Uplink interference classification method, device, equipment and readable medium

Abstract

The embodiment of the present invention relates to the field of wireless network optimization technology, and discloses a classification method, device, equipment and readable medium for uplink interference, the method comprising: obtaining interference noise data on a target frequency band, the interference noise data comprising uplink PRB interference level values ​​of multiple cells to be classified within a preset time period; inputting the interference noise data into a pre-trained classification model, obtaining the interference waveform type of the cells to be classified output by the classification model, and determining the interference time domain characteristics of the cells to be classified according to the uplink PRB interference level values ​​of each cell to be classified within the preset time period; determining the target interference type of each cell to be classified according to the interference waveform type of each cell to be classified and the interference time domain characteristics. In the above manner, the embodiment of the present invention improves the accuracy of uplink interference positioning.

Description

Technical Field

[0001] The embodiments of the present invention relate to the technical field of wireless network optimization, and in particular to an interference classification method, device, equipment and readable medium. Background Art

[0002] Uplink interference refers to interference signals in the uplink frequency band of the mobile network. Mobile base stations are subject to interference from external RF interference sources or unreasonable internal frequency planning, such as co-channel and adjacent channel interference, which has a significant impact on the user's network quality. Therefore, it is necessary to classify uplink interference, locate the cause of interference, and then deal with it.

[0003] At present, the discovery and location of uplink interference mainly rely on user complaints, KPI indicator monitoring and on-site frequency scanning. Due to the fact that the means of interference classification in the existing technology are limited by the uneven ability of personnel, inaccurate analysis, low efficiency, high labor costs, and difficult experience replication, the classification time of interference sources and interference types is seriously delayed and the classification quality cannot be guaranteed, which reduces the timeliness and accuracy of discovering interference problems and affects network quality. Summary of the invention

[0004] In view of the above problems, the embodiments of the present invention provide a method, apparatus, device and readable medium for classifying uplink interference, which are used to solve the problems of inaccurate and low efficiency uplink interference positioning in the prior art.

[0005] According to one aspect of an embodiment of the present invention, a method for classifying uplink interference is provided, the method comprising: obtaining interference noise data on a target frequency band, the interference noise data comprising uplink PRB interference level values ​​of a plurality of cells to be classified within a preset time period;

[0006] Inputting the interference noise data into a pre-trained classification model, and obtaining the interference waveform type of the cell to be classified output by the classification model, wherein the classification model is trained according to a training sample set, the training sample set includes a plurality of labeled samples, and the labeled samples include an uplink PRB interference level value of the sample cell and a corresponding interference waveform type label;

[0007] Determining the interference time domain characteristics of the cells to be classified according to the uplink PRB interference level values ​​of the cells to be classified within the preset time period;

[0008] The target interference type of each of the cells to be classified is determined according to the interference waveform type of each of the cells to be classified and the interference time domain characteristics.

[0009] In an optional manner, before inputting the interference noise data into a pre-trained classification model, the method further includes:

[0010] Acquire the training sample set, wherein the training sample set further includes unmarked training samples, and the unmarked training samples include uplink PRB interference level values ​​of sample cells;

[0011] The preset classification model is trained with the training sample set as input and the waveform type labels of the unlabeled training samples as output to obtain the pre-trained classification model.

[0012] In an optional manner, the method of training a preset classification model using the training sample set as input and the waveform type label of the unlabeled training sample as output, and obtaining the pre-trained classification model for training, further comprises:

[0013] Acquire a test sample set, where the test sample set includes uplink PRB interference level values ​​of multiple cells;

[0014] Inputting the test sample set into the trained classification model, and obtaining the waveform type label output by the trained classification model as a result to be evaluated;

[0015] Determine the silhouette coefficient of the result to be evaluated according to the distance between each cell in each cluster in the result to be evaluated;

[0016] Compare the result to be evaluated with the interference waveform type label of each cell in the test sample set to determine a confusion matrix of the result to be evaluated;

[0017] The trained classification model is evaluated according to the silhouette coefficient and the confusion matrix, and if the evaluation passes, it is determined that the classification model training is complete.

[0018] In an optional manner, the target interference type includes a first interference type and a second interference type, and the time domain feature includes an average value of a busy-time interference level.

[0019] The determining the target interference type of each of the cells to be classified according to the interference waveform type of each of the cells to be classified and the time domain characteristics further includes:

[0020] The target interference type of the cells to be classified whose interference waveform type satisfies the preset first waveform feature and whose average self-busy interference level is greater than the preset interference value threshold is determined as the first interference type, and the target interference types of the remaining cells to be classified are determined as the second interference type.

