Cell index quality difference detection method and device, computer device, and storage medium

Abstract

The present disclosure provides a cell index quality difference detection method and device, computer equipment and storage medium, the method comprising: acquiring time series data of a cell, wherein the time series data comprises a plurality of time points and cell index data corresponding to each time point; performing frequency domain transformation processing on the time series data to obtain a plurality of frequency components; performing mapping processing on each frequency component to obtain corresponding data points, wherein the data points have corresponding target data information; clustering a plurality of data points according to the target data information to obtain at least one cluster, wherein each cluster includes at least part of the data points; and performing quality difference detection on the at least one cluster to obtain a detection result. By implementing the method of the present disclosure, the dependence on parameter selection can be reduced through frequency domain transformation processing and clustering processing, the computing cost is reduced and the ability to process high-dimensional data is improved, thereby improving the accuracy and efficiency of anomaly detection.

Description

Technical Field

[0001] This disclosure relates to the fields of IT business support and artificial intelligence technology, specifically to a method, apparatus, computer equipment, and storage medium for detecting poor quality of community indicators. Background Technology

[0002] With the rapid development of 5G networks, the requirements of different services for network experience are constantly increasing. The traditional single-threshold rule method for identifying poor quality can no longer meet the diverse needs of different regions, scenarios and services.

[0003] In related technologies, when identifying poor cell quality, traditional anomaly detection algorithms, distance- and density-based sample anomaly detection methods, and machine learning algorithms are commonly used to identify abnormal samples in poor cell quality scenarios.

[0004] In this approach, quality defect detection algorithms often rely on statistical assumptions about the data, which limits their effectiveness when dealing with complex distributions or multimodal data. Furthermore, these methods often perform poorly in high-dimensional data scenarios, failing to capture complex relationships within the data. While distance- and density-based methods can adapt better to the distribution characteristics of the data, they are sensitive to parameter selection in practical applications and often face challenges in computational efficiency. Especially with large datasets and high dimensionality, the cost of calculating density and distance increases significantly. Summary of the Invention

[0005] This disclosure aims to at least partially address one of the technical problems in the related art.

[0006] Therefore, the purpose of this disclosure is to propose a method, device, computer equipment, and storage medium for detecting poor quality of community indicators, which can reduce the dependence on parameter selection through frequency domain transformation and clustering processing, reduce computational costs, and improve the ability to process high-dimensional data, thereby improving the accuracy and efficiency of anomaly detection.

[0007] To achieve the above objectives, the method for detecting poor cell index quality proposed in the first aspect of this disclosure includes:

[0008] Obtain time-series data of the cell, wherein the time-series data includes: multiple time points and cell indicator data corresponding to each time point;

[0009] The time series data is subjected to frequency domain transformation to obtain multiple frequency components;

[0010] Each frequency component is mapped to obtain a corresponding data point, wherein the data point has corresponding target data information;

[0011] Based on the target data information, multiple data points are clustered to obtain at least one cluster, wherein each cluster includes at least a portion of the data points; and

[0012] A quality difference detection is performed on at least one cluster to obtain the detection result.

[0013] To achieve the above objectives, the cell index poor quality detection device proposed in the second aspect of this disclosure includes:

[0014] The acquisition module is used to acquire time series data of the cell, wherein the time series data includes: multiple time points and cell indicator data corresponding to each time point;

[0015] The first processing module is used to perform frequency domain transformation on the time series data to obtain multiple frequency components.

[0016] The second processing module is used to perform mapping processing on each frequency component to obtain corresponding data points, wherein the data points have corresponding target data information;

[0017] The third processing module is configured to perform clustering processing on multiple data points based on the target data information to obtain at least one cluster, wherein each cluster includes at least a portion of the data points; and

[0018] The detection module is used to perform quality difference detection on the at least one cluster and obtain the detection result.

[0019] The computer device proposed in the third aspect of this disclosure includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the cell index poor quality detection method proposed in the first aspect of this disclosure.

[0020] The fourth aspect of this disclosure provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the cell index quality poor detection method as proposed in the first aspect of this disclosure.

[0021] The fifth aspect of this disclosure provides a computer program product that, when executed by a processor, performs a cell quality poor detection method as described in the first aspect of this disclosure.

[0022] The method, apparatus, computer equipment, and storage medium for detecting poor cell index quality provided in this disclosure acquire time-series data of a cell, wherein the time-series data includes multiple time points and corresponding cell index data for each time point; performs frequency domain transformation on the time-series data to obtain multiple frequency components; performs mapping processing on each frequency component to obtain corresponding data points, wherein each data point has corresponding target data information; performs clustering processing on the multiple data points based on the target data information to obtain at least one cluster, wherein each cluster includes at least some data points; and performs quality detection on the at least one cluster to obtain a detection result. Therefore, by using frequency domain transformation and clustering processing, the dependence on parameter selection can be reduced, computational costs can be lowered, and the ability to process high-dimensional data can be improved, thereby enhancing the accuracy and efficiency of anomaly detection.

[0023] Additional aspects and advantages of this disclosure will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this disclosure. Attached Figure Description

[0024] The above and / or additional aspects and advantages of this disclosure will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, in which:

[0025] Figure 1 This is a flowchart illustrating a method for detecting poor quality of community indicators according to an embodiment of this disclosure;

[0026] Figure 2 This is a flowchart illustrating a method for detecting poor quality of community indicators according to another embodiment of this disclosure;

[0027] Figure 3 This is a flowchart illustrating a method for detecting poor quality of community indicators according to another embodiment of this disclosure;

[0028] Figure 4 This is a schematic diagram of the structure of a poor quality detection device for community indicators proposed in an embodiment of this disclosure;

[0029] Figure 5 A block diagram of an exemplary computer device suitable for implementing embodiments of the present disclosure is shown. Detailed Implementation

[0030] Embodiments of this disclosure are described in detail below, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are used only to explain this disclosure, and should not be construed as limiting this disclosure. Rather, embodiments of this disclosure include all variations, modifications, and equivalents falling within the spirit and scope of the appended claims.

[0031] Figure 1 This is a flowchart illustrating a method for detecting poor quality of community indicators proposed in an embodiment of this disclosure.

