and extract its three-dimensional location information and signal feature information;

[0085] The terminal device filters out uplink interfering cells from multiple cells. In order to identify the abnormal interference source of 4G / 5G signals from multiple uplink interfering cells, it is necessary to further determine the three-dimensional location information and signal characteristic information of each uplink interfering cell.

[0086] Step S20: Generate three-dimensional point cloud data based on the three-dimensional position information and the signal feature information, and segment the three-dimensional point cloud data to obtain multiple point cloud blocks;

[0087] After acquiring the three-dimensional location information and corresponding signal feature information of the uplink interfering cell, the terminal device will determine the three-dimensional point cloud data of the uplink interfering cell based on the three-dimensional location information and signal feature information, and further segment the three-dimensional point cloud data to obtain multiple point cloud blocks.

[0088] It should be noted that, in this embodiment, point cloud data refers to a set of vectors in a three-dimensional coordinate system. These vectors are typically represented in the form of X, Y, Z three-dimensional coordinates, and are generally used to represent the outer surface shape of an object. Furthermore, in addition to the geometric position information represented by (X, Y, Z), point cloud data can also represent the RGB color, grayscale value, depth, segmentation result, etc., of a point. For example, Pi = {Xi, Yi, Zi, ...} represents a point in space, while Point Cloud = {P1, P2, P3, ..., Pn} represents a set of point cloud data.

[0089] Step S30: Determine the location set of the abnormal interference source based on each point cloud block to identify the abnormal interference source.

[0090] After the terminal device divides the 3D point cloud data into multiple point cloud blocks, it performs operations such as classification and distance calculation on the multiple point cloud blocks. It can then determine the location set of abnormal interference sources from the 3D center position of the multiple point cloud blocks, thereby realizing the identification of abnormal interference sources and improving the efficiency of technicians in investigating abnormal interference sources.

[0091] In this embodiment, the terminal device filters out uplink interfering cells from multiple cells, further determines the three-dimensional location information and signal feature information of each uplink interfering cell, and performs operations such as classification and distance calculation on multiple point cloud blocks. It can determine the location set of abnormal interference sources from the three-dimensional center position of multiple point cloud blocks to realize the identification of abnormal interference sources. The three-dimensional point cloud data of the uplink interfering cell is determined based on the three-dimensional location information and signal feature information, and the three-dimensional point cloud data is further segmented to obtain multiple point cloud blocks.

[0092] Compared to existing methods for identifying abnormal interference sources, this invention uses the extracted three-dimensional location information and signal feature information of uplink interfering cells as a sample analysis system for analyzing abnormal interference sources from illegally installed 4G / 5G signal amplifiers, significantly reducing analysis costs. Based on the uplink point cloud feature scene segmentation method, interference localization is improved from two-dimensional to three-dimensional, enhancing the accuracy of abnormal interference source localization. By determining the location points of suspected illegally installed 4G / 5G signal amplifiers (i.e., abnormal interference sources) within the area, centralized localization of problems within the area is achieved, and the number of abnormal interference sources is further determined, improving the efficiency of investigating abnormal interference sources.

[0093] Furthermore, based on the first embodiment of the method for identifying abnormal interference sources of the present invention described above, a second embodiment of the method for identifying abnormal interference sources of the present invention is proposed.

[0094] Compared to the first embodiment, in this embodiment, step S10 above, "determining the uplink interfering cell," may include:

[0095] Step S101: Collect interference level indicators of each cell and import the interference level indicators into a preset database. Then, select the average interference level within a preset time period from the preset database and determine the cell corresponding to the average interference level that meets the uplink interference cell screening criteria as the uplink interference cell.

[0096] In order to obtain uplink interfering cells, the terminal equipment first collects the uplink interference level index of each cell and imports the uplink interference level index into the database. Then, it extracts the average uplink interference level of each cell in the corresponding time period from the database, and then determines the cell corresponding to the average interference level that meets the uplink interfering cell screening criteria as the uplink interfering cell.

[0097] Specifically, for example, at 1:00 AM every day, the uplink interference level index of each cell at the hourly granularity of the previous day is extracted from the operator's network management big data platform and stored in the database; then, the average uplink interference level of each cell for 6 hours from 8:00 AM to 11:00 AM and from 6:00 PM to 9:00 PM is extracted; if the average uplink interference level is greater than or equal to int_up_high, then the cell with the average uplink interference level at this time is identified as the uplink interference cell. The threshold int_up_high can be flexibly set.

