WO2019062497A1 - 数据分析方法和装置 - Google Patents

数据分析方法和装置 Download PDF

Info

Publication number
WO2019062497A1
WO2019062497A1 PCT/CN2018/104115 CN2018104115W WO2019062497A1 WO 2019062497 A1 WO2019062497 A1 WO 2019062497A1 CN 2018104115 W CN2018104115 W CN 2018104115W WO 2019062497 A1 WO2019062497 A1 WO 2019062497A1
Authority
WO
WIPO (PCT)
Prior art keywords
user plane
plane data
network element
information
feature
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/CN2018/104115
Other languages
English (en)
French (fr)
Inventor
崇卫微
辛阳
吴晓波
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Huawei Technologies Co Ltd
Original Assignee
Huawei Technologies Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Huawei Technologies Co Ltd filed Critical Huawei Technologies Co Ltd
Priority to EP18860414.4A priority Critical patent/EP3684139B1/en
Publication of WO2019062497A1 publication Critical patent/WO2019062497A1/zh
Priority to US16/833,195 priority patent/US11552856B2/en
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W24/00—Supervisory, monitoring or testing arrangements
    • H04W24/02—Arrangements for optimising operational condition
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
    • G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
    • G06F18/2132—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods based on discrimination criteria, e.g. discriminant analysis
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/14—Network analysis or design
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L43/00—Arrangements for monitoring or testing data switching networks
    • H04L43/04—Processing captured monitoring data, e.g. for logfile generation
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/14—Session management
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W24/00—Supervisory, monitoring or testing arrangements
    • H04W24/08—Testing, supervising or monitoring using real traffic
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W88/00—Devices specially adapted for wireless communication networks, e.g. terminals, base stations or access point devices
    • H04W88/18—Service support devices; Network management devices
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/08—Configuration management of networks or network elements
    • H04L41/0894—Policy-based network configuration management

Definitions

  • the embodiments of the present application relate to the field of communications, and in particular, to a method and apparatus for acquiring feature parameters.
  • NWDA network data analytics
  • the embodiment of the present application provides a data analysis method and a data analysis device, which can implement data analysis by using a data analysis network element in a communication network.
  • a first aspect of the present application provides a data analysis method, including: a user plane data processing network element acquires information from at least one feature set of a data analysis network element, wherein each of the at least one feature set information The information of the feature set corresponds to at least one service type or at least one execution rule; the user plane data processing network element receives user plane data; the user plane data processing network element acquires the information according to the information of the at least one feature set a feature parameter of the user plane data; the user plane data processing network element sends the feature parameter to the data analysis network element; the user plane data processing network element acquires the feature parameter from the data analysis network element The user plane data processing network element acquires the service type associated with the user plane data or the execution rule associated with the user plane data according to the response result.
  • the embodiment of the present application implements analyzing data by using a data analysis network element in a communication network.
  • the user plane data processing network element obtains, according to the response result, a service type associated with the user plane data or an execution rule associated with the user plane data, including: The user plane data processing network element acquires the execution rule associated with the user plane data from the policy control network element according to the response result; or the user plane data processing network element acquires the source according to the response result
  • the data analysis performs execution rules of the user plane data association of the network element.
  • the method further includes: the user plane data processing network element processing the location according to the service type associated with the user plane data or the execution rule associated with the user plane data User surface data.
  • the user plane data processing network element is configured according to the service type or the user plane associated with the user plane data.
  • the data association execution rule processing the user plane data including: the user plane data processing network element forwarding the user plane data according to the service priority information indicated in the execution rule associated with the user plane data; or The user plane data processing network element adds the label information of the service type to the user plane data according to the service type associated with the user plane data or the execution rule associated with the user plane data; or the user plane data processing network element is configured according to The service type associated with the user plane data or the execution rule associated with the user plane data adds scheduling priority information to the user plane data; or the service associated with the user plane data processing network element according to the user plane data The type or the execution rule associated with the user plane data performs charging statistics on the user plane data.
  • the user plane data processing network element is configured according to the service type or the user plane associated with the user plane data.