[0032] It should be noted that the execution subject of the poor quality detection method of cell indicators in this embodiment is the poor quality detection device of cell indicators. The device can be implemented by software and / or hardware. The device can be configured in a computer device, which may include, but is not limited to, a terminal, a server, etc., such as a mobile phone, a handheld computer, etc.

[0033] like Figure 1 As shown, the method for detecting poor quality indicators in this community includes:

[0034] S101: Obtain the time series data of the cell, wherein the time series data includes: multiple time points and the cell indicator data corresponding to each time point.

[0035] In this context, a cell refers to the basic coverage area in a cellular communication system, also known as a base station cell.

[0036] Time series data refers to data composed of performance indicators of a cell at different points in time.

[0037] The community indicator data can include key performance indicators (KPIs) and key quality indicators (KQIs) of the community, and there are no restrictions on this.

[0038] In this embodiment of the disclosure, when the time series data of the cell is obtained, reliable data support can be provided for subsequent detection of poor cell index quality.

[0039] S102: Perform frequency domain transformation on the time series data to obtain multiple frequency components.

[0040] Among them, the frequency component can be used to indicate the correlation characteristics of time series data in the frequency domain.

[0041] In other words, in this embodiment of the present disclosure, after acquiring the time series data of the cell, the information in the time domain can be decomposed into the frequency domain through frequency domain transformation, and more hidden complex high-dimensional information in the original time series can be collected, thereby improving the ability of this embodiment of the present disclosure to identify abnormal points with poor quality.

[0042] S103: Map each frequency component to obtain the corresponding data point, where the data point has the corresponding target data information.

[0043] Here, a data point can refer to a point in three-dimensional space that corresponds to the frequency component after mapping processing.

[0044] Among them, target data information can refer to the relevant information corresponding to the data points.

[0045] In other words, after performing frequency domain transformation on time series data to obtain multiple frequency components, the embodiments of this disclosure can perform mapping processing on each frequency component to obtain corresponding data points. The data points have corresponding target data information, thereby providing reliable data support for obtaining clusters in the future.

[0046] S104: Based on the target data information, perform clustering processing on multiple data points to obtain at least one cluster, wherein each cluster includes at least some data points.

[0047] Among them, a cluster refers to a cluster obtained after multiple data points have undergone clustering processing.

[0048] In this embodiment of the present disclosure, when clustering multiple data points according to target data information to obtain at least one cluster, the target data information may be input into a pre-trained machine learning model to obtain the corresponding cluster, or the clustering may be based on a combination of numerical and graphical methods to obtain at least one cluster. No limitation is imposed on this method.

[0049] In this embodiment of the disclosure, after clustering multiple data points according to target data information to obtain at least one cluster, clustering of multiple data points can be achieved, thereby ensuring that data points within the same cluster have similar characteristics, so as to effectively improve the reliability of subsequent quality defect detection.

[0050] S105: Perform a quality defect detection on at least one cluster and obtain the detection result.

[0051] The detection results can be used to indicate the quality of data in clusters.

[0052] It is understood that different types of data points may have significant differences. Therefore, in this embodiment of the present disclosure, multiple data points can be clustered according to the target data information to obtain at least one cluster. Then, quality difference detection is performed on different clusters to obtain the detection results.

[0053] In this embodiment, time-series data of the cell is acquired, including multiple time points and corresponding cell indicator data for each time point. Frequency domain transformation is performed on the time-series data to obtain multiple frequency components. Each frequency component is mapped to obtain corresponding data points, where each data point has corresponding target data information. Based on the target data information, the multiple data points are clustered to obtain at least one cluster, where each cluster includes at least some data points. Finally, quality defect detection is performed on the at least one cluster to obtain detection results. Therefore, frequency domain transformation and clustering processes reduce dependence on parameter selection, lower computational costs, and improve the ability to process high-dimensional data, thereby enhancing the accuracy and efficiency of anomaly detection.

[0054] Figure 2 This is a flowchart illustrating a method for detecting poor quality of community indicators proposed in another embodiment of this disclosure.

[0055] like Figure 2 As shown, the method for detecting poor quality indicators in this community includes:

[0056] S201: Obtain the time series data of the cell, wherein the time series data includes: multiple time points and the cell indicator data corresponding to each time point.

[0057] For a detailed description of S201, please refer to the above embodiments, which will not be repeated here.

[0058] S202: Determine the sample data corresponding to the community indicator data.

[0059] Among them, sample data refers to the data obtained after sampling and collecting data on indicators from different communities.

[0060] In this embodiment of the disclosure, once the sample data corresponding to the cell index data is determined, reliable reference information can be provided for subsequent frequency domain transformation processing of time series data.

[0061] S203: Perform frequency domain transformation on time series data based on multiple time points and multiple sample data to obtain multiple frequency components.

[0062] In other words, in this embodiment of the present disclosure, after acquiring the time series data of the cell, sample data corresponding to the cell indicator data can be determined; based on multiple time points and multiple sample data, the time series data is subjected to frequency domain transformation processing to obtain multiple frequency components. Therefore, the indicative effect of the obtained frequency components can be effectively improved by combining the sample data corresponding to the cell indicator data.

[0063] S204: Determine the frequency components based on reference data information for each mapping dimension.

[0064] The mapping dimension refers to the dimension selected when mapping frequency components.

[0065] In this embodiment of the disclosure, the mapping dimension can be flexibly selected according to the needs of the scenario and the characteristics of the frequency components, and there are no restrictions on it.

[0066] Optionally, in some embodiments, the mapping dimension includes at least one of the following: periodicity dimension, volatility dimension, and complexity dimension. This ensures that the mapping process can meet personalized application scenarios, thereby guaranteeing the practicality of the obtained reference data information.

[0067] Among them, the periodicity dimension is related to the main frequency of the sequence, the volatility dimension can be measured by the frequency amplitude variation in the FFT result, and the complexity dimension can be measured by the spectral entropy of the sequence.