[0098] Further, in step S10 above, "extracting the three-dimensional location information and signal feature information of the uplink interfering cell" includes:

[0099] Step S102: Using the minimized route finding technique, extract the three-dimensional location information of the uplink interfering cell, including longitude, latitude and altitude, and the signal characteristic information, including uplink signal strength, uplink signal-to-noise ratio and terminal transmit power, from the preset positioning platform.

[0100] It should be noted that in this embodiment, the three-dimensional location information of the uplink interfering cell includes longitude, latitude, and altitude, and the signal characteristic information of the uplink interfering cell includes uplink signal strength (ScRsrpUL), uplink signal-to-noise ratio (ScSinrUL), and terminal transmit power (ScPHR). Minimization of Drive Tests (MDT) is a minimal drive test technique introduced in 3GPP 4G R13. It requires reporting the terminal's GPS latitude and longitude information and has significant advantages for precise location service applications. MDT technology is being treated as one of the most prioritized technologies in 5G R16. This positioning data has the characteristics of wide applicability to protocol standards, high accuracy due to the inclusion of multiple positioning information, large data volume, and low correlation difficulty.

[0101] Specifically, for example, after identifying the uplink interfering cell, the terminal device uses MDT (Multi-Targeting Technology) to obtain the three-dimensional location information of the uplink interfering cell, including longitude, latitude, and altitude, and signal characteristic information including ScRsrpUL, ScSinrUL, and ScPHR, from a preset positioning platform. In addition to the above parameters, the terminal device can also obtain the time (TimeStamp) for generating the measurement sampling point, the terminal identification ID (UeId), and the cell's globally unique identifier (Ecgi). In this embodiment, the preset positioning platform is a platform used to store user location information and its wireless network characteristic information.

[0102] Furthermore, in step S10 above, "extracting the three-dimensional location information and signal feature information of the uplink interfering cell" also includes:

[0103] Step S103: Associate the preset OTT positioning data with the network data of the uplink interfering cell to obtain three-dimensional location information including longitude, latitude and altitude, and signal characteristic information including uplink signal strength, uplink signal-to-noise ratio and terminal transmission power. The OTT positioning data is the location information data generated by the uplink interfering cell for Internet services.

[0104] OTT (Over The Top) location data refers to location information data generated by users when conducting internet services. By associating this data with the user's network data, such as MR (Measurement Report) data, feature data that can be used to build a fingerprint database is formed.

[0105] Specifically, for example, terminal devices can correlate OTT data with MR data from uplink interfering cells, thereby obtaining three-dimensional location information including latitude, longitude, and altitude, as well as signal characteristic information including ScRsrpUL, ScSinrUL, and ScPHR. Operators' user positioning capabilities are gradually evolving from two-dimensional to three-dimensional.

[0106] In this embodiment, to obtain uplink interfering cells, the terminal device first collects the uplink interference level indicators of each cell and imports these indicators into a database. Then, it extracts the average uplink interference level of each cell for the corresponding time period from the database. Finally, it identifies the cell with the average interference level that meets the uplink interfering cell screening criteria as the uplink interfering cell. Furthermore, the terminal device can obtain three-dimensional location information including latitude, longitude, and altitude, and signal characteristic information including ScRsrpUL, ScSinrUL, and ScPHR, through MDT; alternatively, the terminal device can also associate OTT data with the MR data of the uplink interfering cell to similarly obtain three-dimensional location information including latitude, longitude, and altitude, and signal characteristic information including ScRsrpUL, ScSinrUL, and ScPHR.

[0107] In this invention, cells with average interference levels that meet the uplink interference cell screening criteria are identified as uplink interference cells. Based on MDT or OTT data, the three-dimensional location information (latitude, longitude, altitude, etc.) and signal characteristic information (ScRsrpUL, ScSinrUL, ScPHR, etc.) of each uplink interference cell are obtained. Therefore, this invention fully utilizes existing positioning data, generating point cloud data containing three-dimensional location information and uplink signal characteristic information through data processing. This reduces the cost of acquiring analysis data and thus improves the accuracy and efficiency of identifying abnormal interference sources.