[0021] In an optional manner, the time domain feature further includes the number of interference hours in a single day, and after determining the target interference type of the remaining cells to be classified as the second interference type, further includes:

[0022] Determine the correspondence between the interference waveform type and the target interference type;

[0023] Obtaining a cell to be classified as a first type of cell, in which the target interference type belongs to the second interference type and the target interference type corresponding to the interference waveform type is the first waveform type, and the number of interference hours per day is greater than a preset interference hour threshold, and determining the target interference type of the first type of cell according to the corresponding relationship, wherein the first waveform type includes spurious interference, blocking interference, and intermodulation harmonics;

[0024] Obtaining that the target interference type belongs to the second interference type in the cell to be located, the cell to be located whose target interference type is the second waveform type and the target interference type corresponding to the interference waveform type is the second type of cell, wherein the second waveform type includes atmospheric waveguide interference, sporadic interference, and school jammer, determining the target interference type of the second type of cell according to the corresponding relationship, and outputting the second type of cell as a target troubleshooting cell;

[0025] The target interference type of the cells to be located that do not belong to the first type of cells or the second type of cells among the cells to be located whose target interference type belongs to the second interference type is determined as the third interference type.

[0026] In an optional manner, the method further includes: taking the training sample set as input and the waveform type label of the unlabeled training sample as output to train a preset classification model, and obtaining the pre-trained classification model:

[0027] Determine the uplink PRB average value and standard deviation of each cell in the training sample set according to the uplink PRB interference level value of each cell in the training sample set;

[0028] Standardizing the uplink PRB interference level value of each cell in the training sample set according to the uplink PRB average value and standard deviation;

[0029] The standardized uplink PRB interference level value of each cell in the marked training sample set is obtained as an initial type center and input into the preset classification model.

[0030] In an optional manner, after determining the target interference type of each cell to be classified according to the interference waveform type of each cell to be classified and the interference time domain feature, the method further includes:

[0031] Obtaining location information of each of the cells to be classified;

[0032] Marking is performed on a preset map display device according to the target interference type and location information of each cell to be classified.

[0033] According to another aspect of an embodiment of the present invention, a device for classifying uplink interference is provided, including:

[0034] A data acquisition module, used to acquire interference noise data on a target frequency band, wherein the interference noise data includes uplink PRB interference level values ​​of a plurality of cells to be classified within a preset time period;

[0035] A waveform classification module, used to input the interference noise data into a pre-trained classification model, and obtain the interference waveform type of the cell to be classified output by the classification model, wherein the classification model is trained according to a training sample set, and the training sample set includes a plurality of marked samples, and the marked samples include an uplink PRB interference level value of the sample cell and a corresponding interference waveform type mark;

[0036] A time domain feature determination module, used to determine the interference time domain features of the cells to be classified according to the uplink PRB interference level values ​​of the cells to be classified within the preset time period;

[0037] The interference classification module is used to determine the target interference type of each cell to be classified according to the interference waveform type of each cell to be classified and the interference time domain characteristics.

[0038] According to another aspect of an embodiment of the present invention, there is provided an uplink interference classification device, comprising: a processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface communicate with each other via the communication bus; the memory is used to store at least one executable instruction, and the executable instruction enables the processor to execute the operation of the above-mentioned uplink interference classification method.

[0039] According to another aspect of an embodiment of the present invention, a computer-readable storage medium is provided, wherein the storage medium stores at least one executable instruction, wherein the executable instruction enables an uplink interference classification device / apparatus to perform the above-mentioned uplink interference classification operation.

[0040] According to another aspect of an embodiment of the present invention, a computer-readable storage medium is provided, wherein at least one executable instruction is stored in the computer-readable storage medium, and when the executable instruction is executed on an uplink interference classification device / apparatus, the uplink interference classification device performs the operation of the uplink interference classification method.

[0041] The embodiment of the present invention obtains interference noise data on a target frequency band, wherein the interference noise data includes a plurality of cells to be classified, and then inputs the interference noise data into a pre-trained classification model to obtain the interference waveform type of the cell to be classified output by the classification model, wherein the pre-trained classification model is trained based on labeled training samples and unlabeled training sample sets, wherein the labeled training sample set includes uplink PRB interference level values ​​and interference waveform type labels of a plurality of cells, and the unlabeled training sample set includes uplink PRB interference level values ​​of a plurality of cells.

[0042] After determining the interference waveform type of each cell to be classified through the above steps, the interference time domain characteristics of the cell to be classified are determined according to the uplink PRB interference level value of each cell to be classified within the preset time period, and finally, the target interference type of each cell to be classified is determined by combining the frequency domain characteristics of the interference waveform type of each cell to be classified and the above interference time domain characteristics.

[0043] The present invention comprehensively judges the interference type of the cell to be classified by combining the frequency domain characteristics and time domain characteristics of the interference of the cell to be classified. Different from the prior art which only considers the interference waveform information of the interfering cell, the present invention improves the accuracy of uplink interference positioning.

[0044] The above description is only an overview of the technical solution of the embodiment of the present invention. In order to more clearly understand the technical means of the embodiment of the present invention, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the embodiment of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] The accompanying drawings are only used to illustrate the embodiments and are not to be considered as limiting the present invention. In addition, the same reference symbols are used to represent the same components throughout the accompanying drawings. In the accompanying drawings:

[0046] Figure 1 A flow chart showing an embodiment of a method for locating uplink interference according to the present invention is shown;

[0047] Figure 2 A flow chart of standardizing sample data in one embodiment is shown;

[0048] Figure 3 A flow chart showing interference type determination in combination with time domain features in one embodiment is shown;

[0049] Figure 4 A schematic diagram showing the structure of an embodiment of a device for locating uplink interference according to the present invention is shown;

[0050] Figure 5A schematic structural diagram of an embodiment of an uplink interference classification device of the present invention is shown. DETAILED DESCRIPTION

[0051] Exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present invention are shown in the drawings, it should be understood that the present invention can be implemented in various forms and should not be limited to the embodiments set forth herein.