[0068] Optionally, in some embodiments, the mapping dimension includes a periodicity dimension; when determining the reference data information for a frequency component based on each mapping dimension, it can be done by: determining a target frequency that makes the frequency component satisfy a condition, wherein satisfying the condition for the frequency component includes: the absolute value of the frequency component based on the target frequency being greater than the absolute value of the frequency component based on other frequencies; and determining the ratio between a preset value and the target frequency as the reference data information under the periodicity dimension. Therefore, when the mapping dimension includes a periodicity dimension, the indicative effect of the obtained reference data information on the periodicity characteristics of the frequency component can be effectively improved.

[0069] The preset value refers to a value set in advance to determine the reference data information under the periodic dimension. For example, it can be 1, or it can be flexibly adjusted according to the application scenario without any restrictions.

[0070] Optionally, in some embodiments, the mapping dimension includes a volatility dimension; when determining the reference data information for each frequency component based on each mapping dimension, it may involve: obtaining the average amplitude of all frequency components; determining the degree of amplitude variation of the frequency components based on the total number of all frequency components and the average amplitude; and determining the degree of amplitude variation as the reference data information under the volatility dimension. Therefore, when the mapping dimension includes a volatility dimension, the accuracy of the obtained reference data information in indicating the degree of amplitude variation of the frequency components can be guaranteed.

[0071] The degree of amplitude variation can be used to indicate the volatility of the frequency component amplitude.

[0072] In this embodiment of the disclosure, when determining the amplitude variation of a frequency component based on the total number and average amplitude of all frequency components, the total number and average amplitude of all frequency components can be input into a pre-trained machine learning model to obtain the corresponding amplitude variation. Alternatively, the amplitude variation of the frequency component can be determined based on a preset formula. There are no restrictions on this.

[0073] Optionally, in some embodiments, the mapping dimension includes a complexity dimension. When determining the reference data information for each frequency component based on each mapping dimension, the process may involve: obtaining the amplitude distribution information of the frequency component; normalizing the amplitude distribution information to obtain a target amplitude distribution; determining the spectral entropy based on the target amplitude distribution, and using the spectral entropy as the reference data information for the complexity dimension. Therefore, when the mapping dimension includes a complexity dimension, the indicative effect of the obtained reference data information on the complexity of the frequency component can be guaranteed.

[0074] Among them, amplitude distribution information can be used to indicate the distribution of frequency components in different dimensions.

[0075] Among them, spectral entropy can be used to reflect the uniformity of the frequency component distribution.

[0076] In this embodiment of the disclosure, when determining the spectral entropy based on the target amplitude distribution, the target amplitude distribution can be substituted into a preset formula to calculate the corresponding spectral entropy.

[0077] S205: Determine the data points corresponding to the frequency components based on multiple reference data information.

[0078] In other words, in the embodiments of this disclosure, after determining the reference data information of the frequency component based on each mapping dimension, the data point corresponding to the frequency component can be determined based on multiple reference data information.

[0079] S206: Use multiple reference data information as target data information.

[0080] In other words, after obtaining multiple frequency components, the embodiments of this disclosure can determine reference data information for each frequency component based on each mapping dimension; determine the data points corresponding to the frequency components based on the multiple reference data information; and use the multiple reference data information as target data information. Therefore, the obtained target data information can indicate relevant data in different mapping dimensions, effectively improving the practicality of the target data information.

[0081] S207: Based on the target data information, perform clustering processing on multiple data points to obtain at least one cluster, wherein each cluster includes at least some data points.

[0082] S208: Perform a quality defect detection on at least one cluster and obtain the detection result.

[0083] For a detailed description of S207 and S208, please refer to the above embodiments, which will not be repeated here.

[0084] In this embodiment, sample data corresponding to cell indicator data is determined; frequency domain transformation is performed on time series data based on multiple time points and multiple sample data to obtain multiple frequency components. Therefore, the indicative effect of the obtained frequency components can be effectively improved by combining the sample data corresponding to the cell indicator data. Reference data information for each mapping dimension is determined for each frequency component; data points corresponding to the frequency components are determined based on multiple reference data; and these multiple reference data are used as target data information. Thus, the obtained target data information can indicate relevant data in different mapping dimensions, effectively improving the practicality of the target data information.

[0085] Figure 3 This is a flowchart illustrating a method for detecting poor quality of community indicators proposed in another embodiment of this disclosure.

[0086] like Figure 3 As shown, the method for detecting poor quality indicators in this community includes:

[0087] S301: Obtain the time series data of the cell, wherein the time series data includes: multiple time points and the cell indicator data corresponding to each time point.

[0088] S302: Perform frequency domain transformation on the time series data to obtain multiple frequency components.

[0089] S303: Map each frequency component to obtain the corresponding data point, where the data point has the corresponding target data information.

[0090] The descriptions of S301-S303 can be found in the above embodiments, and will not be repeated here.

[0091] S304: In each iteration, determine the similarity value between each data point and at least one cluster center.

[0092] The similarity value can be used to describe the similarity between each data point and at least one cluster center.

[0093] In this embodiment of the disclosure, when determining the similarity value between each data point and at least one cluster center, the distance between each data point and at least one cluster center may be determined, and then the corresponding similarity value may be determined based on the distance (the distance is inversely proportional to the similarity value).

[0094] S305: Determine the maximum similarity value among multiple similarity values, and cluster each data point into the cluster where the cluster center corresponding to the maximum similarity value is located.

[0095] In other words, the embodiments of this disclosure can determine the maximum similarity value among multiple similarity values ​​during the clustering process, and cluster each data point into the cluster where the cluster center corresponding to the maximum similarity value is located, so as to ensure the reliability of the final cluster.

[0096] S306: Update the cluster centers corresponding to the maximum similarity values ​​based on the target data information of the data points, and use the updated cluster centers for the next iteration.

[0097] In other words, in this embodiment of the present disclosure, after determining the maximum similarity value among multiple similarity values ​​and clustering each data point into the cluster where the cluster center corresponding to the maximum similarity value is located, the cluster center corresponding to the maximum similarity value can be updated according to the target data information of the data points. The updated cluster center is used for the next iteration process to achieve real-time updating of the cluster center and ensure the reliability of the cluster center in each iteration process.

[0098] S307: Determine at least one cluster that has completed the iteration until the change value of the cluster center is less than a preset threshold and / or the number of iterations reaches the number threshold.