  • the data association execution rule processes the user plane data, including: the user plane data processing network element determines that the destination address of the user plane data is an address of the terminal device, and the terminal device is in an idle state; the user plane data
  • the processing network element sends the paging priority information of the terminal device to the session management network element according to the service type associated with the user plane data or the execution rule associated with the user plane data.
  • the response result includes service type information associated with the user plane data and/or execution rule information associated with the user plane data.
  • the user plane data processing network element acquires the user plane data according to the information of the at least one feature set
  • the feature parameter includes: the user plane data processing network element selects information of the partial feature set from the information of the at least one feature set according to the partial feature; and the user plane data processing network element acquires the feature of the user plane data a parameter, the feature parameter corresponding to information of the partial feature set.
  • the partial feature includes Internet Protocol IP quintuple information including the user plane data.
  • the method further includes: the user plane data processing network element acquiring, according to the information of the at least one feature set, from other user plane data processing network elements or control plane network elements And the information about the user plane data is obtained by the user plane data processing network element according to the information of the at least one feature set, where the user plane data processing network element is according to the at least one The information of the feature set and the associated information acquire feature parameters of the user plane data.
  • the user plane data processing network element includes a feature extraction unit, where the user plane data processing network element acquires the user plane data according to the information of the at least one feature set.
  • the feature extraction unit includes: acquiring, by the feature extraction unit, the feature parameter of the user plane data according to information of at least part of the feature set of the information of the at least one feature set.
  • a second aspect of the present application provides a data analysis method, including: a data analysis network element transmitting information of at least one feature set to a user plane data processing network element, wherein each of the at least one feature set information The information of the feature set corresponds to at least one service type or at least one execution rule; the data analysis network element receives feature parameters of the user plane data from the user plane data processing network element, wherein the at least one feature set The information includes information of the feature set corresponding to the feature parameter, and the data analysis network element determines, according to the feature parameter, service type information associated with the user plane data or execution rule information associated with the user plane data.
  • the method further includes: the data analysis network element, according to the partial feature, selecting information of the at least one feature set from information of the feature set acquired in advance.
  • the part of the feature includes a data network name and/or identifier information of a user plane function network element corresponding to the data network
  • the data analysis network element selects the information of the at least one feature set from the information of the feature set acquired in advance according to the part of the feature, including: the data analysis network element from the information of the pre-acquired feature set And selecting information of a feature set that is consistent with the data network name and/or the identification information of the user plane function network element corresponding to the data network.
  • the data analysis network element determines, according to the feature parameter, service type information associated with the user plane data or
  • the execution rule information associated with the user plane data includes: determining, by the data analysis network element, the service type information or the user plane data associated with the user plane data according to the feature parameter and a matching algorithm corresponding to the feature parameter Associated execution rule information.
  • the service type information associated with the user plane data includes a service type or a service type associated with the user plane data. Instructions.
  • the method further includes: the data analysis network element transmitting a response result of the feature parameter to the user plane data processing network element, where the response result includes the The service type information associated with the user plane data or the execution rule information associated with the user plane data.
  • a third aspect of the present application provides a data analysis method, including: a policy control network element acquiring information from at least one feature set of a data analysis network element, wherein each of the at least one feature set information The set information corresponds to at least one service type or at least one execution rule; the policy control network element sends information of at least part of the feature set of the information of the at least one feature set to the user plane data processing network element.
  • the method further includes: the policy control network element selecting information of the at least part of the feature set from information of the at least one feature set according to a partial feature.
  • the part of the feature includes: a data network name and/or identifier information of a user plane function network element corresponding to the data network
  • the policy control network element selects information of the at least part of the feature set from the information of the at least one feature set according to the part of the feature, including: the policy control network element from the information of the at least one feature set And selecting information of a feature set that is consistent with the data network name and/or the identification information of the user plane function network element corresponding to the data network.
  • the method further includes: the policy control network element transmitting the service type information corresponding to the information of the at least part of the feature set to the user plane data processing network element.