[0052] Figure 1 The flowchart of the embodiment of the method for locating uplink interference of the present invention is shown, and the method is executed by a computer processing device. The computer processing device includes: a mobile phone, a laptop computer, etc. Figure 1 As shown, the method comprises the following steps:

[0053] Step 110: Obtain interference noise data on the target frequency band, where the interference noise data includes uplink PRB interference level values ​​of a plurality of cells to be classified within a preset time period.

[0054] The target frequency band refers to a frequency band range with a certain threshold value, such as 200KHz, around the center frequency of the current working frequency band to be checked. The uplink PRB refers to the interference level value of the physical resource block (ie, PRB) of all uplink subframes in each cell within a preset time period. The preset time period can be, for example, from 8:00 to 22:00 in a day, and the present invention does not limit this.

[0055] Step 120: Input the interference noise data into a pre-trained classification model to obtain the interference waveform type of the cell to be classified output by the classification model, wherein the classification model is trained based on a training sample set, and the training sample set includes multiple labeled samples, and the labeled samples include the uplink PRB interference level value of the sample cell and the corresponding interference waveform type label.

[0056] The pre-trained classification model is obtained by training based on marked training samples and unmarked training sample sets, wherein the marked training sample set includes uplink PRB interference level values ​​and interference waveform type labels of multiple cells, and the unmarked training sample set includes uplink PRB interference level values ​​of multiple cells. The present invention is based on the interference waveform type of the marked cells, and divides each cell to be classified into the interference waveform type represented by the marked cell closest to it with the interference level value as the distance calculation parameter. Among them, the interference waveform of each marked cell is regarded as a separate interference waveform type, that is, the initial center of the cluster.

[0057] Before inputting the interference noise data into the pre-trained classification model, the classification model needs to be trained according to the above steps, further including steps 1201-1202:

[0058] Step 1201: Acquire the training sample set, wherein the training sample set further includes unmarked training samples, and the unmarked training samples include uplink PRB interference level values ​​of sample cells.

[0059] In order to improve the efficiency of classification model training, when determining the unmarked cells to be located, the level value features can be first screened to obtain cells that are likely to have interference, so that the subsequent training model outputs the interference waveform classification results for these cells.

[0060] For example, cells whose average utilization of 100 PRBs of all uplink subframes collected during a preset period of time such as 15 busy hours from 8:00 to 22:00 is not less than -110 decibels and not more than -80 decibels are used as cells in the unmarked training sample set.

[0061] Then, all 100PRB level data of the above-mentioned cells throughout the day are obtained as the unlabeled data here. Furthermore, since the level data of the unlabeled cells obtained may be missing, abnormal, etc., data cleaning is required. For the common case of missing data, the mean method is used for processing, that is, the interference data of the cell within one day is obtained, and the mean of its data distribution is taken to fill the missing data. Correspondingly, each cell in the marked training sample set is a known cell that causes interference, and its interference waveform has been marked by type. In an embodiment of the present invention, the interference waveform of each cell in the marked training sample set is used as a separate waveform type.

[0062] The interference waveform can be obtained according to the level data, and its abscissa can be all PRBs from 1 to 100 of all uplink subframes in a preset time period, and the ordinate is the average value of the level values ​​corresponding to the 100 PRBs from 1 to 100.

[0063] Step 1202: train a preset classification model with the training sample set as input and the waveform type labels of the unlabeled training samples as output to obtain the pre-trained classification model.

[0064] In the process of clustering, the waveform type of the marked cell is used as a separate cluster center. In the process of classification, each cell in the unmarked training sample set is assigned to the interference waveform type represented by the marked cell with the closest level value characteristics. In order to ensure the accuracy of clustering, before each marked cell is used as the corresponding initial type center, the level value data of each marked cell should be standardized as a basis for comparison.

[0065] The standardization process includes at least Figure 2 Steps 1211-1213 are shown, Figure 2A flow chart for normalizing sample data in one embodiment is shown.

[0066] Step 1211: Determine the uplink PRB average value and standard deviation of each cell in the training sample set according to the uplink PRB interference level value of each cell in the training sample set.

[0067] Step 1212: Standardize the uplink PRB interference level value of each cell in the training sample set according to the uplink PRB average value and standard deviation.

[0068] The standardization process first determines the uplink PRB average value Xi and standard deviation Si of each cell, and then calculates it according to the following formula:

[0069] Z ji =(X ji -X j )÷S j

[0070] Where: j = 1, 2, ..., k, represents the cell in the jth marked training sample; i = 1, 2, ..., 100, represents the i-th PRB in a subframe, Z ji is the standardized variable value, X ji is the actual variable value.