[0099] The specific values ​​of the preset threshold and the number of times threshold can be flexibly adjusted according to the application scenario, and there are no restrictions on them.

[0100] In other words, in this embodiment of the present disclosure, after mapping each frequency component to obtain the corresponding data point, the similarity value between each data point and at least one cluster center can be determined in each iteration; the maximum similarity value among multiple similarity values ​​can be determined, and each data point can be clustered into the cluster where the cluster center corresponding to the maximum similarity value is located; the cluster center corresponding to the maximum similarity value can be updated according to the target data information of the data point, and the updated cluster center is used for the next iteration; until the change value of the cluster center is less than a preset threshold and / or the number of iterations reaches a threshold, at least one cluster is determined to be completed. Therefore, the clustering effect of data points can be effectively improved by combining the similarity value between each data point and at least one cluster center, and the practicality of the clusters obtained by the iteration can be guaranteed based on the preset threshold and / or the number of iterations threshold.

[0101] S308: Determine the spectral characteristics corresponding to each cluster.

[0102] Among them, spectral characteristics can be used to indicate the characteristics of multiple data points in a cluster in the spectral dimension.

[0103] In this embodiment of the disclosure, when the spectral characteristics corresponding to each cluster are determined, a reliable execution basis can be provided for the subsequent determination of the poor quality detection algorithm.

[0104] S309: Determine the quality defect detection algorithm corresponding to the spectral characteristics.

[0105] The poor quality detection algorithm refers to the algorithm used for poor quality detection as determined in the embodiments of this disclosure, such as isolated forest, STL decomposition, or DBSCAN, etc., and there are no restrictions on it.

[0106] It is understood that different spectral characteristics of clusters may affect the quality defect detection process. Therefore, in the embodiments of this disclosure, a quality defect detection algorithm corresponding to the spectral characteristics can be determined to ensure the quality defect detection effect of the obtained quality defect detection algorithm on the corresponding clusters.

[0107] S310: Use the quality difference detection algorithm corresponding to the spectral characteristics to perform quality difference detection on the corresponding clusters and obtain the detection results.

[0108] In other words, in this embodiment of the present disclosure, after obtaining at least one cluster, the spectral characteristics corresponding to each cluster can be determined; a quality defect detection algorithm corresponding to the spectral characteristics can be determined; and the quality defect detection algorithm corresponding to the spectral characteristics can be used to perform quality defect detection on the corresponding cluster to obtain the detection result. Therefore, a quality defect detection algorithm suitable for each cluster can be determined by combining the spectral characteristics corresponding to each cluster, thereby ensuring the accuracy and reliability of the obtained detection results.

[0109] In this embodiment, during each iteration, the similarity value between each data point and at least one cluster center is determined; the maximum similarity value among multiple similarity values ​​is determined, and each data point is clustered into the cluster where the cluster center corresponding to the maximum similarity value is located; the cluster center corresponding to the maximum similarity value is updated according to the target data information of the data point, and the updated cluster center is used for the next iteration; until the change value of the cluster center is less than a preset threshold and / or the number of iterations reaches a threshold, at least one cluster is determined to be completed. Thus, the clustering effect of data points can be effectively improved by combining the similarity value between each data point and at least one cluster center, and the practicality of the clusters obtained by iteration is guaranteed based on the preset threshold and / or the number of iterations threshold. The spectral characteristics corresponding to each cluster are determined; a quality defect detection algorithm corresponding to the spectral characteristics is determined; and the quality defect detection algorithm corresponding to the spectral characteristics is used to perform quality defect detection on the corresponding cluster to obtain the detection result. Thus, a quality defect detection algorithm suitable for each cluster can be determined by combining the spectral characteristics corresponding to each cluster, thereby ensuring the accuracy and reliability of the obtained detection results.

[0110] Based on the above embodiments, this disclosure proposes a novel method for identifying poor cell performance indicators (SMIs) based on spectral clustering. It utilizes the SpectraCluster Anomaly Detection (SCAD) algorithm to achieve efficient, accurate, and adaptable identification of dynamic SMIs. First, time-series data of key performance indicators (KPIs) and key quality indicators (KQIs) of the cells are collected, including but not limited to coverage, interference, capacity, and structure dimensions. Since the sequences of all cell KPIs are time-varying, Fourier transform can decompose the information from the time domain to the frequency domain, capturing more hidden, complex, and high-dimensional information from the original time series, thereby improving the ability of the SMI identification algorithm to identify anomalies. By applying Fast Fourier Transform (FFT) to the collected time-series data, the SCAD algorithm converts the original sequence data into scatter coordinates (x, y, z) in three-dimensional space, capturing and quantifying the spectral features of the sequence. Subsequently, the algorithm uses the K-means clustering method to group these scatter points, thereby identifying sequence categories with different quality characteristics. Based on the clustering results, this method further selects the most suitable quality defect identification algorithm for each sequence class from among anomaly detection algorithms such as Isolation Forest, STL (Seasonal and Trend Decomposition with Residual Detection), and DBSCAN (Density-based Spatial Clustering with Noise). This strategy aims to employ the most appropriate detection method for sequences with different quality characteristics to improve the accuracy and efficiency of quality defect identification.

[0111] The SCAD algorithm not only deeply considers the spectral characteristics of time series data and significantly reduces its dependence on parameter selection through cluster analysis, but also enhances its ability to process high-dimensional data. By combining advanced data processing techniques with machine learning algorithms, the SCAD algorithm provides an innovative solution for identifying poor-quality indicators in residential communities.

[0112] (1) Data collection:

[0113] This proposal comprehensively collects key performance indicators (KPIs) and key quality indicators (KQIs) of communication network cells, as well as multi-dimensional data such as coverage, interference, capacity, and structure (e.g., the average number of RRC connections in the KPIs). The collected sequences can be represented as s1, s2, ..., s n Where n is the time point. Each sequence s i Corresponding to a specific time point i, it reflects the cell status and performance indicators at that time point.

[0114] S = {s1, s2, ..., s} n}

[0115] By continuously collecting these sequences, the operational status of the cell can be monitored and analyzed in real time, thereby enabling accurate identification and assessment of quality differences.