  • the method further includes: the policy control network element transmitting, to the user plane data processing network element, an execution rule corresponding to the information of the at least part of the feature set.
  • the method further includes: the policy control network element acquiring service type information corresponding to the information of the at least part of the feature set of the data analysis network element; The policy control network element generates an execution rule corresponding to the information of the at least part of the feature set according to the acquired service type information; the policy control network element sends the information of the at least part of the feature set to the user plane data processing network element Corresponding execution rules.
  • a fourth aspect of the present application provides a data analysis method, including: a user plane data processing network element acquires information from a feature set of a data analysis network element, where the information of the feature set corresponds to a service type or a The user plane data processing network element receives the user plane data; the user plane data processing network element acquires the feature parameter of the user plane data according to the information of the feature set; the user plane data processing network element Transmitting, to the data analysis network element, the feature parameter; the user plane data processing network element acquiring a response result of the feature parameter from the data analysis network element; the user plane data processing network element according to the response As a result, the service type associated with the user plane data or the execution rule associated with the user plane data is obtained.
  • the fourth aspect of the present application may also include the first to fourth possible implementations of the first aspect.
  • a fifth aspect of the present application provides a data analysis method, including: a data analysis network element transmitting information of a feature set to a user plane data processing network element, where the information of the feature set corresponds to a service type or a type Performing a rule; the data analysis network element receives a feature parameter of the user plane data from the user plane data processing network element, wherein the feature parameter corresponds to information of the feature set; and the data analysis network element is configured according to the The feature parameter determines service type information associated with the user plane data or execution rule information associated with the user plane data.
  • a sixth aspect of the present application provides a data analysis method, including: a policy control network element acquiring information from a feature set of a data analysis network element, wherein the information of the feature set corresponds to a service type or an execution a rule: the policy control network element sends the information of the feature set to a user plane data processing network element.
  • a seventh aspect of the present application provides an apparatus for data analysis, including: a transceiver unit, configured to acquire information from at least one feature set of a data analysis network element and to receive user plane data, wherein the at least one The information of each feature set in the information of the feature set corresponds to the at least one service type or the at least one execution rule; the processing unit is configured to acquire the feature parameter of the user plane data according to the information of the at least one feature set; The transceiver unit is further configured to send the feature parameter to the data analysis network element and obtain a response result of the feature parameter from the data analysis network element; the processing unit is further configured to acquire the location according to the response result.
  • the service type associated with the user plane data or the execution rule associated with the user plane data is configured to acquire information from at least one feature set of a data analysis network element and to receive user plane data, wherein the at least one The information of each feature set in the information of the feature set corresponds to the at least one service type or the at least one execution rule; the processing unit is configured to acquire the
  • An eighth aspect of the present invention provides an apparatus for data analysis, including: a transceiver unit, configured to send information of at least one feature set to a user plane data processing network element, and receive a user from the user plane data processing network element a feature parameter of the face data, wherein the information of each feature set in the information of the at least one feature set corresponds to at least one service type or at least one execution rule, and the information of the at least one feature set includes the feature parameter And a processing unit, configured to determine, according to the feature parameter, service type information associated with the user plane data or execution rule information associated with the user plane data.
  • the ninth aspect of the present application provides an apparatus for data analysis, including: a processing unit and a transceiver unit, wherein the processing unit is configured to acquire, by using the transceiver unit, information from at least one feature set of a data analysis network element, The information of each feature set of the at least one feature set corresponds to at least one service type or at least one execution rule.
  • the processing unit is further configured to process the network element to the user plane through the transceiver unit. Transmitting information of at least a portion of the feature set of the at least one feature set.
  • the information of the feature set is a set of feature indexes.
  • the feature parameter is a feature vector.
  • the feature parameter is a set of feature values.
  • a tenth aspect of the present application provides an apparatus for data analysis, comprising: a storage unit configured to store computer instructions; and a processing unit configured to perform the first to sixth aspects according to computer instructions stored in the storage unit Aspects and any of a variety of possible implementations.
  • a computer storage medium stores instructions that, when run on a computer, cause the computer to perform the first to sixth aspects and various Any of the possible implementations.
