Network anomaly prediction method, device and storage medium

By acquiring data from both the network and user sides and using a network anomaly prediction model for parameter extraction and prediction, the problem of low efficiency in network quality monitoring in existing technologies has been solved, achieving efficient and accurate network anomaly detection.

CN116056117BActive Publication Date: 2024-11-19CHINA UNITED NETWORK COMM GRP CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202310102412.5
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-01-20
Publication Date
2024-11-19
Estimated Expiration
2043-01-20

AI Technical Summary

Technical Problem

Existing technologies are inefficient in network quality monitoring, struggle to handle the ever-increasing volume of network communication data, and lack accuracy and timeliness in anomaly detection.

Method used

By acquiring historical time-period data from both the network and user sides, network quality parameters, such as instant messaging latency and video download speed, are extracted and input into a network anomaly prediction model for prediction, including outlier detection, parameter fitting, and time series prediction, thereby improving detection efficiency and accuracy.

Benefits of technology

It enables timely detection of network anomalies, improves detection efficiency and accuracy, reduces manual intervention, and enhances operational efficiency.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN116056117B_ABST
    Figure CN116056117B_ABST
Patent Text Reader

Abstract

The present application provides a network anomaly prediction method, device and storage medium, which relates to the field of communication technology and can solve the problem of low detection efficiency of the prior art. The method includes: obtaining network quality data; the network quality data includes network side data and user side data in a historical time period; the user side data includes user service configuration data and terminal equipment capability data; extracting parameters from the network quality data to obtain network quality parameters in a historical time period; the network quality parameters include at least one of instant communication delay, video download rate and downlink round-trip delay; inputting the network quality parameters into a network anomaly prediction model to obtain a prediction result at the target time; the prediction result includes abnormal cells and abnormal problem information. The present application can improve the efficiency of network anomaly prediction.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of communication technology, and in particular to a network anomaly prediction method, device and storage medium. Background Art

[0002] With the development of the communications industry and the rapid iteration of networks, the increasingly complex networks, the geometric growth in the number of mobile communication base stations, and the real-time communication data of massive users all need to be stored in a timely manner. Wireless network quality monitoring is to organize the timely communication data, transmit it in real time, and mine abnormal data to ensure the normal operation of communication base stations.

[0003] In the prior art, relevant technicians usually monitor network quality data and detect abnormal data therein to determine network abnormality problems. However, this solution has low detection efficiency and is difficult to cope with the network communication data with doubled data volume. Summary of the invention

[0004] The present application provides a network anomaly prediction method, device and storage medium, which can improve the efficiency of network anomaly prediction.

[0005] In order to achieve the above purpose, this application adopts the following technical solutions:

[0006] In a first aspect, the present application provides a network anomaly prediction method, the method comprising: obtaining network quality data; the network quality data comprises network side data and user side data within a historical time period; the user side data comprises user service configuration data and terminal device capability data; performing parameter extraction on the network quality data to obtain network quality parameters within the historical time period; the network quality parameters comprise at least one of instant communication delay, video download rate and downlink round-trip delay; inputting the network quality parameters into a network anomaly prediction model to obtain a prediction result at a target time; the prediction result comprises abnormal cells and abnormal problem information.

[0007] Based on the above technical solution, the network anomaly prediction device obtains network quality data including network side data and user side data in a historical time period, thereby extracting parameters from the network quality data to obtain network quality parameters in the historical time period. The network quality parameters may include at least one of instant messaging delay, video download rate, and downlink round-trip delay. Afterwards, the network anomaly prediction device may input the network quality parameters into the network anomaly prediction model to obtain the prediction result at the target time. The prediction result includes abnormal cells and abnormal problem information. Compared with the solution in the prior art in which technicians manually perform wireless network quality monitoring and detection, the present application improves the detection efficiency and detection accuracy. At the same time, the network anomaly prediction device in the present application can also make predictions based on the network quality parameters in the historical time period to obtain the prediction results at the target time, so that network abnormal problems can be detected more promptly, improving the timeliness of anomaly detection.

[0008] In combination with the above-mentioned first aspect, in a possible implementation method, the network anomaly prediction model includes an outlier detection sub-model, a parameter fitting sub-model and a time series prediction sub-model; the outlier detection sub-model is used to delete abnormal parameters in the network quality parameters; the parameter fitting sub-model is used to perform data fitting on the network quality parameters; and the time series prediction sub-model is used to predict the network quality parameters at the target time.

[0009] In combination with the above-mentioned first aspect, in a possible implementation manner, the method includes: inputting the network quality parameter into the outlier detection sub-model to obtain the first network quality parameter after removing the abnormal parameter; inputting the first network quality parameter into the parameter fitting sub-model to obtain the second network quality parameter after the parameter fitting; inputting the second network quality parameter into the time series prediction sub-model to obtain the third network quality parameter at the target time; and determining the prediction result according to the third network quality parameter.

[0010] In combination with the first aspect above, in a possible implementation, the method further includes: performing data preprocessing operations on the network quality data; the data preprocessing operations include at least one of the following: duplicate value data processing, missing value data processing, outlier data processing, and data standardization processing.

[0011] In a second aspect, the present application provides a network anomaly prediction device, which includes: a communication unit, used to obtain network quality data; the network quality data includes network side data and user side data within a historical time period; the user side data includes user service configuration data and terminal equipment capability data; a processing unit, used to extract parameters from the network quality data to obtain network quality parameters within the historical time period; the network quality parameters include at least one of instant messaging delay, video download rate, and downlink round-trip delay; the processing unit is also used to input the network quality parameters into a network anomaly prediction model to obtain a prediction result at a target time; the prediction result includes abnormal cells and abnormal problem information.

[0012] In combination with the above-mentioned second aspect, in a possible implementation method, the network anomaly prediction model includes an outlier detection sub-model, a parameter fitting sub-model and a time series prediction sub-model; the outlier detection sub-model is used to delete abnormal parameters in the network quality parameters; the parameter fitting sub-model is used to perform data fitting on the network quality parameters; and the time series prediction sub-model is used to predict the network quality parameters at the target time.