[0116] By deeply extracting the characteristics of community scenarios, such as district / county, scene, manufacturer, and frequency band features, the system accurately identifies and distinguishes the characteristics of quality problems under different scenarios. The data collection process can adopt sliding detection and sliding collection methods to ensure the real-time updating and calculation of key indicator data, thereby realizing real-time identification of quality defects in community operation status and service quality.

[0117] (2) Fast Fourier Transform and Coordinate Mapping:

[0118] A. Fast Fourier Transform

[0119] In this proposal, Fast Fourier Transform (FFT) and coordinate mapping play a crucial role. They are able to extract key spectral features from time series data. Based on the key spectral features obtained from the FFT, the data features can be mapped into scatter points for further K-means clustering and quality defect detection.

[0120] The Fast Fourier Transform (FFT) is a crucial step in processing time series data, especially in transforming this data into the frequency domain. Through the FFT, the original time series data can be converted into a form representing its frequency components.

[0121] The core of the FFT transform is to convert time series data into a frequency domain representation. For the sequence data S = {s1, s2, ..., s...} collected in the previous chapter... n (e.g., the RRC average connection count in KPI metrics), its Fast Fourier Transform (FFT) representation is as follows:

[0122] Where k = 1, ..., n

[0123] Where S(k) represents the k-th frequency component in the frequency domain, which is the average number of RRC connections. -πijk / n It is the rotation factor, where i is the imaginary unit, representing the phase change in the frequency domain, and s j Let j represent the j samples with the average number of connections in the RRC, and n represent the number of samples.

[0124] B. Coordinate Mapping

[0125] Coordinate mapping is performed on the FFT-transformed data, that is, mapping each frequency component to scattered coordinates (x, y, z) in three-dimensional space. This allows for intuitive observation and analysis of the spectral characteristics of the data in geometric space. This step is crucial for identifying sequence categories with different quality characteristics.

[0126] X-coordinate (periodicity): Periodicity is related to the dominant frequency of the sequence. This can be determined by finding the value of k that maximizes |S(k)|. The mapped X-coordinate can be expressed as:

[0127]

[0128] Y-axis (volatility): Volatility is measured by the frequency amplitude variation in the FFT result.

[0129]

[0130] Where μ is the average amplitude of all frequency components, and N is the total number of frequency components.

[0131] Z-coordinate (complexity): Complexity is measured by the spectral entropy of the sequence, which reflects the uniformity of the frequency component distribution. The calculation of spectral entropy remains unchanged.

[0132]

[0133] Here, P(f) is the amplitude distribution of the normalized frequency components, ensuring that the sum is 1.

[0134] Through the two-step operation of Fast Fourier Transform and coordinate mapping, the sequence data to be detected in the cell can be transformed into scattered points (x,y,z) through Fast Fourier Transform and corresponding coordinate mapping, so as to facilitate the subsequent K-means clustering and model quality defect identification operations.

[0135] (3) K-means clustering

[0136] After Fast Fourier Transform (FFT) processing, the key performance and quality index sequences of the cell are transformed into a series of three-dimensional scattered coordinates (x, y, z), where each coordinate axis maps the spectral characteristics of the sequence in three dimensions: periodicity, volatility, and complexity. Based on this transformation, the K-means algorithm is further used to perform precise clustering analysis on these scattered points, achieving effective classification of sequences with different spectral characteristics. This clustering process not only highlights the natural distribution of data points in multi-dimensional space but also provides a clear basis for dividing sequences with different quality characteristics, which is a key step in improving the accuracy and efficiency of sequence quality defect identification.

[0137] In the SCAD algorithm, the K-means clustering algorithm is responsible for processing the 3D scatter coordinate dataset after the Fast Fourier Transform (FFT). Each point (x) i y i , z i ).

[0138] The goal of the K-means clustering algorithm is to... The points are divided into K clusters, such that points within a cluster are as similar as possible (i.e., close in distance), while points between clusters are as different as possible. This algorithm can be formalized as an optimization problem, with the objective function being to minimize the sum of the distances from all points to their respective cluster centers.

[0139] Mathematically, the K-means algorithm attempts to find a set of cluster centers. and each data point d i Assigning to the nearest cluster center can be represented as:

[0140]

[0141] Among them, ||d i -c j || 2 Represents data point d i To cluster center c j The square of the Euclidean distance. The optimization process is completed iteratively, with each iteration consisting of two steps:

[0142] A. Allocation: For each data point d i Find the nearest cluster center c j and d i Assigned to the corresponding clusters.

[0143]

[0144] Where γ(i) represents data point d i The index of the cluster to which it belongs.

[0145] B. Update: For each cluster, update the cluster center c. j This represents the mean position of all data points belonging to this cluster.

[0146]

[0147] Among them, S j It is assigned to cluster center c j The set of all data points, |S j | represents the number of points in the set.

[0148] After several iterations, the algorithm terminates and outputs the final cluster partitioning when convergence conditions are met, such as when the change in cluster centers is less than a predetermined threshold or the number of iterations reaches the upper limit.

[0149] (4) Model matching and quality defect identification

[0150] The K-means clustering algorithm is applied to these transformed scatter points to identify and classify datasets with similar characteristics. For each cluster's specific spectral characteristics, the most suitable quality defect detection algorithm, including Isolation Forest, STL decomposition, or DBSCAN, will be applied. This strategy will significantly improve the accuracy and efficiency of quality defect identification, thus providing precise support for the maintenance and optimization of cell metrics.

[0151] A. STL: STL is an algorithm suitable for processing time series data with obvious periodicity and seasonality trends. The first cluster of data shows some degree of periodicity, but with relatively low volatility and complexity. Using STL decomposition can help identify these periodic and trend anomalies in time series data and to more clearly see the basic patterns and anomalous fluctuations in the data. Outlier detection can be performed on the residuals after decomposition. The residuals represent the volatility after removing trends and seasonality; outliers usually appear as significant peaks or troughs in the residuals. Outliers can be identified by combining STL with Z-score anomaly detection methods.