[0108] Furthermore, based on the first and second embodiments of the method for identifying abnormal interference sources of the present invention described above, a third embodiment of the method for identifying abnormal interference sources of the present invention is proposed.

[0109] In this embodiment, step S20 above, "generating three-dimensional point cloud data based on the three-dimensional position information and the signal feature information," may include:

[0110] Step S201: Determine the longitude, latitude, and longitude of the uplink interfering cell as a three-dimensional location representation;

[0111] Step S202: The uplink signal strength, the uplink signal-to-noise ratio, and the terminal transmit power are mapped to a preset space in a uniform distribution manner to obtain the color RGB representation of the point cloud data;

[0112] Step S203: Determine the three-dimensional point cloud data based on the three-dimensional position representation and the color RGB representation.

[0113] After acquiring the three-dimensional location information (latitude, longitude, altitude, etc.) and signal characteristic information (ScRsrpUL, ScSinrUL, and ScPHR) of the uplink interfering cell, the terminal device uses Longitude, Latitude, and Altitude as the three-dimensional location representation of the three-dimensional point cloud data. ScRsrpUL, ScSinrUL, and ScPHR are mapped to the 0-255 range using a uniform distribution to obtain the RGB color representation of the three-dimensional point cloud data. In this embodiment, to ensure sufficient data, data information from the uplink interfering cell within day t can be used. The value of t can be flexibly set according to the actual situation; in this embodiment, t is set to 7 by default.

[0114] Specifically, for example, when mapping ScRsrpUL, ScSinrUL, and ScPHR to the 0–255 range using a uniform distribution, the output values ​​after mapping are:

[0115] X new =255*(XX) min ) / (Xmax –X min )

[0116] Among them, X min X max X and X represent the minimum value, maximum value, and sample values, respectively.

[0117] Furthermore, in step S20 above, "segmenting the three-dimensional point cloud data to obtain multiple point cloud blocks" may include:

[0118] Step S204: Divide the three-dimensional point cloud data into multiple sub-regions using PointNet++, extract and iterate features for the sub-regions using PointNet to obtain high-dimensional cloud points, and perform inverse distance weight difference on the high-dimensional cloud points to obtain the corresponding low-dimensional cloud points.

[0119] Step S205: Perform feature fusion and feature extraction on the low-dimensional cloud points, and perform global normalization on the weights of each low-dimensional cloud point after feature fusion and feature extraction to obtain multiple point cloud blocks.

[0120] It should be noted that, in this embodiment, as Figure 4 As shown, after determining the three-dimensional point cloud data based on the three-dimensional location information and signal feature information, the three-dimensional point cloud data segmentation capability of PointNet++ is used to segment the point cloud data with uplink signal feature information of the cell, and multiple point cloud blocks are generated in three-dimensional space based on their location attribute features and uplink RGB features.

[0121] Specifically, for example, the input includes three-dimensional location features such as longitude, latitude, and altitude, as well as uplink signal features such as uplink signal strength, uplink signal-to-noise ratio, and mobile phone transmission power, totaling six channels of point cloud data. PointNet++ first samples and divides the point cloud into regions. Within each small region, it uses a basic PointNet network to extract features and iterates to obtain high-dimensional cloud points. Then, it performs inverse distance interpolation on the high-dimensional cloud points to obtain corresponding low-dimensional cloud points. Next, it performs feature fusion and feature extraction on these low-dimensional cloud points to obtain high-dimensional cloud points. Finally, it performs global normalization on the weights of each low-dimensional cloud point after feature fusion and feature extraction to obtain multiple point cloud blocks, ultimately segmenting the data into N point cloud blocks. In this embodiment, it can achieve three-dimensional clustering and segmentation of samples with similar uplink features and location features in the analyzed region.

[0122] In this embodiment, the terminal device uses Longitude, Latitude, and Altitude as the three-dimensional position representations of the three-dimensional point cloud data. ScRsrpUL, ScSinrUL, and ScPHR are uniformly distributed and mapped to the 0-255 range to obtain the RGB color representation of the three-dimensional point cloud data. Based on the three-dimensional position representation and the RGB color representation, the three-dimensional point cloud data is determined. Then, PointNet++ is used to segment the point cloud data containing signal feature information, generating multiple point cloud blocks in three-dimensional space based on their positional attributes and uplink RGB features. This invention enables uplink scene segmentation using the PointNet++ deep network, increasing the dimensionality of uplink interference localization from two-dimensional to three-dimensional, thus improving the accuracy and efficiency of identifying abnormal interference sources.