[0013] In combination with the above-mentioned second aspect, in a possible implementation method, the processing unit is used to: input the network quality parameter into the outlier detection sub-model to obtain the first network quality parameter after removing the abnormal parameter; input the first network quality parameter into the parameter fitting sub-model to obtain the second network quality parameter after the parameter fitting; input the second network quality parameter into the time series prediction sub-model to obtain the third network quality parameter at the target time; determine the prediction result according to the third network quality parameter.

[0014] In combination with the above-mentioned second aspect, in a possible implementation method, the processing unit is also used to: perform data preprocessing operations on the network quality data; the data preprocessing operations include at least one of the following: duplicate value data processing, missing value data processing, outlier data processing and data standardization processing.

[0015] In a third aspect, the present application provides a network anomaly prediction device, comprising: a processor and a communication interface; the communication interface and the processor are coupled, and the processor is used to run a computer program or instruction to implement the network anomaly prediction method described in the first aspect and any possible implementation method of the first aspect.

[0016] In a fourth aspect, the present application provides a computer-readable storage medium, which stores instructions. When the instructions are executed on a terminal, the terminal executes the network anomaly prediction method described in the first aspect and any possible implementation of the first aspect.

[0017] In a fifth aspect, the present application provides a computer program product comprising instructions. When the computer program product runs on a network anomaly prediction device, the network anomaly prediction device executes the network anomaly prediction method as described in the first aspect and any possible implementation of the first aspect.

[0018] In a sixth aspect, the present application provides a chip, the chip comprising a processor and a communication interface, the communication interface and the processor are coupled, and the processor is used to run a computer program or instructions to implement the network anomaly prediction method as described in the first aspect and any possible implementation method of the first aspect.

[0019] Specifically, the chip provided in the present application also includes a memory for storing computer programs or instructions.

[0020] It should be noted that the above computer instructions may be stored in whole or in part on a computer-readable storage medium, wherein the computer-readable storage medium may be packaged together with the processor of the device, or may be packaged separately from the processor of the device, which is not limited in this application.

[0021] The descriptions of the second to sixth aspects of the present application can refer to the detailed description of the first aspect; and the beneficial effects of the descriptions of the second to sixth aspects can refer to the beneficial effect analysis of the first aspect, which will not be repeated here.

[0022] In this application, the name of the above-mentioned network anomaly prediction device does not limit the device or functional module itself. In actual implementation, these devices or functional modules may appear with other names. As long as the functions of each device or functional module are similar to those of this application, they belong to the scope of the claims of this application and their equivalent technologies.

[0023] These and other aspects of the present application will become more apparent from the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 A schematic diagram of the architecture of a network anomaly prediction system provided in an embodiment of the present application;

[0025] Figure 2 A functional architecture diagram of a network anomaly prediction system provided in an embodiment of the present application;

[0026] Figure 3 A schematic diagram of the architecture of a problem delimiting function provided in an embodiment of the present application;

[0027] Figure 4 A flowchart of a network anomaly prediction method provided in an embodiment of the present application;

[0028] Figure 5A flowchart of another network anomaly prediction method provided in an embodiment of the present application;

[0029] Figure 6 A fitting data diagram of a network quality parameter provided in an embodiment of the present application;

[0030] Figure 7 A flowchart of another network anomaly prediction method provided in an embodiment of the present application;

[0031] Figure 8 A schematic diagram of the structure of a network anomaly prediction device provided in an embodiment of the present application;

[0032] Fig. 9 A schematic diagram of the structure of another network anomaly prediction device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0033] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0034] The term "and / or" in this article is merely a description of the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone.

[0035] The terms "first" and "second" and the like in the specification and drawings of this application are used to distinguish different objects, or to distinguish different processing of the same object, rather than to describe a specific order of objects.

[0036] In addition, the terms "including" and "having" and any variations thereof mentioned in the description of the present application are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device comprising a series of steps or units is not limited to the listed steps or units, but may optionally include other steps or units that are not listed, or may optionally include other steps or units that are inherent to these processes, methods, products or devices.

[0037] It should be noted that, in the embodiments of the present application, words such as "exemplary" or "for example" are used to indicate examples, illustrations or descriptions. Any embodiment or design described as "exemplary" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary" or "for example" is intended to present related concepts in a specific way.

[0038] In the description of the present application, unless otherwise specified, “plurality” means two or more.

[0039] With the development of the communications industry and the rapid iteration of networks, the number of active Internet users with 4G terminals has reached 986 million, and the number of 5G terminals has exceeded 200 million. Faced with increasingly complex networks, the geometric growth of the number of mobile communication base stations, as well as massive user real-time communication data, all need to be stored in a timely manner. Wireless network quality monitoring is to organize and transmit real-time communication data, and mine abnormal data to ensure the normal operation of communication base stations.

[0040] At present, the conventional method of wireless network quality monitoring is to use a large number of network planning and optimization engineers to gather key quality indicators (KQI) and key performance indicators (KPI) according to the time dimension, analyze the operation rules of each cell based on historical data, and then determine network abnormalities by detecting abnormal data. However, the above solution has problems such as low operating efficiency, low accuracy in determining abnormal discrete data, and large real-time transmission delay.

[0041] Faced with the rapid development of wireless network scale and number of users, the above methods are increasingly difficult to reflect the actual network conditions of users in a timely manner, and the processing efficiency is getting lower and lower after the data volume doubles. At the same time, the cell positioning of cross-professional core networks and transmission network problems is more complicated, and the perception of network problems caused by terminals or user behavior is even more difficult to locate.