[0152] In this proposal, for the collected cell time series data S = {s1, s2, ..., s...} n If we divide the data into three clusters, then S = {S1, S2, S3}. The feature sequence of S1 is represented in three-dimensional space as having relatively low periodicity, complexity, and volatility. Therefore, K-means can be used to obtain the result sequence of the first cluster as S1 (the feature sequence contained in the first cluster of the subdivision results; let S1 contain the feature sequence S...). i-1 ).

[0153] First, STL decomposition is performed, and each time series S... i It is broken down into three components:

[0154] s i-1 =T i +S i +R i

[0155] Among them, T i S represents the trend component. i Indicates seasonal components, R i This represents the residual component.

[0156] Next, calculate the residual sequence R = {R1, R2, ..., R...} n Z-score of}

[0157]

[0158] Here μ R σ is the average value of the residual sequence R. RZ is the standard deviation of the residual sequence R. i It is time point i.

[0159] Z-score in the residual sequence. By comparing the Z-score of each residual, outliers that deviate significantly from the average level can be identified.

[0160] B. iForest: Within the SCAD framework, the S2 feature sequence of the second cluster exhibits moderate periodicity and high volatility in three-dimensional space. This indicates that the sequence has certain regularity but is accompanied by significant fluctuations. Based on these characteristics, the Isolation Forest (iForest) algorithm is applicable. The Isolation Forest algorithm constructs multiple random trees (isolated trees) and uses the degree of isolation of data points as an indicator of anomalies. Outliers are usually easier to isolate due to their rarity in the data structure; therefore, the degree of isolation can be used as a basis for anomaly scoring.

[0161] The anomaly detection process for isolated forests includes the following steps:

[0162] Constructing isolated trees (iTrees): For a sequence s in set S2 i-2 The algorithm recursively segments the data by randomly selecting features and their corresponding segmentation values, thus constructing multiple isolated trees.

[0163] Calculating the anomaly score: The algorithm calculates the anomaly score based on the path length of each data point in the isolated tree. A shorter path length implies a higher probability of an anomaly. The formula for calculating the anomaly score is:

[0164]

[0165] Among them, E(h(s) i-2 )) represents point s i-2 The average path length on an isolated tree, c(n), is a normalization factor for the average path length of a tree consisting of n points in a dataset.

[0166] Outlier labeling: Based on anomaly scores, points s in the sequence i-2 If the score exceeds the preset threshold, it is considered an outlier.

[0167] C. DBSCAN: The DBSCAN algorithm is selected for anomaly detection in the third type of cluster S3. This type of cluster exhibits high periodicity, high volatility, and high complexity in three-dimensional space, indicating that its time series data may follow complex patterns, and the identification of anomalies depends on the spatial density characteristics of the data.

[0168] The DBSCAN algorithm identifies clustered regions and isolated points based on the number of neighboring points within a specific "neighborhood" for each data point. The algorithm's hyperparameters include:

[0169] ∈: Defines the radius size of the neighborhood, that is, neighboring points are searched within this radius. Generally, this hyperparameter can be defined by the distance map method.

[0170] MinPts: The minimum number of neighboring points (including the point itself) required within the ∈-neighborhood of a point to determine whether the point is a core point. The setting of MinPts can generally be determined according to the dimension of the dataset.

[0171] The DBSCAN algorithm process can be described as follows:

[0172] Calculation of the number of points in the neighborhood: For each point s i-3 [[ID=​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​By applying differentiated anomaly detection algorithms to different clusters within the SCAD (Spectrum Clustering Anomaly Detection) model, the system adapts to the diverse characteristics of communication network cell indicator sequence data. The K-means clustering algorithm is used to cluster the sequence data based on key features such as periodicity, volatility, and complexity. This allows for the selection of the most suitable anomaly detection algorithm for each cluster, including STL decomposition combined with Z-score, Isolation Forest (iForest), and DBSCAN algorithms, respectively targeting sequences with obvious periodicity and seasonality, sequences exhibiting moderate periodicity and high volatility, and sequences displaying high complexity. Through this approach, the SCAD model not only improves the accuracy of anomaly detection but also enhances its adaptability and flexibility in handling complex and variable communication network cell indicator data.

[0182] In summary, the technical points proposed in this disclosure include:

[0183] (1) A method for identifying poor cell performance indicators based on spectrum clustering. After collecting time series data of key performance indicators (KPI), KQI, coverage, interference, capacity and structure of cells, the method extracts frequency domain features through Fast Fourier Transform (FFT) technology. The method is characterized by further studying the relationship between data, that is, defining the complexity, periodicity and volatility of data, and classifying the spectrum characteristics in a more granular way based on the relationship between data in the three dimensions of complexity, periodicity and volatility. It creates a mechanism and theoretical basis for selecting different poor quality identification algorithms for each type of sequence, which breaks the shortcomings of traditional poor quality detection technology that does not delve into the relationship between data and only applies a single machine learning algorithm to anomaly detection in all scenarios.

[0184] (2) A spectrum clustering anomaly detection (SCAD) algorithm converts the FFT result into frequency domain data and performs K-means clustering based on three dimensions: complexity, periodicity, and volatility. Then, for these frequency domain data, the algorithm uses STL, iForest, and DBSCAN to identify quality defects for data that are characterized by relatively low periodicity, complexity, and volatility in three-dimensional space, data that exhibits medium periodicity and high volatility in three-dimensional space, and data that exhibits high periodicity, high volatility, and high complexity in three-dimensional space. This effectively improves the anomaly detection accuracy of the SCAD algorithm.

[0185] The advantages of the community indicator quality detection method proposed in this disclosure include, but are not limited to:

[0186] Compared to traditional quality defect detection techniques, the SCAD algorithm fully explores the relationships between data and defines multi-dimensional features such as periodicity, volatility, and complexity of time series data. By combining Fast Fourier Transform (FFT) with K-means clustering, it classifies data based on these multi-dimensional features, achieving accurate identification and location of anomalies. This comprehensive multi-dimensional feature analysis capability is often lacking in traditional methods. Secondly, by selecting appropriate anomaly detection algorithms for data clusters with different characteristics—combining STL decomposition with Z-score, Isolation Forest (iForest), and DBSCAN—the SCAD model ensures high adaptability and accuracy, surpassing identification methods based on a single threshold or statistical characteristic. This proposal provides an innovative solution for the diverse regional, scenario, and service needs in 5G networks through its efficient, accurate, and adaptable quality defect identification capabilities.