  • a computer program product comprising instructions which, when run on a computer, cause the computer to perform any of the first to sixth aspects and various possible implementations described above method.
  • FIG. 1 is a schematic diagram of a communication system implementing an embodiment of the present application.
  • FIG. 2 is a schematic flowchart of a data analysis method according to a first embodiment of the present application.
  • FIG. 3 is a schematic flowchart of a data analysis method according to a second embodiment of the present application.
  • FIG. 4 is a schematic flowchart of a data analysis method according to a third embodiment of the present application.
  • FIG. 5 is a schematic flowchart of a data analysis method according to a fourth embodiment of the present application.
  • FIG. 6 is a schematic flowchart of a data analysis method according to a fifth embodiment of the present application.
  • FIG. 7 is a schematic flowchart of a data analysis method according to a sixth embodiment of the present application.
  • FIG. 8 is a schematic diagram of a data analysis apparatus according to an embodiment of the present application.
  • the embodiments of the present application can be used for long term evolution (LTE), 5G or next generation networks, fixed networks, home base station networks, mobile networks that are not accessed by 3GPP (such as wifi), and the like.
  • LTE long term evolution
  • 5G or next generation networks fixed networks
  • 3GPP such as wifi
  • an example is applied to a 5G network.
  • FIG. 1 is a schematic diagram of a communication system capable of implementing an embodiment of the present application.
  • the terminal device 101 accesses the core network through an access network (AN) device 102.
  • AN access network
  • the terminal device 101 includes but is not limited to: user equipment (UE), user unit, user station, mobile station, mobile station, remote station, remote terminal device, mobile terminal device, user terminal device, terminal device, Wireless communication device, user agent, user device, cellular phone, cordless phone, session initiation protocol (SIP) phone, wireless local loop (WLL) station, personal digital assistant (PDA) ), handheld devices with wireless communication capabilities, computing devices, processing devices connected to wireless modems, in-vehicle devices, wearable devices, terminal devices in the Internet of Things, home appliances, virtual reality devices, terminal devices in future 5G networks
  • PLMN public land mobile network
  • Access network device 102 can be a device that communicates with terminal device 101.
  • the access network device can provide communication coverage for a particular geographic area and can communicate with terminal devices located within the coverage area (cell).
  • Access network device 102 can communicate with any number of terminal devices.
  • the access network device can support different standard communication protocols or can support different communication modes.
  • the access network device 102 is an evolved base station (eNodeB), or a wireless fidelity access point (WiFi AP), or a global microwave access interoperability base station ( Worldwide interoperability for microwave access base station (WiMAX BS), or a wireless controller in a cloud radio access network (CRAN), or the network device may be an access network device or a future in a future 5G network An access network device or the like in an evolved PLMN.
  • eNodeB evolved base station
  • WiFi AP wireless fidelity access point
  • WiMAX BS global microwave access interoperability base station
  • CDRF cloud radio access network
  • the network device may be an access network device or a future in a future 5G network
  • the core network may include: a control plane function (CPF) network element, a user plane function (UPF) network element 103, a policy control function (PCF) network element 104, and an NWDA network element 105.
  • the control plane function network element may include an access management function (AMF) network element 106 and a session management function (SMF) network element 107.
  • the user plane data transmission between the terminal device 101 and the data network (DN) 108 can be implemented by the access network device 102 and the user plane function network element 103.
  • the PCF network element 104 has the function of policy control decision making to provide a policy for the network.
  • the NWDA network element 105 is used for big data learning and analysis.
  • the AMF network element 106 is used for mobility management, lawful interception, or access authorization and authentication.
  • the SMF network element 107 is used to implement session and bearer management, address allocation, and the like.
  • the DN 108 is a network for transmitting data.
  • the DN 108 may be an internet protocol (IP) multimedia subsystem (IMS) server or a packet data network (PDN) or an application server (application). Server, App server).