[0042] In view of this, the present application provides a network anomaly prediction method, in which a network anomaly prediction device obtains network quality data including network side data and user side data within a historical time period, thereby extracting parameters from the network quality data and obtaining network quality parameters within the historical time period. The network quality parameters may include at least one of instant messaging delay, video download rate, and downlink round-trip delay. Afterwards, the network anomaly prediction device may input the network quality parameters into a network anomaly prediction model to obtain a prediction result at a target time. The prediction result includes abnormal cells and abnormal problem information. Compared with the solution of manually performing wireless network quality monitoring and detection by technicians in the prior art, the present application improves detection efficiency and detection accuracy. At the same time, the network anomaly prediction device in the present application can also make predictions based on the network quality parameters within the historical time period to obtain prediction results at the target time, so that network abnormal problems can be detected more timely, and the timeliness of abnormality detection is improved.

[0043] Figure 1 This is an architecture diagram of a network anomaly prediction system 10 provided in an embodiment of the present application. Figure 1As shown, the network anomaly prediction system 10 includes: a network anomaly prediction device 101 and at least one terminal device 102 .

[0044] The network anomaly prediction device 101 and the terminal device 102 are connected via a communication link, which may be a wired communication link or a wireless communication link, which is not limited in this application. The number of the terminal devices 102 in this application may be one or more.

[0045] Exemplarily, the embodiments of the present application can be applied to various communication systems, such as code division multiple access (CDMA), time division multiple access (TDMA), frequency division multiple access (FDMA), orthogonal frequency division multiple access (OFDMA), single carrier frequency division multiple access (SC-FDMA) and other systems. The term "system" can be interchangeable with "network". The CDMA system can implement wireless technologies such as universal terrestrial radio access (UTRA) and CDMA2000. UTRA can include wideband CDMA (WCDMA) technology and other CDMA variant technologies. CDMA2000 can cover interim standard (IS) 2000 (IS-2000), IS-95 and IS-856 standards. The TDMA system can implement wireless technologies such as global system for mobile communication (GSM). The OFDMA system can implement wireless technologies such as evolved universal radio terrestrial access (evolved UTRA, E-UTRA), ultra mobile broadband (ultra mobile broadband, UMB), IEEE 802.11 (Wi-Fi), IEEE 802.16 (WiMAX), IEEE802.20, Flash OFDMA, etc. UTRA and E-UTRA are UMTS and UMTS evolution versions. 3GPP's long term evolution (long term evolution, LTE) and various versions based on LTE evolution are new versions of UMTS using E-UTRA. 5G communication system and new radio (new radio, NR) are communication systems under study. In addition, the communication system can also be applied to future-oriented communication technologies, and the technical solutions provided in the embodiments of the present application are applicable.

[0046] It should be noted that the network anomaly prediction method provided in the embodiment of the present application can be applied to the above-mentioned network anomaly prediction device 101, and the above-mentioned network anomaly prediction device 101 can be an independent communication device, such as an access network device, a core network device and other communication devices. The above-mentioned network anomaly prediction device 101 can also be a functional module coupled in a communication device, such as a communication network element. The above-mentioned network anomaly prediction device 101 can also be a computer program (application, APP) for executing the network anomaly prediction method. The above-mentioned network anomaly prediction device 101 can also be a server connected to the communication device.

[0047] Exemplarily, when the network anomaly prediction device 101 is a server connected to a communication device, the network anomaly prediction device 101 includes:

[0048] Processor, the processor can be a general-purpose central processing unit (CPU), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits used to control the execution of the program of the present application.

[0049] Transceiver, a transceiver can be a device using any transceiver type for communicating with other devices or communication networks, such as Ethernet, radio access network (RAN), wireless local area networks (WLAN), etc.

[0050] Memory, the memory may be a read-only memory (ROM) or other types of static storage devices that can store static information and instructions, a random access memory (RAM) or other types of dynamic storage devices that can store information and instructions, or an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical disc, laser disc, optical disc, digital versatile disc, Blu-ray disc, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto. The memory may exist independently and be connected to the processor through a communication line. The memory may also be integrated with the processor.

[0051] When the network anomaly prediction device 101 is an access network device, the network anomaly prediction device 101 may be a device located at the access network side of the above-mentioned communication system and having a wireless transceiver function or a chip or chip system that can be set in the device. The access network equipment includes, but is not limited to: access points (APs) in WiFi systems, such as home gateways, routers, servers, switches, bridges, etc., evolved NodeBs (eNBs), radio network controllers (RNCs), NodeBs (NBs), base station controllers (BSCs), base transceiver stations (BTSs), home base stations (e.g., home evolvedNodeBs, or home NodeBs, HNBs), base band units (BBUs), wireless relay nodes, wireless backhaul nodes, transmission points (TRPs or TPs), etc., and may also be 5G base stations, such as gNBs in new radio (NR) systems, or transmission points (TRPs or TPs), one or a group of (including multiple antenna panels) antenna panels of base stations in 5G systems, or network nodes constituting gNBs or transmission points, such as base band units (BBUs), or distributed units (DUs), road side units with base station functions (road side Access network equipment also includes base stations in different networking modes, such as master evolved NodeB (MeNB), secondary eNB (SeNB, or secondary gNB, SgNB). Access network equipment also includes different types, such as ground base stations, aerial base stations, and satellite base stations.

[0052] The network anomaly prediction device 101 can also be a core network device, which is located on the core network side of the above-mentioned communication system. The core network device can be a physical device or a virtual device. The core network devices of the 4G network include but are not limited to: mobility management entity (MME), serving gateway (SGW), public data network gateway (PGW). The core network devices of the 5G network include but are not limited to: access and mobility management function network element (AMF), session management function network element (SMF), user plane function network element (UPF).