[0123] Furthermore, based on the first, second, and third embodiments of the method for identifying abnormal interference sources of the present invention described above, a fourth embodiment of the method for identifying abnormal interference sources of the present invention is proposed.

[0124] In this embodiment, step S30 above, "determining the location set of the abnormal interference source based on each of the point cloud blocks to complete the identification of the abnormal interference source," may include:

[0125] Step S301: Obtain the center position and uplink feature value of each point cloud block;

[0126] Step S302: Classify the point cloud blocks based on the feature values, and determine abnormal point cloud blocks from the classified point cloud blocks according to the center position;

[0127] Step S303: Determine the location set of abnormal interference sources based on the three-dimensional center position of the abnormal point cloud block, so as to complete the identification of the abnormal interference sources.

[0128] After segmenting the 3D point cloud data into multiple point cloud blocks, the terminal device acquires the center position and uplink feature value of each point cloud block. Then, it classifies the point cloud blocks based on the uplink feature value of each point cloud block, and judges the classified point cloud blocks based on the center position of each point cloud block to identify abnormal point cloud blocks from multiple classified point cloud blocks. After identifying the abnormal point cloud blocks, it obtains the 3D center position of the abnormal point cloud block. At this time, the 3D center position of the abnormal point cloud block is the location set of abnormal interference sources, thereby realizing the identification of abnormal interference sources.

[0129] Further, in step S302 above, "classifying the point cloud blocks based on the feature values, and determining abnormal point cloud blocks from the classified point cloud blocks according to the center position" may include:

[0130] Step S3021: Based on the uplink average signal strength, uplink average signal-to-noise ratio and uplink terminal average transmit power of each point cloud block, the point cloud blocks are grouped according to a preset classification label to obtain multiple target point cloud block groups. The target point cloud block groups include: a first target point cloud block group whose uplink average signal strength and uplink average signal-to-noise ratio satisfy a preset first classification threshold, and a second target point cloud block group whose uplink average signal strength and uplink average signal-to-noise ratio satisfy a preset second classification threshold.

[0131] Step S3022: Perform an approximate Cartesian equivalent point transformation on the longitude center position, latitude center position, and altitude center position;

[0132] Step S3023: Based on the longitude center position, latitude center position, and altitude center position after the approximate Cartesian equivalent point transformation, the KNN algorithm is used to determine the proportion of point cloud blocks belonging to the second target point cloud block group among the preset number of point cloud blocks closest to the point cloud block to be detected in the first target point cloud block group. When the proportion exceeds a preset proportion threshold, the point cloud block to be detected is identified as an abnormal point cloud block.

[0133] It should be noted that, in this embodiment, the center location includes: longitude center location, latitude center location, and altitude center location. In this embodiment, the longitude center location is the average longitude of the point cloud block, and similarly, the latitude center location is the average latitude of the point cloud block, and the altitude center location is the average altitude of the point cloud block. The uplink characteristic values ​​include: uplink average signal strength, uplink average signal-to-noise ratio, and uplink terminal average transmit power.

[0134] The terminal device groups each point cloud block based on the uplink average signal strength, uplink average signal-to-noise ratio, and uplink terminal average transmit power, according to a preset classification label. Point cloud blocks whose uplink average signal strength and uplink average signal-to-noise ratio meet a preset first classification threshold are identified as the first target point cloud block group, and point cloud blocks whose uplink average signal strength and uplink average signal-to-noise ratio meet a preset second classification threshold are identified as the second target point cloud block group. That is, the signal strength and signal quality (i.e., signal-to-noise ratio) of the point cloud blocks in the first target point cloud block group are both very good, while the signal strength and signal quality of the point cloud blocks in the second target point cloud block group are both poor.

[0135] The preset first classification threshold is:

[0136] RsrpUL>rsrp_high,SinrUL>sinr_high

[0137] The preset threshold for the second category is:

[0138] RsrpUL <rsrp_low,SinrUL<sinr_low

[0139] Where RsrpUL is the uplink average signal strength and SinrUL is the uplink average signal-to-noise ratio. Furthermore, in this embodiment, point cloud blocks that do not meet either the preset first classification threshold or the preset second classification threshold are identified as the third target point cloud block group.