  • IP internet protocol
  • IMS internet protocol multimedia subsystem
  • PDN packet data network
  • application application server
  • Server App server
  • a network element obtains information from another network element (for example, a B network element), and may refer to that the A network element directly receives information from the B network element, or may The A network element receives information from the B network element through other network elements (for example, the C network element).
  • the C network element can transparently transmit information, and can also process the information, for example, carrying the information in different messages for transmission or filtering the information. Only the filtered information is sent to the A network element.
  • the A network element sends information to the B network element, which may be that the A network element directly sends information to the B network element, and may also refer to the A network element passing through other network elements (for example, the C network. Meta) sends information to the B network element.
  • the B network element may be that the A network element directly sends information to the B network element, and may also refer to the A network element passing through other network elements (for example, the C network. Meta) sends information to the B network element.
  • FIG. 2 is a schematic flowchart of a data analysis method according to a first embodiment of the present application.
  • the data analysis method includes:
  • the data analysis network element obtains training data.
  • the data analysis network element may be the NWDA network element in FIG.
  • the data analysis network element may also be another network element having a network data analysis function, which is not limited herein.
  • the data analysis network element can obtain training data from other network elements respectively, and the data analysis network element can also obtain training data from the same network element set.
  • other network elements may be telecommunication network devices or third-party servers.
  • the telecommunication network device may be at least one of the following devices: a terminal device, an access network device, a control plane function network element (for example, an AMF network element or an SMF network element), a UPF network element, a PCF network element, and a network management.
  • a network element for example, a business support system (BSS) or an operation support system (OSS) or a management support system (MSS)
  • UDM unified data management
  • Yuan and IMS network elements may be at least one of an App Server, an OTT (over the top) server, and a vertical industry control center.
  • the data analysis network element can directly obtain training data by performing data interaction with other network elements.
  • the data analysis network element can also obtain training data indirectly through other network elements.
  • the data analysis network element obtains data from a third-party server through a network exposure function (NEF) network element.
  • NEF network exposure function
  • the data analysis network element can obtain training data from other network elements in real time, and the data analysis network element can also obtain training data when the data analysis network element and/or other network elements are idle.
  • the training data acquired by the data analysis network element may be original data.
  • the training data obtained by the data analysis network element may also be the data preprocessed by other network elements.
  • the application server cleans sensitive information in the original data for the purpose of protecting user privacy, and sends the processed data to the data analysis network element.
  • the training data acquired by the data analysis network element may be network data, such as an address of a terminal device, a cell identifier (cell ID), time information, or a network congestion status.
  • the training data obtained by the data analysis network element may also be application data, such as: IP quintuple, user plane data size, user plane data interval, service type, service experience, or extended field.
  • the data analysis network element may also associate the acquired application data with the network data to obtain the associated training data. For example, the data analysis network element associates the application data with the network data according to the address and/or time information of the terminal device.
  • the address of the terminal device can be an IP address or an Ethernet address.
  • the data analysis network element can obtain the training data by using the service type as the granularity. For example, the data analysis network element obtains training data of the video service, training data of the payment service, or training data based on the voice over long term evolution (VOLTE) service, respectively.
  • VOLTE voice over long term evolution
  • the data analysis network element can also obtain the training data by using the network element as the granularity.
  • the training data acquired from the terminal device may include: a terminal type, an address of the terminal device, an operating system version, a temperature of the terminal device, a power quantity of the terminal device, or a cell wireless channel quality measured by the terminal device, and the like;
  • the obtained training data may include: an identifier (ID) of the session management network element, an address of the terminal device, or a data network name (DNN), etc.
  • the training data obtained from the user plane function network element may include: User plane function network element ID, tunnel end point identifier (TEID), congestion level, IP quintuple, user plane data size, or number of user plane data;
  • training data obtained from the access network device can be Including: cell identity, quality of service (Qos) parameters, real-time wireless channel quality (eg, reference signal receiving power (RSRP), or reference singular received quality (RSRQ), Or signal to interference plus noise (signal to interference plus noise Rat
  • the data analysis network element analyzes the training data to obtain information of the feature set.
  • the data analysis network element analyzes the training data obtained in step 201 by using a big data analysis method, and acquires information of at least one feature set.