[0053] The terminal device 102 may also be a device with a wireless communication function or a wired communication function, which may be deployed on land, including indoors or outdoors, handheld or vehicle-mounted. It may also be deployed on the water surface (such as a ship, etc.). It may also be deployed in the air (such as on an airplane, a balloon, and a satellite, etc.). The terminal device 102 is also called user equipment (UE), mobile station (MS), mobile terminal (MT), and terminal, etc., and is a device that provides voice and / or data connectivity to a user. For example, the terminal device 102 includes a handheld device with a wireless connection function, a vehicle-mounted device, etc. At present, the terminal device 102 can be: a mobile phone, a tablet computer, a laptop computer, a PDA, a mobile internet device (MID), a wearable device (such as a smart watch, a smart bracelet, a pedometer, etc.), a vehicle-mounted device (such as a car, a bicycle, an electric car, an airplane, a ship, a train, a high-speed train, etc.), a virtual reality (VR) device, an augmented reality (AR) device, a wireless terminal device in industrial control, a smart home device (such as a refrigerator, a television, an air conditioner, an electric meter, etc.), an intelligent robot, a workshop device, a wireless terminal device in self-driving, a wireless terminal device in remote medical surgery, a wireless terminal device in a smart grid, a wireless terminal device in transportation safety, a wireless terminal device in a smart city, or a wireless terminal device in a smart home, a flying device (such as an intelligent robot, a hot air balloon, a drone, an airplane), etc. In one possible application scenario of the present application, the terminal device is a terminal device that often works on the ground, such as a vehicle-mounted device. In the present application, for the sake of convenience, the chip deployed in the above-mentioned device, such as the system-on-a-chip (SOC), baseband chip, etc., or other chips with communication functions can also be referred to as terminal devices.

[0054] The terminal device 102 may be a vehicle with corresponding communication functions, or a vehicle-mounted communication device, or other embedded communication devices, or may be a user's handheld communication device, including a mobile phone, a tablet computer, etc.

[0055] As an example, in an embodiment of the present application, the terminal device 102 may also be a wearable device. Wearable devices may also be referred to as wearable smart devices, which are a general term for wearable devices that are intelligently designed and developed using wearable technology for daily wear, such as glasses, gloves, watches, clothing, and shoes. A wearable device is a portable device that is worn directly on the body or integrated into the user's clothes or accessories. Wearable devices are not just hardware devices, but also achieve powerful functions through software support, data interaction, and cloud interaction. Broadly speaking, wearable smart devices include those that are fully functional, large in size, and can achieve complete or partial functions without relying on smartphones, such as smart watches or smart glasses, as well as those that only focus on a certain type of application function and need to be used in conjunction with other devices such as smartphones, such as various types of smart bracelets and smart jewelry for vital sign monitoring.

[0056] The network anomaly prediction device 101 is used to obtain network quality data.

[0057] The network quality data includes network-side data and user-side data in the historical time period. The user-side data includes user service configuration data and terminal equipment capability data.

[0058] For example, the network side data may be network dimension indicator evaluation data, which is divided into network element performance data and end-to-end signaling notification data, wherein each data includes a large number of data fields and data types.

[0059] The user service configuration data may be configuration data of a user package service, and is used to represent subscription service information of the terminal device 102 .

[0060] Exemplarily, the network anomaly prediction system 10 provided in the embodiment of the present application can build a unified data pool based on the Postgres distributed database, thereby building a wireless network basic database of network side data by parsing network management configuration XML data, basic engineering parameter data and network element KPI data, and at the same time, by connecting to the manufacturer's customer experience management enabling (service & experience quality, SEQ) end-to-end signaling system underlying data, a user multi-dimensional perception database with minute-level granularity can be built.

[0061] The network anomaly prediction system 10 provided in the embodiment of the present application can also aggregate user B side data, and associate it with the terminal capability library after desensitization operation, so as to establish a user characteristic database for storing user side data.

[0062] Exemplarily, the present application can also reduce the difficulty of implementing the network anomaly prediction system 10 by screening the data for value and storing the core indicator data in a unified manner.

[0063] In a possible implementation, the network anomaly prediction device 101 may obtain user-side data from the terminal device 102 .

[0064] Exemplarily, the terminal device 102 sends the user-side data to the network anomaly prediction apparatus 101 , and correspondingly, the network anomaly prediction apparatus 101 receives the user-side data from the terminal device 102 .

[0065] The network anomaly prediction device 101 is also used to extract parameters from the network quality data to obtain network quality parameters within a historical time period.

[0066] The network quality parameter includes at least one of an instant messaging delay, a video download rate, and a downlink roundtrip time (RTT).

[0067] Exemplarily, the network anomaly prediction device 101 may obtain network quality parameters within a historical time period by extracting fields used for user perceived risk prediction.

[0068] For example, the network quality parameters may be derived from signaling / user plane data of interfaces such as the evolved packet core (EPC) network S1-MME interface, S1U interface, S6A interface, S11 interface, and GN interface.

[0069] The network anomaly prediction device 101 is also used to input the network quality parameter into the network anomaly prediction model to obtain the prediction result at the target time.

[0070] The prediction results include abnormal cells and abnormal problem information.

[0071] In a possible implementation, the network anomaly prediction model includes an outlier detection sub-model, a parameter fitting sub-model, and a time series prediction sub-model.

[0072] Among them, the outlier detection sub-model is used to delete abnormal parameters in the network quality parameters. The parameter fitting sub-model is used to perform data fitting on the network quality parameters. The time series prediction sub-model is used to predict the network quality parameters at the target time.

[0073] Exemplarily, the network anomaly prediction system 10 provided in the embodiment of the present application can obtain the fluctuation pattern of the user perception flow by aggregating and analyzing the minute-level perception indicators of sampled users, thereby performing intelligent prediction of the user perception flow through an AI algorithm to achieve high-granularity control of the user perception trend.

[0074] Among them, the network anomaly prediction system 10 can remove historically perceived outliers in network quality parameters through an outlier detection sub-model. For example, the outlier detection sub-model can be implemented through algorithms such as density-based spatial clustering of applications with noise (DBSCAN), box plot algorithm, and Gaussian mixture model (GMM).

[0075] Afterwards, the network anomaly prediction system 10 can fit the indicator flow trend of the network quality parameter through the parameter fitting sub-model, for example, through the least squares method and the multivariate linear regression algorithm.

[0076] For the fitted network quality parameters, the network anomaly prediction system 10 can predict the perception indicators at the target time based on the historical perception indicator trends of the network quality parameters in multiple time periods through the time series prediction submodel. The time series prediction submodel can be implemented by the prophet time series prediction algorithm.