[0071] The actual variable value here is the average value of the level of each marked cell, which can be used as the uplink PRB interference level value of the initial type center after being standardized.

[0072] Step 1213: Obtain the standardized uplink PRB interference level value of each cell in the marked training sample set as the initial type center and input it into the preset classification model.

[0073] That is, after the standardization is completed, each cell in the labeled training sample set is taken as an initial type center, and each cell in the unlabeled sample set is allocated to the nearest initial category center according to the principle of minimum distance of level value features.

[0074] Furthermore, after training the preset classification model, it is necessary to test the training results of the classification model to determine whether the requirements have been met. The process of evaluating the training results of the classification model can at least further include the following steps 1221-1225.

[0075] Step 1221: Acquire a test sample set, where the test sample set includes uplink PRB interference level values ​​of multiple cells.

[0076] Step 1222: input the test sample set into the trained classification model, and obtain the waveform type label output by the trained classification model as the result to be evaluated.

[0077] Step 1223: Determine the silhouette coefficient of the result to be evaluated according to the distances between the cells in the clusters in the result to be evaluated.

[0078] The silhouette coefficient is used to evaluate the quality of waveform type labels. Its calculation method can be as follows: calculate the average distance between sample i and other samples in the same cluster as the intra-cluster dissimilarity of sample i. At the same time, calculate the average distance between sample i and other clusters c j The average distance b of all samples ij , called sample i and cluster c j The dissimilarity of b i =min{b i1 , b i2 , ..., b ik} is the inter-cluster dissimilarity of sample i.

[0079] Then the final silhouette coefficient of sample i is expressed as:

[0080]

[0081] Step 1224: Compare the result to be evaluated with the interference waveform type label of each cell in the test sample set to determine the confusion matrix of the result to be evaluated.

[0082] Step 1225: Evaluate the trained classification model according to the silhouette coefficient and the confusion matrix, and if the evaluation passes, determine that the classification model training is complete.

[0083] The model is evaluated based on the silhouette coefficient, and the silhouette coefficient is compared with the preset threshold (such as 1). The closer it is to 1, the better the training is. The evaluation based on the confusion matrix is ​​based on the preset parameters contained in the confusion matrix, such as accuracy, recall, specificity, and precision. If the parameters meet the corresponding thresholds, the training is considered complete.

[0084] Step 130: Determine the interference time domain characteristics of the cells to be classified according to the uplink PRB interference level value of each of the cells to be classified within the preset time period.

[0085] Therefore, in one embodiment, the time domain feature herein includes an average value of the busy-hour interference level.

[0086] Step 140: Determine the target interference type of each cell to be classified according to the interference waveform type and the interference time domain characteristics of each cell to be classified. The interference waveform type of each cell to be located determined in the above steps extracts the spectrum characteristics of each cell in a preset period or a certain time, but considering that the existence of interference generally lasts for a period of time and changes according to the signals of the interfering and interfered cells and the network conditions, it is also necessary to consider the interference time domain characteristics of the interfering cell, such as the interference value level during busy hours, or the length of time that a certain level of interference lasts, etc.

[0087] Step 140 further includes step 1401: determining the target interference type of the to-be-classified cell whose interference waveform type satisfies the preset first waveform feature and whose self-busy interference level average value is greater than the preset interference value threshold as the first interference type, and determining the target interference type of the remaining to-be-classified cells as the second interference type.

[0088] First, the preset first waveform feature is a frequency domain feature that satisfies the network interference, which is high in the middle and low on both sides or low in the middle and high on both sides. The preset interference value threshold here is 5 decibels. The first type of interference here refers to the network interference type, and correspondingly, the second waveform type interference type is the network interference type. The network interference type can also be divided into more types according to the cause of the interference, such as stray interference, blocking interference, intermodulation harmonic interference, etc., and the time domain feature here also includes the number of interference hours per day, so step 140 can further include Figure 3 Steps 1411-1413.

[0089] Step 1411: Determine the correspondence between the interference waveform type and the target interference type.

[0090] The target interference type may be determined by corresponding to each interference waveform type stored in a preset database.

[0091] Step 1412: Obtain the target interference type of the to-be-located cell belonging to the second interference type, the target interference type corresponding to the interference waveform type is the first waveform type, and the number of interference hours per day is greater than the preset interference hour threshold value as a first type cell, and determine the target interference type of the first type cell according to the corresponding relationship, wherein the first waveform type includes spurious interference, blocking interference, and intermodulation harmonics.

[0092] For certain specific interference types, such as the above-mentioned spurious interference, blocking interference, and intermodulation harmonics, the corresponding interference type cannot be determined solely based on its waveform type. It is also necessary to further combine the time domain characteristics for judgment, that is, whether the occurrence and duration of the interference waveform meet the characteristics of a certain type of interference, that is, whether the number of interference hours per day is greater than 10 hours.