[0187] By introducing a real-time rolling quality defect detection mechanism, this invention continuously monitors and calculates anomalies in the feature sequence, enabling timely detection and repair of abnormal cell indicators. This mechanism not only improves network stability but also allows for rapid response in the event of anomalies, ensuring the quality and reliability of communication services.

[0188] SCAD proposes an efficient framework for identifying quality issues in high-dimensional data, combining spectral feature extraction, cluster analysis, and multi-algorithm anomaly detection techniques. This framework not only handles complex multidimensional time-series data but also performs data analysis and anomaly detection at different levels, providing a novel method for quality issue identification.

[0189] Figure 4 This is a schematic diagram of the structure of a poor quality detection device for community indicators proposed in one embodiment of this disclosure.

[0190] like Figure 4 As shown, the community indicator quality detection device 40 includes:

[0191] The acquisition module 401 is used to acquire time series data of the cell, wherein the time series data includes: multiple time points and cell indicator data corresponding to each time point;

[0192] The first processing module 402 is used to perform frequency domain transformation on the time series data to obtain multiple frequency components.

[0193] The second processing module 403 is used to perform mapping processing on each frequency component to obtain the corresponding data point, wherein the data point has corresponding target data information;

[0194] The third processing module 404 is used to perform clustering processing on multiple data points based on target data information to obtain at least one cluster, wherein each cluster includes: at least some data points; and

[0195] The detection module 405 is used to perform quality difference detection on at least one cluster and obtain the detection result.

[0196] It should be noted that the aforementioned explanation of the method for detecting poor quality of community indicators also applies to the community indicator quality detection device in this embodiment, and will not be repeated here.

[0197] In this embodiment, time-series data of the cell is acquired, including multiple time points and corresponding cell indicator data for each time point. Frequency domain transformation is performed on the time-series data to obtain multiple frequency components. Each frequency component is mapped to obtain corresponding data points, where each data point has corresponding target data information. Based on the target data information, the multiple data points are clustered to obtain at least one cluster, where each cluster includes at least some data points. Finally, quality defect detection is performed on the at least one cluster to obtain detection results. Therefore, frequency domain transformation and clustering processes reduce dependence on parameter selection, lower computational costs, and improve the ability to process high-dimensional data, thereby enhancing the accuracy and efficiency of anomaly detection.

[0198] Figure 5 A block diagram of an exemplary computer device suitable for implementing embodiments of the present disclosure is shown. Figure 5 The computer device 12 shown is merely an example and should not be construed as limiting the functionality and scope of use of the embodiments disclosed herein.

[0199] like Figure 5 As shown, the computer device 12 is represented in the form of a general-purpose computing device. The components of the computer device 12 may include, but are not limited to: one or more processors or processing units 16, system memory 28, and a bus 18 connecting different system components (including system memory 28 and processing unit 16).

[0200] Bus 18 represents one or more of several bus architectures, including a memory bus or memory controller, a peripheral bus, a graphics acceleration port, a processor, or a local bus using any of the various bus architectures. Examples of these architectures include, but are not limited to, the Industry Standard Architecture (ISA) bus, the Micro Channel Architecture (MAC) bus, the Enhanced ISA bus, the Video Electronics Standards Association (VESA) local bus, and the Peripheral Component Interconnect (PCI) bus.

[0201] Computer device 12 typically includes a variety of computer system readable media. These media can be any available media that can be accessed by computer device 12, including volatile and non-volatile media, removable and non-removable media.

[0202] Memory 28 may include computer system readable media in the form of volatile memory, such as random access memory (RAM) 30 and / or cache memory 32. Computer device 12 may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, storage system 34 may be used to read and write non-removable, non-volatile magnetic media (…). Figure 5 Not shown; usually referred to as a "hard drive".

[0203] although Figure 5 Not shown, a disk drive for reading and writing to a removable non-volatile disk (e.g., a "floppy disk") and an optical disc drive for reading and writing to a removable non-volatile optical disc (e.g., a Compact Disc Read-Only Memory (CD-ROM), a Digital Video Disc Read-Only Memory (DVD-ROM), or other optical media). In these cases, each drive may be connected to bus 18 via one or more data media interfaces. Memory 28 may include at least one program product having a set (e.g., at least one) of program modules configured to perform the functions of the embodiments of this disclosure.

[0204] A program / utility 40 having a set (at least one) of program modules 42 may be stored, for example, in memory 28. Such program modules 42 include, but are not limited to, an operating system, one or more application programs, other program modules, and program data. Each or some combination of these examples may include an implementation of a network environment. Program modules 42 typically perform the functions and / or methods described in the embodiments of this disclosure.

[0205] Computer device 12 can also communicate with one or more external devices 14 (e.g., keyboard, pointing device, display 24, etc.), and with one or more devices that enable human interaction with the computer device 12, and / or with any device that enables the computer device 12 to communicate with one or more other computing devices (e.g., network card, modem, etc.). This communication can be performed via input / output (I / O) interface 22. Furthermore, computer device 12 can also communicate with one or more networks (e.g., local area network (LAN), wide area network (WAN), and / or public networks, such as the Internet) via network adapter 20. As shown, network adapter 20 communicates with other modules of computer device 12 via bus 18. It should be understood that, although not shown in the figures, other hardware and / or software modules can be used in conjunction with computer device 12, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0206] The processing unit 16 executes various functional applications and data processing by running programs stored in the system memory 28, such as implementing the cell index quality poor detection method mentioned in the foregoing embodiments.

[0207] To implement the above embodiments, this disclosure also proposes a non-transitory computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the cell index quality poor detection method proposed in the foregoing embodiments of this disclosure.

[0208] To implement the above embodiments, this disclosure also proposes a computer program product that, when executed by an instruction processor, performs the cell index quality poor detection method as proposed in the foregoing embodiments of this disclosure.

[0209] Other embodiments of this disclosure will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This disclosure is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this disclosure are indicated by the following claims.

[0210] It should be understood that this disclosure is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this disclosure is limited only by the appended claims.