[0077] In this way, the network anomaly prediction system 10 can perform anomaly detection on the predicted perception indicators through dynamic thresholds and absolute thresholds, thereby obtaining the prediction results at the target time.

[0078] For example, Figure 2 As shown, the network anomaly prediction system 10 in the present application has the following functions. First, the network anomaly prediction device 101 in the network anomaly prediction system 10 can obtain communication data within a historical time period from at least one terminal device 102, and uniformly read various types of communication data through the server.

[0079] Afterwards, the network anomaly prediction device 101 can perform unified data analysis and clean the analyzed data, and store them in the corresponding data warehouse.

[0080] Finally, the network anomaly prediction device 101 aggregates the data in the data warehouse according to time, so as to predict the operation law of the next cycle and detect abnormal data through the operation law of the network quality parameters in the historical data.

[0081] Among them, Figure 3 As shown, the network anomaly prediction system 10 in the present application can implement problem demarcation by peeling the onion method.

[0082] First, the network anomaly prediction device 101 in the network anomaly prediction system 10 can analyze outlier users based on user package data, terminal type data and master cell data. The relevant data can come from the wireless network side, transmission network side, core network side, user side, terminal side and other aspects.

[0083] Based on the KPI / KQI combined with the AI ​​positioning engine, the network anomaly prediction device 101 can perform data processing and feature processing on the data in the KPI / KQI data warehouse, and select the corresponding model for data grouping, perform model training and model prediction on the grouped data, and perform cross-validation. After the verification is completed, the trained model is output.

[0084] Through the trained model, the network anomaly prediction device 101 can realize anomaly detection of difficult problems, coverage conditions, interference conditions, communication quality, capacity, network switching and other issues.

[0085] Through systematic logic design, this application can integrate complex data of multiple dimensions, so as to unify the association of multi-level user data, facilitate collection and analysis, and help the iterative evolution of the development platform. At the same time, each level has a certain degree of independence and can be split and utilized like a module, making problem location more flexible and clear.

[0086] It should be pointed out that the various embodiments of the present application can learn from or refer to each other. For example, the same or similar steps, method embodiments, system embodiments and device embodiments can all refer to each other without limitation.

[0087] Figure 4 A flowchart of a network anomaly prediction method provided in an embodiment of the present application. Figure 4 As shown, the method comprises the following steps:

[0088] Step 401: The network anomaly prediction device obtains network quality data.

[0089] The network quality data includes network-side data and user-side data in the historical time period. The user-side data includes user service configuration data and terminal equipment capability data.

[0090] It should be noted that the historical time period can be set according to actual conditions, such as one week. In this case, the network quality data can be data for each day of the week.

[0091] For example, the network side data may be network dimension indicator evaluation data, which is divided into network element performance data and end-to-end signaling notification data, wherein each data includes a large number of data fields and data types.

[0092] The user service configuration data may be configuration data of a user package service, and is used to represent subscription service information of a terminal device.

[0093] In a possible implementation, the network anomaly prediction apparatus may obtain user-side data from a terminal device.

[0094] Exemplarily, the terminal device sends user-side data to the network anomaly prediction apparatus, and correspondingly, the network anomaly prediction apparatus receives the user-side data from the terminal device.

[0095] Step 402: The network anomaly prediction device extracts parameters from the network quality data to obtain network quality parameters within a historical time period.

[0096] The network quality parameter includes at least one of an instant messaging delay, a video download rate, and a downlink roundtrip time (RTT).

[0097] Exemplarily, the network anomaly prediction device may obtain network quality parameters within a historical time period by extracting fields used for user-perceived risk prediction.

[0098] For example, the network quality parameters may be derived from signaling / user plane data of interfaces such as the evolved packet core (EPC) network S1-MME interface, S1U interface, S6A interface, S11 interface, and GN interface.

[0099] Step 403: The network anomaly prediction device inputs the network quality parameter into the network anomaly prediction model to obtain the prediction result at the target time.

[0100] The prediction results include abnormal cells and abnormal problem information.

[0101] Exemplarily, the network anomaly prediction device may also process network quality parameters by means of feature engineering.

[0102] Among them, feature engineering includes operations such as discrete variables, binning / partitioning, feature extraction, feature selection, and dimensionality reduction.

[0103] In a possible implementation, after obtaining the prediction result of the target time, the network anomaly prediction device may also send abnormal work order information to the abnormal cell according to the prediction result, and correspondingly, the abnormal cell receives the abnormal work order information from the network anomaly prediction device.

[0104] Among them, the abnormal work order information is used to indicate the maintenance of abnormal problems. In this way, relevant technicians can promptly discover abnormal cells with abnormalities, conduct on-site tests and feedback, and manage and maintain the identified parameter configuration problems and external interference problems.

[0105] Exemplarily, the prediction results may include information such as network standard, cell unit, station number, station name, cell name, cell type, equipment manufacturer, partition type, base station type, problem type, whether closed loop, closed loop deadline, closed loop date, dispatch date, cause classification, solution, etc. For example, the base station type may be macro base station, indoor base station, pole base station, etc., and the problem type may be interference problem, same frequency switching, traffic mutation, connection problem, disconnection problem, downlink packet loss problem, traffic mutation, etc. Cause classification may include high interference, parameter configuration problem, base station failure, user behavior, etc. Solutions may include interference elimination, parameter adjustment, maintenance processing, technical transformation, troubleshooting, maintenance troubleshooting, and other solutions.

[0106] In a possible implementation, the network anomaly prediction model includes an outlier detection sub-model, a parameter fitting sub-model, and a time series prediction sub-model.

[0107] Among them, the outlier detection sub-model is used to delete abnormal parameters in the network quality parameters. The parameter fitting sub-model is used to perform data fitting on the network quality parameters. The time series prediction sub-model is used to predict the network quality parameters at the target time.