[0140] It should be noted that, in this embodiment, the characteristic of the illegally installed signal amplifier is that it occupies the signal of the same cell, resulting in most users having poor signal strength and quality in adjacent locations where users are concentrated, while a small number of users have good signal strength and quality. Therefore, in this embodiment, the KNN (K-Nearest Neighbor) algorithm can be used to detect point cloud blocks that are marked as the first target point cloud block group with good cell signal strength and quality, and to determine whether the K nearest point cloud blocks are mostly point cloud blocks of the second target point cloud block group with poor signal strength and quality. This identifies the isolated feature and thus locates the specific three-dimensional location information of the illegally installed signal amplifier, i.e., the abnormal interference source.

[0141] Specifically, for example, the longitude center position, latitude center position, and altitude center position are transformed into approximate Cartesian equivalent points. Based on the longitude center position, latitude center position, and altitude center position after the approximate Cartesian equivalent point transformation, the KNN algorithm is used to determine the proportion of point cloud blocks belonging to the second target point cloud block group among the preset number of point cloud blocks closest to the point cloud block to be detected in the first target point cloud block group. If this proportion exceeds a preset proportion threshold, the point cloud block to be detected is identified as an abnormal point cloud block. In this embodiment, the preset number is set to 10, but it can also be flexibly set according to the actual situation. The preset proportion threshold can also be flexibly set according to the actual situation.

[0142] It should be noted that, in this embodiment, when performing approximate Cartesian equivalent point transformations for the longitude center, latitude center, and altitude center, the following rules are followed:

[0143] x = Alt * cos(Lat) * sin(Lon)

[0144] y = Alt*sin(Lat)

[0145] z = Alt * cos(Lat) * cos(Lon)

[0146] Where Alt is the center of elevation, Lat is the center of latitude, and Lon is the center of longitude.

[0147] In this embodiment, the terminal device groups point cloud blocks based on uplink average signal strength, uplink average signal-to-noise ratio, and uplink terminal average transmit power, according to preset classification labels. Point cloud blocks with good signal strength and quality are marked as the first target point cloud block group, and point cloud blocks with poor signal strength and quality are marked as the second target point cloud block group. The longitude center position, latitude center position, and altitude center position are approximated by Cartesian equivalent point transformation. Based on the longitude center position, latitude center position, and altitude center position after the approximation by Cartesian equivalent point transformation, the KNN algorithm is used to determine the proportion of point cloud blocks belonging to the second target point cloud block group among the preset number of point cloud blocks closest to the point cloud block to be detected in the first target point cloud block group. If this proportion exceeds a preset proportion threshold, the point cloud block to be detected is identified as an abnormal point cloud block. After identifying the abnormal point cloud block, the three-dimensional center position of the abnormal point cloud block is obtained. At this time, the three-dimensional center position of the abnormal point cloud block is the location set of abnormal interference sources, thus completing the identification of abnormal interference sources.

[0148] In this invention, point cloud blocks are classified according to preset classification labels, and the KNN algorithm is used to detect the classified point cloud blocks to identify abnormal point cloud blocks from multiple point cloud blocks. The three-dimensional center position of the abnormal point cloud block is then determined as the location set of abnormal interference sources, realizing rapid localization of abnormal interference sources. This allows inspectors to quickly locate the abnormal interference sources and also determines the number of abnormal interference sources, improving the efficiency of abnormal interference source investigation. This invention also fills a gap in existing technology by establishing a complete analysis process for abnormal interference source localization and identification, including data preparation, label preparation, and model recognition.

[0149] Furthermore, embodiments of the present invention also propose a system for identifying abnormal interference sources, referring to... Figure 3 , Figure 3 This is a functional module diagram illustrating an embodiment of the present invention for identifying abnormal interference sources. (See diagram below.) Figure 3 As shown, the abnormal interference source identification system of the present invention includes:

[0150] The information extraction module 10 is used to identify uplink interfering cells and extract the three-dimensional location information and signal feature information of the uplink interfering cells;

[0151] The segmentation module 20 is used to generate three-dimensional point cloud data based on the three-dimensional position information and the signal feature information, and to segment the three-dimensional point cloud data to obtain multiple point cloud blocks;

[0152] The abnormal interference source identification module 30 is used to determine the location set of the abnormal interference sources based on each point cloud block, so as to identify the abnormal interference sources.