[0211] It should be noted that in the description of this disclosure, the terms "first," "second," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance. Furthermore, in the description of this disclosure, unless otherwise stated, "a plurality of" means two or more.

[0212] Any process or method description in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or more executable instructions for implementing a particular logical function or process, and the scope of preferred embodiments of this disclosure includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the function involved, as will be understood by those skilled in the art to which embodiments of this disclosure pertain.

[0213] It should be understood that various parts of this disclosure can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0214] Those skilled in the art will understand that all or part of the steps of the methods in the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, the program includes one or a combination of the steps of the method embodiments.

[0215] Furthermore, the functional units in the various embodiments of this disclosure can be integrated into a processing module, or each unit can exist physically separately, or two or more units can be integrated into a module. The integrated module can be implemented in hardware or as a software functional module. If the integrated module is implemented as a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium.

[0216] The storage media mentioned above can be read-only memory, disk, or optical disk, etc.

[0217] In the description of this specification, references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this disclosure. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0218] Although embodiments of the present disclosure have been shown and described above, it is to be understood that the above embodiments are exemplary and should not be construed as limiting the present disclosure. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present disclosure.

Claims

1. A method for detecting poor quality indicators in residential communities, characterized in that, The method includes: Obtain time-series data of the cell, wherein the time-series data includes: multiple time points and cell indicator data corresponding to each time point; The time series data is subjected to frequency domain transformation to obtain multiple frequency components; Each frequency component is mapped to obtain a corresponding data point, wherein the data point has corresponding target data information; Based on the target data information, multiple data points are clustered to obtain at least one cluster, wherein each cluster includes at least a portion of the data points; and Quality difference detection is performed on at least one cluster to obtain the detection results; The mapping process for each frequency component yields a corresponding data point, wherein each data point has corresponding target data information, including: The frequency components are determined based on reference data information for each mapping dimension; The data points corresponding to the frequency components are determined based on multiple sets of reference data information; The multiple reference data information are used as the target data information; The mapping dimension includes a periodic dimension; wherein, determining the frequency component based on reference data information under each mapping dimension includes: Determine a target frequency that makes the frequency component satisfy a condition, wherein the condition for the frequency component to satisfy a condition includes: the absolute value of the frequency component based on the target frequency is greater than the absolute value of the frequency component based on other frequencies; The ratio between the preset value and the target frequency is determined as the reference data information under the periodic dimension.

2. The method according to claim 1, characterized in that, The frequency domain transformation of the time series data yields multiple frequency components, including: Determine the sample data corresponding to the community indicator data; The time series data is subjected to frequency domain transformation based on the multiple time points and the multiple sample data to obtain the multiple frequency components.

3. The method according to claim 1, characterized in that, The mapping dimension also includes at least one of the following: volatility dimension and complexity dimension.

4. The method according to claim 3, characterized in that, The mapping dimension includes: a volatility dimension; wherein, determining the frequency component based on reference data information under each mapping dimension includes: Obtain the average amplitude of all frequency components; The degree of amplitude variation of the frequency components is determined based on the total number of all frequency components and the average amplitude. The degree of amplitude variation is determined as reference data information under the volatility dimension.

5. The method according to claim 3, characterized in that, The mapping dimension includes a complexity dimension; wherein, determining the frequency component based on reference data information under each mapping dimension includes: Obtain the amplitude distribution information of the frequency components; The amplitude distribution information is normalized to obtain the target amplitude distribution; Based on the target amplitude distribution, the spectral entropy is determined, and the spectral entropy is used as reference data information under the complexity dimension.

6. The method according to claim 1, characterized in that, The step of clustering multiple data points based on the target data information to obtain at least one cluster includes: In each iteration, the similarity value between each data point and at least one cluster center is determined; Determine the maximum similarity value among the multiple similarity values, and cluster each data point into the cluster where the cluster center corresponding to the maximum similarity value is located; The cluster centers corresponding to the maximum similarity value are updated based on the target data information of the data points, and the updated cluster centers are used in the next iteration process. The iteration continues until the change value of the cluster center is less than a preset threshold and / or the number of iterations reaches a threshold, at least one cluster that has completed the iteration is determined.

7. The method according to claim 1, characterized in that, The step of performing quality difference detection on the at least one cluster to obtain the detection result includes: Determine the spectral characteristics corresponding to each of the clusters; Determine the quality difference detection algorithm corresponding to the spectral characteristics; The quality difference detection algorithm corresponding to the spectral characteristics is used to perform quality difference detection on the corresponding cluster to obtain the detection result.

8. A device for detecting poor quality of community indicators, characterized in that, The device includes: The acquisition module is used to acquire time series data of the cell, wherein the time series data includes: multiple time points and cell indicator data corresponding to each time point; The first processing module is used to perform frequency domain transformation on the time series data to obtain multiple frequency components. The second processing module is used to perform mapping processing on each frequency component to obtain corresponding data points, wherein the data points have corresponding target data information; The third processing module is configured to perform clustering processing on multiple data points based on the target data information to obtain at least one cluster, wherein each cluster includes at least a portion of the data points; and The detection module is used to perform quality difference detection on the at least one cluster and obtain the detection result; The second processing module is specifically used for: The frequency components are determined based on reference data information for each mapping dimension; The data points corresponding to the frequency components are determined based on multiple sets of reference data information; The multiple reference data information are used as the target data information; The mapping dimension includes a periodic dimension; wherein, determining the frequency component based on reference data information under each mapping dimension includes: Determine a target frequency that makes the frequency component satisfy a condition, wherein the condition for the frequency component to satisfy a condition includes: the absolute value of the frequency component based on the target frequency is greater than the absolute value of the frequency component based on other frequencies; The ratio between the preset value and the target frequency is determined as the reference data information under the periodic dimension.

9. A computer device, characterized in that, include: At least one processor; as well as A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-7.

10. A non-transitory computer-readable storage medium storing computer instructions, characterized in that, in, The computer instructions are used to cause the computer to perform the method according to any one of claims 1-7.

11. A computer program product, characterized in that, Includes a computer program that, when executed by a processor, implements the steps of the method according to any one of claims 1-7.

Images