[0108] Based on the above technical solution, the network anomaly prediction device obtains network quality data including network side data and user side data in a historical time period, thereby extracting parameters from the network quality data to obtain network quality parameters in the historical time period. The network quality parameters may include at least one of instant messaging delay, video download rate, and downlink round-trip delay. Afterwards, the network anomaly prediction device may input the network quality parameters into the network anomaly prediction model to obtain the prediction result at the target time. The prediction result includes abnormal cells and abnormal problem information. Compared with the solution in the prior art in which technicians manually perform wireless network quality monitoring and detection, the present application improves the detection efficiency and detection accuracy. At the same time, the network anomaly prediction device in the present application can also make predictions based on the network quality parameters in the historical time period to obtain the prediction results at the target time, so that network abnormal problems can be detected more promptly, improving the timeliness of anomaly detection.

[0109] The following describes the process by which the network anomaly prediction device predicts the prediction result at the target time.

[0110] As a possible embodiment of the present application, Figure 4 ,like Figure 5 As shown, the above step 403 can be implemented through the following steps 501 to 504.

[0111] Step 501: The network anomaly prediction device inputs the network quality parameter into the outlier detection sub-model to obtain the first network quality parameter after removing the abnormal parameter.

[0112] Among them, the outlier detection sub-model is used to remove abnormal parameters in the network quality parameters.

[0113] Exemplarily, the outlier detection sub-model can be implemented by algorithms such as density-based spatial clustering of applications with noise (DBSCAN), box plot algorithm, and Gaussian mixture model (GMM).

[0114] Step 502: The network anomaly prediction device inputs the first network quality parameter into a parameter fitting sub-model to obtain a second network quality parameter after parameter fitting.

[0115] The parameter fitting sub-model is used to perform data fitting on the first network quality parameter.

[0116] Exemplarily, the parameter fitting sub-model can be implemented by algorithms such as the least squares method and the multivariate linear regression algorithm.

[0117] Exemplarily, the second network quality parameter after parameter fitting is as follows: Figure 6 The first row of fitted data is the instant messaging delay, the second row of fitted data is the video download rate, and the third row of fitted data is the downlink round-trip delay.

[0118] Based on the above steps, the network anomaly prediction device in the present application can convert the original network quality parameters into feature data, namely, the second network quality parameters, so that the feature data can better describe potential problems, thereby improving the prediction accuracy of the time series prediction sub-model.

[0119] Step 503: The network anomaly prediction device inputs the second network quality parameter into the time series prediction sub-model to obtain the third network quality parameter at the target time.

[0120] Among them, the time series prediction sub-model is used to predict the third network quality parameter at the target time.

[0121] Exemplarily, the time series prediction sub-model can be implemented by the prophet time series prediction algorithm.

[0122] In a possible implementation, the network anomaly prediction device can also perform model training to obtain a trained time series prediction sub-model.

[0123] Exemplarily, the model training process is as follows:

[0124] First, the network anomaly prediction device can use the time series data decomposition method to convert the third network quality parameter S t Decompose into trend items t 、Periodic items (daily t , weekly t ) and the error term ε.

[0125] Afterwards, the network anomaly prediction device uses the decomposed trend item trend t and periodic items (daily t , weekly t ), predict the trend value trend at the target time (for example, one week later) t+7 and periodic forecast value (daily t+7 , weekly t+7 ).

[0126] In this way, the network anomaly prediction device can predict the trend value according to the trend t+7 、Periodic forecast value (daily t+7 , weekly t+7 ) and the error term ε are used to predict the third network quality parameter at the target time.

[0127] Exemplarily, the third network quality parameter at the target time satisfies the following formula 1:

[0128] S t+7 =trend t+7 +daily t+7 +weekly t+7 +ε Formula 1

[0129] Among them, S t+7 is the third network quality parameter at the target time (t+7 means seven days after time t, and the application can also set the predicted target time according to the actual situation), trend t+7 is the trend forecast value obtained based on the time series forecast sub-model, (daily t+7 , weekly t+7 ) is the period prediction value obtained based on the time series prediction sub-model, and ε is the error term.

[0130] Step 504: The network anomaly prediction device determines a prediction result according to the third network quality parameter.

[0131] In a possible implementation, the network anomaly prediction device may determine a dynamic threshold and an absolute threshold, and judge and detect the third network quality parameter according to the dynamic threshold and the absolute threshold, so as to obtain a prediction result.

[0132] Based on the above technical solution, the network anomaly prediction device provided in the embodiment of the present application can remove abnormal parameters in the network quality parameters through the outlier detection sub-model, and perform data trend fitting through the parameter fitting sub-model. After that, the network anomaly prediction device inputs the fitted network quality parameters into the time series prediction sub-model, predicts the third network quality parameters at the target time, and determines the prediction result based on the third network quality parameters. In this way, the above technical solution of the present application can transform wireless network optimization from passive complaint-side detection to active prevention and identification, greatly saving the evaluation time and labor costs of network optimization personnel, thereby realizing reasonable planning of resource ratios and improving operational efficiency.

[0133] The following introduces the process of data preprocessing of the network anomaly prediction device.

[0134] As a possible embodiment of the present application, Figure 4 ,like Figure 7 As shown, the method may further include the following step 701.

[0135] Step 701: The network anomaly prediction device performs data preprocessing operations on network quality data.

[0136] The data preprocessing operation includes at least one of the following: duplicate value data processing, missing value data processing, outlier data processing and data standardization processing.

[0137] Exemplarily, the network anomaly prediction device may perform data preprocessing operations according to a cross-industry standard process for data mining (CRISP-DM).

[0138] It should be noted that data preprocessing is mainly the process of constructing the final data set required by this application from the processed network quality data, which is usually based on the following principles:

[0139] a. Convert the collected data into the format required by the model;

[0140] b. Process the data features, generate corresponding fields, and perform data matching and integration;

[0141] c. Clean the entries with missing values ​​and outliers in the data;

[0142] d. Generate a high-quality dataset that is clean and can be read by the model.

[0143] Among them, duplicate data is meaningless for network anomaly prediction, and may even cause the network anomaly prediction model to overfit during the training iteration process, ultimately affecting the prediction accuracy of the network anomaly prediction model. Therefore, the network anomaly prediction device can delete duplicate data in the network quality data.