[0153] Furthermore, the information extraction module 10 includes:

[0154] The first information extraction unit is used to extract, through the minimized path measurement technique, three-dimensional location information of the uplink interfering cell, including longitude, latitude and altitude, and signal characteristic information including uplink signal strength, uplink signal-to-noise ratio and terminal transmit power, from the preset positioning platform.

[0155] Furthermore, the information extraction module 10 also includes:

[0156] The second information extraction unit is used to associate the preset OTT positioning data with the network data of the uplink interfering cell to obtain three-dimensional location information including longitude, latitude and altitude, and signal feature information including uplink signal strength, uplink signal-to-noise ratio and terminal transmission power. The OTT positioning data is the location information data generated by the uplink interfering cell for Internet services.

[0157] Furthermore, the segmentation module 20 includes:

[0158] A three-dimensional location characterization determination unit is used to determine the longitude, latitude, and latitudinal dimension of the uplink interfering cell as a three-dimensional location characterization.

[0159] A mapping unit is used to map the uplink signal strength, the uplink signal-to-noise ratio, and the terminal transmit power to a preset space in a uniform distribution manner to obtain the color RGB representation of the point cloud data;

[0160] A three-dimensional point cloud data unit is defined, which is used to determine three-dimensional point cloud data based on the three-dimensional position representation and the color RGB representation.

[0161] Furthermore, the segmentation module 20 also includes:

[0162] The inverse distance weighted difference unit is used to divide the three-dimensional point cloud data into multiple sub-regions using PointNet++, and to extract and iterate the features of the sub-regions using PointNet to obtain high-dimensional cloud points, and to perform inverse distance weighted difference on the high-dimensional cloud points to obtain the corresponding low-dimensional cloud points.

[0163] The global normalization processing unit is used to perform feature fusion and feature extraction on the low-dimensional cloud points, and to perform global normalization processing on the weights of each cloud point after feature fusion and feature extraction to obtain multiple point cloud blocks.

[0164] Furthermore, the abnormal interference source identification module 30 includes:

[0165] The acquisition unit is used to acquire the center position and uplink feature value of each point cloud block;

[0166] The abnormal point cloud block unit is used to classify the point cloud blocks based on the feature values, and to determine the abnormal point cloud blocks from the classified point cloud blocks according to the center position.

[0167] An abnormal interference source identification unit is used to determine the location set of abnormal interference sources based on the three-dimensional center position of the abnormal point cloud block, so as to complete the identification of the abnormal interference source.

[0168] Furthermore, the center location includes: longitude center location, latitude center location, and altitude center location, and the uplink characteristic values ​​include: uplink average signal strength, uplink average signal-to-noise ratio, and uplink terminal average transmit power;

[0169] The unit for determining abnormal point cloud blocks includes:

[0170] A grouping subunit is configured to group each point cloud block into multiple target point cloud block groups based on the uplink average signal strength, the uplink average signal-to-noise ratio, and the uplink terminal average transmit power of each point cloud block, according to a preset classification label. The target point cloud block groups include: a first target point cloud block group whose uplink average signal strength and uplink average signal-to-noise ratio satisfy a preset first classification threshold, and a second target point cloud block group whose uplink average signal strength and uplink average signal-to-noise ratio satisfy a preset second classification threshold.

[0171] An approximate Cartesian equivalent point conversion word unit is used to perform an approximate Cartesian equivalent point conversion on the longitude center position, latitude center position, and altitude center position;

[0172] The method for identifying anomalous point cloud block sub-units is based on the longitude center position, latitude center position, and altitude center position after approximate Cartesian equivalent point transformation. Using the KNN algorithm, it determines the proportion of point cloud blocks belonging to the second target point cloud block group among the preset number of point cloud blocks closest to the point cloud block to be detected in the first target point cloud block group. When the proportion exceeds a preset proportion threshold, the point cloud block to be detected is identified as an anomalous point cloud block.

[0173] Furthermore, the information extraction module 10 includes:

[0174] The cell unit is identified as an uplink interference cell. It is used to collect interference level indicators of each cell and import the interference level indicators into a preset database. The average interference level within a preset time period is selected from the preset database, and the cell corresponding to the average interference level that meets the uplink interference cell screening criteria is identified as an uplink interference cell.