[0093] Step 1413: Obtain the target interference type of the second interference type in the cell to be located, and the target interference type corresponding to the interference waveform type is the second waveform type, which is a second type cell, wherein the second waveform type includes atmospheric waveguide interference, sporadic interference, and school jammers. Determine the target interference type of the second type cell according to the corresponding relationship, and output the second type cell as a target troubleshooting cell.

[0094] Optionally, relevant information of the target troubleshooting cell, such as cell identification, location coordinates, interference collection data, etc., can be further obtained and output to relevant troubleshooting personnel, thereby helping the troubleshooting personnel to perform troubleshooting on site and eliminate interference as soon as possible.

[0095] Step 1414: Determine the target interference type of the cells to be located that do not belong to the first type of cells or the second type of cells among the cells to be located whose target interference type belongs to the second interference type as the third interference type.

[0096] After locating the target interference type of each cell to be located, each cell to be located can also be recorded as a black spot cell in a preset interference cell fingerprint library, thereby improving the efficiency of subsequent interference troubleshooting. And according to different interference types, they are displayed in a differentiated manner through a preset device. Specifically, at least the following steps 1501-1502 may be included.

[0097] Step 1501: Obtain location information of each cell to be classified.

[0098] Step 1502: Marking the target interference type and location information of each cell to be classified on a preset map display device.

[0099] The marking methods of cells with different interference types, such as marking colors and marking patterns, can be different to distinguish and prompt. The relevant information of the cells with preset interference types that need to be troubleshooted can be specially marked or their location information can be sent to relevant personnel for processing.

[0100] Figure 4 FIG. 2 is a schematic diagram showing the structure of an embodiment of a device for locating uplink interference according to the present invention. Figure 4 As shown, the device 300 includes: a data acquisition module 310, a waveform classification module 320, a time domain feature determination module 330, and an interference classification module 340.

[0101] In an optional manner, the waveform classification module 320 is further used to obtain the training sample set, wherein the training sample set also includes a de-marked training sample, and the de-marked training sample includes an uplink PRB interference level value of a sample cell;

[0102] The preset classification model is trained with the training sample set as input and the waveform type labels of the unlabeled training samples as output to obtain the pre-trained classification model.

[0103] In an optional manner, the waveform classification module 320 is also used to obtain a test sample set, which includes uplink PRB interference level values ​​of multiple cells; input the test sample set into the trained classification model, and obtain the waveform type label output by the trained classification model as the result to be evaluated; determine the contour coefficient of the result to be evaluated based on the distance between each cell in each cluster in the result to be evaluated; compare the result to be evaluated with the interference waveform type label of each cell in the test sample set to determine the confusion matrix of the result to be evaluated; evaluate the trained classification model based on the contour coefficient and the confusion matrix, and determine that the classification model training is completed if the evaluation passes.

[0104] In an optional manner, the interference classification module 340 is further configured to:

[0105] The target interference type of the cells to be classified whose interference waveform type satisfies the preset first waveform feature and whose average self-busy interference level is greater than the preset interference value threshold is determined as the first interference type, and the target interference types of the remaining cells to be classified are determined as the second interference type.

[0106] In an optional manner, the interference classification module 340 is further configured to:

[0107] Determine the correspondence between the interference waveform type and the target interference type;

[0108] Obtain the target interference type belonging to the second interference type in the to-be-located cell, and the target interference type corresponding to the interference waveform type is the first waveform type, and the number of interference hours per day is greater than the preset interference hour threshold, as the first category cell, and determine the target interference type of the first category cell according to the corresponding relationship, wherein the first waveform type includes stray interference, blocking interference, and intermodulation harmonics; obtain the target interference type belonging to the second interference type in the to-be-located cell, and the target interference type corresponding to the interference waveform type is the second waveform type as the second category cell, wherein the second waveform type includes atmospheric waveguide interference, sporadic interference, and school jammers, determine the target interference type of the second category cell according to the corresponding relationship, and output the second category cell as the target troubleshooting cell; determine the target interference type of the to-be-located cell that does not belong to the first category cell or the second category cell among the to-be-located cells whose target interference type belongs to the second interference type as the third interference type.

[0109] In an optional manner, the waveform classification module 320 is further configured to:

[0110] The uplink PRB average value and standard deviation of each cell in the training sample set are determined according to the uplink PRB interference level value of each cell in the training sample set; the uplink PRB interference level value of each cell in the training sample set is standardized according to the uplink PRB average value and standard deviation; and the uplink PRB interference level value of each cell in the marked training sample set after standardization is obtained as an initial type center and input into the preset classification model.

[0111] In an optional manner, the interference classification module 340 is further configured to:

[0112] Acquire the location information of each cell to be classified; and mark each cell to be classified on a preset map display device according to the target interference type and location information.

[0113] The specific working process of the uplink interference classification device in the embodiment of the present invention is the same as the specific process steps of the uplink interference method described above, and will not be repeated here.

[0114] The uplink interference classification device of the embodiment of the present invention comprehensively determines the interference type of the cell to be classified by combining the frequency domain characteristics and time domain characteristics of the interference of the cell to be classified, thereby improving the accuracy of uplink interference positioning. Figure 5 The schematic diagram of the structure of an embodiment of the uplink interference classification device of the present invention is shown. The specific embodiment of the present invention does not limit the specific implementation of the uplink interference classification device.