[0144] Missing value data processing can be performed according to actual conditions. For example, when the missing rate of column data or row data is high (for example, greater than 20%), the reliability of the column data or row data is poor, so the network anomaly prediction device can delete the data.

[0145] For continuous missing values, the network anomaly prediction device can use the median to fill in the missing data. For discrete type values, the network anomaly prediction device can use the mode to fill in the data.

[0146] Outlier data processing can be performed by standard deviation method to identify outliers. The standard deviation σ can characterize the degree of discreteness of the factor, which is determined based on the mean value Xmean of the factor. In the discrete value processing process, the network anomaly prediction device can measure the distance between the factor and the mean value by the mean value Xmean±3σ. Data outside this range is an outlier.

[0147] Through data standardization processing, the network anomaly prediction device can eliminate the impact of different sample data with different magnitudes on network anomaly prediction. For example, the network anomaly prediction device can perform data preprocessing on network quality data through standardization and normalization algorithms.

[0148] Based on the above technical solution, the network anomaly prediction device in the present application can perform data preprocessing operations on network quality data to avoid problems such as overfitting of the model, thereby improving the prediction accuracy of network anomaly prediction.

[0149] The embodiment of the present application can divide the network anomaly prediction device into functional modules or functional units according to the above method example. For example, each functional module or functional unit can be divided according to each function, or two or more functions can be integrated into one processing module. The above integrated module can be implemented in the form of hardware or in the form of software functional modules or functional units. Among them, the division of modules or units in the embodiment of the present application is schematic, which is only a logical function division. There may be other division methods in actual implementation.

[0150] like Figure 8 FIG. 8 is a schematic diagram of the structure of a network anomaly prediction device 80 provided in an embodiment of the present application. The network anomaly prediction device 80 includes:

[0151] The communication unit 802 is used to obtain network quality data; the network quality data includes network side data and user side data in a historical time period; the user side data includes user service configuration data and terminal device capability data.

[0152] The processing unit 801 is used to extract parameters from the network quality data to obtain network quality parameters within a historical time period; the network quality parameters include at least one of instant messaging delay, video download rate, and downlink round-trip delay.

[0153] The processing unit 801 is further used to input the network quality parameter into the network anomaly prediction model to obtain the prediction result at the target time; the prediction result includes abnormal cells and abnormal problem information.

[0154] In a possible implementation, the network anomaly prediction model includes an outlier detection sub-model, a parameter fitting sub-model and a time series prediction sub-model; the outlier detection sub-model is used to delete abnormal parameters in the network quality parameters; the parameter fitting sub-model is used to perform data fitting on the network quality parameters; and the time series prediction sub-model is used to predict the network quality parameters at the target time.

[0155] In one possible implementation, the processing unit 801 is used to: input the network quality parameter into the outlier detection sub-model to obtain the first network quality parameter after removing the abnormal parameter; input the first network quality parameter into the parameter fitting sub-model to obtain the second network quality parameter after the parameter fitting; input the second network quality parameter into the time series prediction sub-model to obtain the third network quality parameter at the target time; and determine the prediction result according to the third network quality parameter.

[0156] In a possible implementation, the processing unit 801 is further used to: perform data preprocessing operations on the network quality data; the data preprocessing operations include at least one of the following: duplicate value data processing, missing value data processing, abnormal value data processing, and data standardization processing.

[0157] When implemented by hardware, the communication unit 802 in the embodiment of the present application can be integrated on the communication interface, and the processing unit 801 can be integrated on the processor. Fig. 9 shown.

[0158] Fig. 9Another possible structural diagram of the network anomaly prediction device involved in the above embodiment is shown. The network anomaly prediction device 90 includes: a processor 902 and a communication interface 903. The processor 902 is used to control and manage the actions of the network anomaly prediction device 90, for example, to execute the steps performed by the above processing unit 801, and / or to execute other processes of the technology described herein. The communication interface 903 is used to support the communication between the network anomaly prediction device 90 and other network entities, for example, to execute the steps performed by the above communication unit 802. The network anomaly prediction device 90 may also include a memory 901 and a bus 904, and the memory 901 is used to store program codes and data of the network anomaly prediction device 90.

[0159] Among them, the memory 901 can be a memory in the network anomaly prediction device 90, etc. The memory may include a volatile memory, such as a random access memory; the memory may also include a non-volatile memory, such as a read-only memory, a flash memory, a hard disk or a solid-state drive; the memory may also include a combination of the above types of memory.

[0160] The processor 902 may be a processor that implements or executes various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of the present application. The processor may be a central processing unit, a general-purpose processor, a digital signal processor, an application-specific integrated circuit, a field programmable gate array, or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. The processor may implement or execute various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of the present application. The processor may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, and the like.

[0161] The bus 904 may be an Extended Industry Standard Architecture (EISA) bus, etc. The bus 904 may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Fig. 9 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0162] Fig. 9 The network anomaly prediction device 90 in the embodiment can also be a chip. The chip includes one or more than two (including two) processors 902 and a communication interface 903 .

[0163] In some embodiments, the chip further includes a memory 901, which may include a read-only memory and a random access memory and provide operation instructions and data to the processor 902. A portion of the memory 901 may also include a non-volatile random access memory (NVRAM).

[0164] In some implementations, the memory 901 stores the following elements, execution modules or data structures, or a subset thereof, or an extended set thereof.

[0165] In the embodiment of the present application, the corresponding operation is performed by calling the operation instruction stored in the memory 901 (the operation instruction may be stored in the operating system).

[0166] Through the description of the above implementation methods, technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above functional modules is used as an example. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. The specific working process of the system, device and unit described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0167] An embodiment of the present application provides a computer program product including instructions. When the computer program product is run on a computer, the computer is enabled to execute the network anomaly prediction method in the above method embodiment.