[0175] The specific implementation methods of each functional module of the abnormal interference source identification system of the present invention are basically the same as those of the above-described abnormal interference source identification methods, and will not be repeated here.

[0176] Furthermore, this embodiment of the invention also proposes a computer-readable storage medium storing an identification program for abnormal interference sources. When the identification program for abnormal interference sources is executed by a processor, it implements the steps of the abnormal interference source identification method described above.

[0177] The various embodiments of the abnormal interference source identification system and computer-readable storage medium of the present invention can be referred to the various embodiments of the abnormal interference source identification method of the present invention, and will not be repeated here.

[0178] Furthermore, embodiments of the present invention also provide a computer program product, the computer program product including a computer program, which, when executed by a processor, implements the steps of the abnormal interference source identification method as described in any of the embodiments of the abnormal interference source identification method above.

[0179] The specific embodiments of the computer program product of the present invention are basically the same as the embodiments of the above-described method for identifying abnormal interference sources, and will not be described in detail here.

[0180] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0181] The sequence numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0182] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0183] The above are merely preferred embodiments of the present invention and do not limit the scope of the patent. Any equivalent structural or procedural transformations made based on the description and drawings of the present invention, or direct or indirect applications in other related technical fields, are similarly included within the scope of patent protection of the present invention.

Claims

1. A method for identifying abnormal interference sources, characterized in that, The method for identifying abnormal interference sources includes the following steps: Identify the uplink interfering cell and extract its three-dimensional location information and signal feature information. The three-dimensional location information includes longitude, latitude, and altitude, and the signal feature information includes uplink signal strength, uplink signal-to-noise ratio, and terminal transmit power. Three-dimensional point cloud data is generated based on the three-dimensional position information and the signal feature information, and the three-dimensional point cloud data is segmented to obtain multiple point cloud blocks; The location set of abnormal interference sources is determined based on each of the point cloud blocks in order to identify the abnormal interference sources; The step of generating three-dimensional point cloud data based on the three-dimensional position information and the signal feature information includes: The longitude, latitude, and altitude of the uplink interfering cell are determined as a three-dimensional location representation; The uplink signal strength, the uplink signal-to-noise ratio, and the terminal transmit power are mapped to a preset space in a uniform distribution manner to obtain the color RGB representation of the point cloud data, wherein the preset space is a pixel range of 0 to 255; The three-dimensional point cloud data is determined based on the three-dimensional position representation and the color RGB representation; The step of segmenting the three-dimensional point cloud data to obtain multiple point cloud blocks includes: The 3D point cloud data is divided into multiple sub-regions using PointNet++, and high-dimensional cloud points are obtained by feature extraction and iteration of the sub-regions using PointNet. The high-dimensional cloud points are then subjected to inverse distance weight difference to obtain the corresponding low-dimensional cloud points. Feature fusion and feature extraction are performed on the low-dimensional cloud points, and the weights of each low-dimensional cloud point after feature fusion and feature extraction are globally normalized to obtain multiple point cloud blocks. The step of determining the location set of abnormal interference sources based on each of the point cloud blocks to identify the abnormal interference sources includes: Obtain the center position and uplink feature value of each point cloud block; The point cloud blocks are classified based on the feature values, and abnormal point cloud blocks are identified from the classified point cloud blocks according to the center position. The location set of abnormal interference sources is determined based on the three-dimensional center position of the abnormal point cloud block, so as to complete the identification of the abnormal interference sources.

2. The method for identifying abnormal interference sources as described in claim 1, characterized in that, The step of extracting the three-dimensional location information and signal feature information of the uplink interfering cell includes: By using the minimized route finding technique, the three-dimensional location information of the uplink interfering cell, including longitude, latitude and altitude, and the signal characteristic information, including uplink signal strength, uplink signal-to-noise ratio and terminal transmit power, are extracted from the preset positioning platform.

3. The method for identifying abnormal interference sources as described in claim 1, characterized in that, The step of extracting the three-dimensional location information and signal feature information of the uplink interfering cell includes: The preset OTT positioning data is associated with the network data of the uplink interfering cell to obtain three-dimensional location information including longitude, latitude and altitude, and signal characteristic information including uplink signal strength, uplink signal-to-noise ratio and terminal transmission power. The OTT positioning data is the location information data generated by the uplink interfering cell for Internet services.