[0115] like Figure 5 As shown, the uplink interference classification device may include: a processor (processor) 402 , a communication interface (Communications Interface) 404 , a memory (memory) 406 , and a communication bus 408 .

[0116] The processor 402, the communication interface 404, and the memory 406 communicate with each other via the communication bus 408. The communication interface 404 is used to communicate with other devices such as a client or other server network elements. The processor 402 is used to execute the program 410, which can specifically execute the relevant steps in the above-mentioned embodiment of the classification method for uplink interference.

[0117] Specifically, the program 410 may include a program code, which includes computer executable instructions. The processor 402 may be a central processing unit CPU, or an application specific integrated circuit ASIC (Application Specific Integrated Circuit), or one or more integrated circuits configured to implement the embodiment of the present invention. The one or more processors included in the uplink interference classification device may be processors of the same type, such as one or more CPUs; or may be processors of different types, such as one or more CPUs and one or more ASICs.

[0118] The memory 406 is used to store the program 410. The memory 406 may include a high-speed RAM memory, and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.

[0119] Program 410 can be specifically called by processor 402 to enable the uplink interference classification device to perform the following operations: obtain interference noise data on the target frequency band, the interference noise data including uplink PRB interference level values ​​of multiple cells to be classified within a preset time period; input the interference noise data into a pre-trained classification model, and obtain the interference waveform type of the cell to be classified output by the classification model, wherein the classification model is trained according to a training sample set, and the training sample set includes multiple labeled samples, and the labeled samples include the uplink PRB interference level value of the sample cell and the corresponding interference waveform type label; determine the interference time domain characteristics of the cell to be classified according to the uplink PRB interference level value of each cell to be classified within the preset time period; determine the target interference type of each cell to be classified according to the interference waveform type and the interference time domain characteristics of each cell to be classified.

[0120] In an optional manner, the program 410 is called by the processor 402 to enable the uplink interference classification device to perform the following operations:

[0121] Acquire the training sample set, wherein the training sample set also includes unmarked training samples, and the unmarked training samples include uplink PRB interference level values ​​of sample cells; use the training sample set as input and the waveform type labels of the unmarked training samples as output to train a preset classification model to obtain the pre-trained classification model.

[0122] In an optional manner, the program 410 is called by the processor 402 to enable the uplink interference classification device to perform the following operations:

[0123] Acquire a test sample set, the test sample set including uplink PRB interference level values ​​of multiple cells; input the test sample set into the trained classification model, and obtain a waveform type label output by the trained classification model as a result to be evaluated; determine a contour coefficient of the result to be evaluated according to the distance between each cell in each cluster in the result to be evaluated;

[0124] Compare the result to be evaluated with the interference waveform type label of each cell in the test sample set to determine a confusion matrix of the result to be evaluated;

[0125] The trained classification model is evaluated according to the silhouette coefficient and the confusion matrix, and if the evaluation passes, it is determined that the classification model training is complete.

[0126] In an optional manner, the program 410 is called by the processor 402 to enable the uplink interference classification device to perform the following operations:

[0127] The target interference type of the cells to be classified whose interference waveform type satisfies the preset first waveform feature and whose average self-busy interference level is greater than the preset interference value threshold is determined as the first interference type, and the target interference types of the remaining cells to be classified are determined as the second interference type.

[0128] In an optional manner, the program 410 is called by the processor 402 to enable the uplink interference classification device to perform the following operations:

[0129] Determine the correspondence between the interference waveform type and the target interference type;

[0130] Obtaining a cell to be classified as a first type of cell, in which the target interference type belongs to the second interference type and the target interference type corresponding to the interference waveform type is the first waveform type, and the number of interference hours per day is greater than a preset interference hour threshold, and determining the target interference type of the first type of cell according to the corresponding relationship, wherein the first waveform type includes spurious interference, blocking interference, and intermodulation harmonics;

[0131] Obtaining that the target interference type belongs to the second interference type in the cell to be located, the cell to be located whose target interference type is the second waveform type and the target interference type corresponding to the interference waveform type is the second type of cell, wherein the second waveform type includes atmospheric waveguide interference, sporadic interference, and school jammer, determining the target interference type of the second type of cell according to the corresponding relationship, and outputting the second type of cell as a target troubleshooting cell;

[0132] The target interference type of the cells to be located that do not belong to the first type of cells or the second type of cells among the cells to be located whose target interference type belongs to the second interference type is determined as the third interference type.

[0133] In an optional manner, the program 410 is called by the processor 402 to enable the uplink interference classification device to perform the following operations:

[0134] Obtaining location information of each of the cells to be classified;

[0135] Marking is performed on a preset map display device according to the target interference type and location information of each cell to be classified.

[0136] The uplink interference classification device of the embodiment of the present invention comprehensively determines the interference type of the cell to be classified by combining the frequency domain characteristics and time domain characteristics of the interference of the cell to be classified, thereby improving the accuracy of uplink interference positioning.