[0168] The embodiment of the present application also provides a computer-readable storage medium, in which instructions are stored. When the instructions are executed on a computer, the computer executes the network anomaly prediction method in the method flow shown in the above method embodiment.

[0169] Among them, the computer readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer readable storage media (a non-exhaustive list) include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a register, a hard disk, an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above, or any other form of computer readable storage medium known in the art. An exemplary storage medium is coupled to a processor so that the processor can read information from the storage medium and write information to the storage medium. Of course, the storage medium can also be a component of the processor. The processor and the storage medium can be located in an application-specific integrated circuit (ASIC). In the embodiments of the present application, a computer-readable storage medium may be any tangible medium that contains or stores a program, which may be used by or in conjunction with an instruction execution system, apparatus, or device.

[0170] Since the network anomaly prediction device, computer-readable storage medium, and computer program product in the embodiments of the present application can be applied to the above method, the technical effects that can be obtained can also refer to the above method embodiments, and the embodiments of the present application will not be repeated here.

[0171] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0172] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0173] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0174] The above are only specific implementations of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A network anomaly prediction method, characterized in that: The method comprises: Acquire network quality data; the network quality data includes network side data and user side data in a historical time period; the user side data includes user service configuration data and terminal device capability data; Extracting parameters from the network quality data to obtain network quality parameters within a historical time period; the network quality parameters include at least one of instant messaging delay, video download rate, and downlink round-trip delay; The network quality parameters are input into the network anomaly prediction model to obtain the prediction result at the target time, including: The network quality parameter is input into the outlier detection sub-model to obtain a first network quality parameter after removing the abnormal parameter; the network abnormality prediction model includes the outlier detection sub-model, the parameter fitting sub-model and the time series prediction sub-model; the outlier detection sub-model is used to delete the abnormal parameter in the network quality parameter; Inputting the first network quality parameter into the parameter fitting sub-model to obtain a second network quality parameter after parameter fitting; the parameter fitting sub-model is used to perform data fitting on the network quality parameter; Inputting the second network quality parameter into the time series prediction sub-model to obtain a third network quality parameter at the target time; the time series prediction sub-model is used to predict the network quality parameter at the target time; Determine the prediction result according to the third network quality parameter; the prediction result includes abnormal cells and abnormal problem information; The model training process is as follows: Through the time series data decomposition method, the third network quality parameter S t Decompose into trend items t 、Periodic items (daily t , weekly t ) and the error term ε; According to the decomposed trend item trend t and periodic items (daily t , weekly t ), predict the trend forecast value trend at the target time t+7 and periodic forecast value (daily t+7 , weekly t+7 ); According to the trend forecast value trend t+7 、Periodic forecast value (daily t+7 , weekly t+7 ) and the error term ε predict the third network quality parameter at the target time; The third network quality parameter at the target time satisfies the following formula: S t+7 =trend t+7 +daily t+7 +weekly t+7 +ε Among them, S t+7 is the third network quality parameter at the target time, trend t+7 is the trend forecast value obtained based on the time series forecast sub-model, (daily t+7 , weekly t+7 ) is the period prediction value obtained based on the time series prediction sub-model, and ε is the error term.

2. The method according to claim 1, characterized in that The method further comprises: The network quality data is subjected to data preprocessing operations; the data preprocessing operations include at least one of the following: duplicate value data processing, missing value data processing, abnormal value data processing, and data standardization processing.

3. A network anomaly prediction device, characterized in that: including a communication unit and a processing unit; The communication unit is used to obtain network quality data; the network quality data includes network side data and user side data in a historical time period; the user side data includes user service configuration data and terminal device capability data; The processing unit is used to extract parameters from the network quality data to obtain network quality parameters within a historical time period; the network quality parameters include at least one of instant messaging delay, video download rate, and downlink round-trip delay; The processing unit is further used to: input the network quality parameter into a network anomaly prediction model to obtain a prediction result at a target time, including: The network quality parameter is input into the outlier detection sub-model to obtain a first network quality parameter after removing the abnormal parameter; the network abnormality prediction model includes the outlier detection sub-model, the parameter fitting sub-model and the time series prediction sub-model; the outlier detection sub-model is used to delete the abnormal parameter in the network quality parameter; Inputting the first network quality parameter into the parameter fitting sub-model to obtain a second network quality parameter after parameter fitting; the parameter fitting sub-model is used to perform data fitting on the network quality parameter; Inputting the second network quality parameter into the time series prediction sub-model to obtain a third network quality parameter at the target time; the time series prediction sub-model is used to predict the network quality parameter at the target time; Determine the prediction result according to the third network quality parameter; the prediction result includes abnormal cells and abnormal problem information; The model training process is as follows: Through the time series data decomposition method, the third network quality parameter S t Decompose into trend items t 、Periodic items (daily t , weekly t ) and the error term ε; According to the decomposed trend item trend t and periodic items (daily t , weekly t ), predict the trend forecast value trend at the target time t+7 and periodic forecast value (daily t+7 , weekly t+7 ); According to the trend forecast value trend t+7 、Periodic forecast value (daily t+7 , weekly t+7 ) and the error term ε predict the third network quality parameter at the target time; The third network quality parameter at the target time satisfies the following formula: S t+7 =trend t+7 +daily t+7 +weekly t+7 +ε Among them, S t+7 is the third network quality parameter at the target time, trend t+7 is the trend forecast value obtained based on the time series forecast sub-model, (daily t+7 , weekly t+7 ) is the period prediction value obtained based on the time series prediction sub-model, and ε is the error term.

4. The device according to claim 3, characterized in that The processing unit is also used for: The network quality data is subjected to data preprocessing operations; the data preprocessing operations include at least one of the following: duplicate value data processing, missing value data processing, abnormal value data processing, and data standardization processing.

5. A network anomaly prediction device, characterized in that: include: A processor and a communication interface; the communication interface is coupled to the processor, and the processor is used to run a computer program or instruction to implement the network anomaly prediction method as described in claim 1 or 2.

6. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores instructions, and when a computer executes the instructions, the computer executes the network anomaly prediction method according to claim 1 or 2.