4. The method for identifying abnormal interference sources as described in claim 1, characterized in that, The center location includes: longitude center location, latitude center location, and altitude center location; the uplink characteristic values ​​include: uplink average signal strength, uplink average signal-to-noise ratio, and uplink terminal average transmit power. The step of classifying the point cloud blocks based on the feature values ​​and determining abnormal point cloud blocks from the classified point cloud blocks according to the center position includes: Based on the uplink average signal strength, uplink average signal-to-noise ratio, and uplink terminal average transmit power of each point cloud block, the point cloud blocks are grouped according to a preset classification label to obtain multiple target point cloud block groups. The target point cloud block groups include: a first target point cloud block group whose uplink average signal strength and uplink average signal-to-noise ratio satisfy a preset first classification threshold, and a second target point cloud block group whose uplink average signal strength and uplink average signal-to-noise ratio satisfy a preset second classification threshold. The longitude center position, latitude center position, and altitude center position are transformed into approximately Cartesian equivalent points. Based on the longitude center position, latitude center position, and altitude center position after approximate Cartesian equivalent point transformation, the KNN algorithm is used to determine the proportion of point cloud blocks belonging to the second target point cloud block group among the preset number of point cloud blocks closest to the point cloud block to be detected in the first target point cloud block group. When the proportion exceeds a preset proportion threshold, the point cloud block to be detected is identified as an abnormal point cloud block.

5. The method for identifying abnormal interference sources as described in claim 1, characterized in that, The step of determining the uplink interfering cell includes: The interference level indicators of each cell are collected and imported into a preset database. The average interference level within a preset time period is extracted from the preset database, and the cell corresponding to the average interference level that meets the uplink interference cell screening criteria is identified as the uplink interference cell.

6. A system for identifying abnormal interference sources, characterized in that, The system for identifying abnormal interference sources includes: An information extraction module is used to identify uplink interfering cells and extract their three-dimensional location information and signal feature information. The three-dimensional location information includes longitude, latitude, and altitude, and the signal feature information includes uplink signal strength, uplink signal-to-noise ratio, and terminal transmit power. The segmentation module is used to generate three-dimensional point cloud data based on the three-dimensional position information and the signal feature information, and to segment the three-dimensional point cloud data to obtain multiple point cloud blocks. An abnormal interference source identification module is used to determine the location set of the abnormal interference sources based on each point cloud block, so as to identify the abnormal interference sources; The segmentation module is further configured to determine the longitude, latitude, and altitude of the uplink interfering cell as a three-dimensional location representation; and to map the uplink signal strength, the uplink signal-to-noise ratio, and the terminal transmit power to a preset space in a uniform distribution manner to obtain the color RGB representation of the point cloud data, wherein the preset space is a pixel range of 0~255; and to determine the three-dimensional point cloud data based on the three-dimensional location representation and the color RGB representation. The segmentation module is further configured to divide the three-dimensional point cloud data into multiple sub-regions using PointNet++, extract and iterate features from the sub-regions using PointNet to obtain high-dimensional cloud points, and perform inverse distance weight difference on the high-dimensional cloud points to obtain the corresponding low-dimensional cloud points. Feature fusion and feature extraction are performed on the low-dimensional cloud points, and the weights of each low-dimensional cloud point after feature fusion and feature extraction are globally normalized to obtain multiple point cloud blocks. The abnormal interference source identification module is also used to obtain the center position and uplink feature value of each point cloud block; The point cloud blocks are classified based on the feature values, and abnormal point cloud blocks are identified from the classified point cloud blocks according to the center position. The location set of abnormal interference sources is determined based on the three-dimensional center position of the abnormal point cloud block, so as to complete the identification of the abnormal interference sources.

7. A terminal device, characterized in that, The terminal device includes a memory, a processor, and an identification program for abnormal interference sources stored in the memory and executable on the processor. When the identification program for abnormal interference sources is executed by the processor, it implements the steps of the identification method for abnormal interference sources as described in any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a program for identifying abnormal interference sources, which, when executed by a processor, implements the steps of the method for identifying abnormal interference sources as described in any one of claims 1 to 5.

Images