[0137] An embodiment of the present invention provides a computer-readable storage medium, which stores at least one executable instruction. When the executable instruction runs on an uplink interference classification device / apparatus, the uplink interference classification device / apparatus executes the uplink interference classification method in any of the above-mentioned method embodiments.

[0138] The specific working process of the computer-readable storage medium in the embodiment of the present invention is the same as the specific process steps of the above-mentioned uplink interference classification method, which will not be repeated here.

[0139] The computer-readable storage medium of the embodiment of the present invention comprehensively determines the interference type of the cell to be classified by combining the frequency domain characteristics and time domain characteristics of the interference of the cell to be classified, thereby improving the accuracy of uplink interference positioning.

[0140] An embodiment of the present invention provides a computer program, which can be called by a processor to enable an uplink interference classification device to perform the uplink interference classification method in any of the above method embodiments.

[0141] The embodiment of the present invention provides a computer program product, which includes a computer program stored on a computer-readable storage medium, and the computer program includes program instructions. When the program instructions are run on a computer, the computer is caused to perform the classification method of uplink interference in any of the above-mentioned method embodiments. The algorithm or display provided herein is not inherently related to any specific computer, virtual system or other device. Various general systems can also be used together with the teachings based on this. According to the above description, the structure required to construct such a system is obvious. In addition, the embodiment of the present invention is not directed to any specific programming language. It should be understood that the content of the present invention described herein can be implemented using various programming languages, and the above description of specific languages ​​is to disclose the best implementation mode of the present invention.

[0142] In the description provided herein, a large number of specific details are described. However, it is understood that embodiments of the present invention can be practiced without these specific details. In some instances, well-known methods, structures and techniques are not shown in detail so as not to obscure the understanding of this description.

[0143] Similarly, it should be understood that in order to streamline the present invention and aid in understanding one or more of the various inventive aspects, in the above description of exemplary embodiments of the present invention, various features of the embodiments of the present invention are sometimes grouped together into a single embodiment, figure, or description thereof. However, this method of disclosure should not be interpreted as reflecting an intention that the claimed invention requires more features than those expressly recited in each claim.

[0144] Those skilled in the art will appreciate that the modules in the devices in the embodiments may be adaptively changed and arranged in one or more devices different from the embodiments. The modules or units or components in the embodiments may be combined into one module or unit or component, and may be divided into a plurality of submodules or subunits or subcomponents. Except that at least some of such features and / or processes or units are mutually exclusive, all features disclosed in this specification (including the accompanying claims, abstracts and drawings) and all processes or units of any method or device disclosed in this manner may be combined in any combination. Unless otherwise expressly stated, each feature disclosed in this specification (including the accompanying claims, abstracts and drawings) may be replaced by an alternative feature providing the same, equivalent or similar purpose.

[0145] It should be noted that the above embodiments illustrate the present invention rather than limit it, and that those skilled in the art may devise alternative embodiments without departing from the scope of the appended claims. In the claims, any reference symbol between brackets shall not be construed as a limitation on the claims. The word "comprising" does not exclude the presence of elements or steps not listed in the claims. The word "one" or "an" preceding an element does not exclude the presence of a plurality of such elements. The present invention may be implemented by means of hardware comprising a number of different elements and by means of a suitably programmed computer. In a unit claim enumerating a number of devices, several of these devices may be embodied by the same hardware item. The use of the words first, second, and third, etc. does not indicate any order. These words may be interpreted as names. The steps in the above embodiments, unless otherwise specified, should not be understood as limitations on the order of execution.

Claims

1. A method for classifying uplink interference, characterized in that: The method comprises: Obtain interference noise data on a target frequency band, wherein the interference noise data includes uplink PRB interference level values ​​of a plurality of cells to be classified within a preset time period; The interference noise data is input into a pre-trained classification model, and the interference waveform type of the cell to be classified output by the classification model is obtained, wherein the classification model is trained according to a training sample set, and the training sample set includes a plurality of labeled samples, and the labeled samples include an uplink PRB interference level value of the sample cell and a corresponding interference waveform type label; before the interference noise data is input into the pre-trained classification model, it further includes: obtaining the training sample set, wherein the training sample set also includes unlabeled training samples, and the unlabeled training samples include an uplink PRB interference level value of the sample cell; taking the training sample set as input and the waveform type label of the unlabeled training sample as output, the preset classification model is trained to obtain the pre-trained classification model; wherein the uplink PRB average value and the label of the uplink PRB of each cell in the training sample set are determined according to the uplink PRB interference level value of each cell in the training sample set. Standard deviation; standardize the uplink PRB interference level value of each cell in the training sample set according to the uplink PRB mean value and standard deviation; obtain the uplink PRB interference level value of each cell in the standardized labeled training sample set as the initial type center and input it into the preset classification model; obtain a test sample set, the test sample set includes the uplink PRB interference level values ​​of multiple cells; input the test sample set into the trained classification model, and obtain the waveform type label output by the trained classification model as the result to be evaluated; determine the contour coefficient of the result to be evaluated according to the distance between each cell in each cluster in the result to be evaluated; compare the result to be evaluated with the interference waveform type label of each cell in the test sample set to determine the confusion matrix of the result to be evaluated; evaluate the trained classification model according to the contour coefficient and the confusion matrix, and determine that the classification model training is completed if the evaluation passes; Determining the interference time domain characteristics of the cells to be classified according to the uplink PRB interference level values ​​of the cells to be classified within the preset time period; The target interference type of each of the cells to be classified is determined according to the interference waveform type of each of the cells to be classified and the interference time domain characteristics.

2. The method according to claim 1, characterized in that The target interference type includes a first interference type and a second interference type, the time domain feature includes an average value of a busy-time interference level, The determining the target interference type of each of the cells to be classified according to the interference waveform type of each of the cells to be classified and the time domain characteristics further includes: The target interference type of the cells to be classified whose interference waveform type satisfies the preset first waveform feature and whose average self-busy interference level is greater than the preset interference value threshold is determined as the first interference type, and the target interference type of the remaining cells to be classified is determined as the second interference type.

3. The method according to claim 2, characterized in that The time domain feature also includes the number of interference hours in a single day. After determining the target interference type of the remaining cells to be classified as the second interference type, the method further includes: Determine the correspondence between the interference waveform type and the target interference type; Obtaining a cell to be classified as a first type of cell, in which the target interference type belongs to the second interference type and the target interference type corresponding to the interference waveform type is the first waveform type, and the number of interference hours per day is greater than a preset interference hour threshold, and determining the target interference type of the first type of cell according to the corresponding relationship, wherein the first waveform type includes spurious interference, blocking interference, and intermodulation harmonics; Obtaining that the target interference type belongs to the second interference type in the cell to be located, the cell to be located whose target interference type is the second waveform type and the target interference type corresponding to the interference waveform type is the second type of cell, wherein the second waveform type includes atmospheric waveguide interference, sporadic interference, and school jammer, determining the target interference type of the second type of cell according to the corresponding relationship, and outputting the second type of cell as a target troubleshooting cell; The target interference type of the cells to be located that do not belong to the first type of cells or the second type of cells among the cells to be located whose target interference type belongs to the second interference type is determined as the third interference type.

4. The method according to claim 1, characterized in that: After determining the target interference type of each of the cells to be classified according to the interference waveform type of each of the cells to be classified and the interference time domain characteristics, the method further includes: Obtaining location information of each of the cells to be classified; Marking is performed on a preset map display device according to the target interference type and location information of each cell to be classified.

5. An uplink interference classification device, characterized in that: The device comprises: A data acquisition module, used to acquire interference noise data on a target frequency band, wherein the interference noise data includes uplink PRB interference level values ​​of a plurality of cells to be classified within a preset time period; A waveform classification module is used to input the interference noise data into a pre-trained classification model, and obtain the interference waveform type of the cell to be classified output by the classification model, wherein the classification model is trained according to a training sample set, and the training sample set includes multiple labeled samples, and the labeled samples include the uplink PRB interference level value of the sample cell and the corresponding interference waveform type label; before the interference noise data is input into the pre-trained classification model, it further includes: obtaining the training sample set, wherein the training sample set also includes unlabeled training samples, and the unlabeled training samples include the uplink PRB interference level value of the sample cell; using the training sample set as input and the waveform type label of the unlabeled training sample as output to train the preset classification model to obtain the pre-trained classification model; wherein the uplink PRB of each cell in the training sample set is determined according to the uplink PRB interference level value of each cell in the training sample set Mean value and standard deviation; standardize the uplink PRB interference level value of each cell in the training sample set according to the uplink PRB mean value and standard deviation; obtain the uplink PRB interference level value of each cell in the standardized labeled training sample set as the initial type center and input it into the preset classification model; obtain a test sample set, the test sample set includes the uplink PRB interference level values ​​of multiple cells; input the test sample set into the trained classification model, and obtain the waveform type label output by the trained classification model as the result to be evaluated; determine the contour coefficient of the result to be evaluated according to the distance between each cell in each cluster in the result to be evaluated; compare the result to be evaluated with the interference waveform type label of each cell in the test sample set to determine the confusion matrix of the result to be evaluated; evaluate the trained classification model according to the contour coefficient and the confusion matrix, and determine that the classification model training is completed if the evaluation passes; A time domain feature determination module, used to determine the interference time domain features of the cells to be classified according to the uplink PRB interference level values ​​of the cells to be classified within the preset time period; The interference classification module is used to determine the target interference type of each cell to be classified according to the interference waveform type of each cell to be classified and the interference time domain characteristics.

6. An uplink interference classification device, characterized in that: include: A processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface communicate with each other via the communication bus; The memory is used to store at least one executable instruction, and the executable instruction enables the processor to perform the operation of the uplink interference classification method as described in any one of claims 1-5.

7. A computer-readable storage medium, characterized in that: The storage medium stores at least one executable instruction. When the executable instruction is executed on the uplink interference classification device / apparatus, the uplink interference classification device / apparatus performs the operation of the uplink interference classification method as described in any one of claims 1-5.

Images