  • the information of the feature set may be a specific feature set, or may be information corresponding to the feature set, for example, an index of the feature set.
  • the information of each of the at least one feature set corresponds to at least one service type or at least one execution rule.
  • the execution rule may be at least one of an enforcement policy, a control policy, a charging policy, and a policy & charging control (PCC) rule.
  • PCC rules can include quality of service policies.
  • the data analysis network element separately analyzes the training data of various service types to obtain the information of the feature set corresponding to each service type.
  • One type of service may uniquely correspond to information of one feature set; one type of service may also correspond to information of multiple feature sets; information of one feature set may uniquely correspond to one type of service, and information of one feature set is also Can correspond to multiple business types.
  • the corresponding service type may also be determined by combining other conditions or characteristics other than the information of the feature set.
  • the data analysis network element may determine the execution rule information of the service type according to the service type corresponding to the information of the feature set, and the execution rule information of the service type is the execution rule information corresponding to the information of the feature set. For example, when the service type is a payment service, the processing priority of the payment type service is determined.
  • the execution rule information may be a specific execution rule, or may be information related to the execution rule, for example, used to obtain execution rule information.
  • the data analysis network element may also determine execution rule information corresponding to the information of the feature set according to the content in the information of the feature set.
  • the data analysis network element may determine an execution rule corresponding to the feature set according to at least one of a radio channel quality, a congestion level, a packet loss rate, and a handover threshold in the feature set.
  • the data analysis network element may determine information related to execution rules based on the training data, such as: radio channel quality, congestion level, packet loss rate, or handover threshold.
  • the data analysis network element can obtain the characteristics of the service granularity. For example, if the size of the user plane data corresponding to the training data of a certain service is a certain value or belongs to a specific range, the data analysis network element may use the size of the user plane data as a feature of the service. For another example, if the user plane data corresponding to the training data of a certain service is from a data network of a certain feature or characteristics, the data analysis network element may use the data network name as another feature of the service.
  • the data analysis network element can acquire the feature of the terminal device granularity. For example, the user plane data characteristics of a certain service corresponding to terminal devices produced by different vendors may be different. When it is necessary to distinguish such services of different terminal devices, the data analysis network element may use the type of the terminal device as a feature of the service. Optionally, the data analysis network element can distinguish the type of the terminal device according to a type allocation code (TAC) in an international mobile equipment identity (IMEI). Therefore, the data analysis network element can use the terminal.
  • TAC type allocation code
  • IMEI international mobile equipment identity
  • the IMEI of the device is a feature.
  • user plane data characteristics of a certain service corresponding to terminal devices of different operating systems may be different. Therefore, the data analysis network element can also use the type of the operating system of the terminal device as a feature of the service.
  • the data analysis network element can also acquire the characteristics of the user granularity. For example, if a user is a merchant user, the terminal device held by the user performs a certain service for a long time in a fixed location range. Therefore, the data analysis network element can use the location information of the terminal device as a feature of the service.
  • the data analysis network element may obtain public network address information and port number information of the terminal device from the application server. The data analysis network element can then query the network address translation (NAT) network element for the intranet address information corresponding to the public network address.
  • the data analysis network element acquires user identification information, location information, and the like corresponding to the intranet address information from the network management.
  • NAT network address translation
  • the data analysis network element analyzes the training data to obtain the following feature set:
  • terminal device type terminal device location, time, the size of the first packet in the data flow, the mean of all packet sizes in the data stream, and the entropy value of all packet sizes in the data stream>.
  • the data analysis network element analyzes the training data to obtain the following feature set:
  • terminal equipment type terminal equipment location
  • cell identity time
  • time the average of the time interval of all data packets in the data stream
  • entropy value the entropy value of the time interval of all data packets in the data stream.
  • feature set of the video service and the payment service in the present application is merely an example for convenience of understanding.
  • the feature set of the video service and the payment service in the present application may also be other content, which is not limited in this application. .
  • the data analysis network element may aggregate the feature sets corresponding to the various service types to obtain a total feature set, and set an index for each feature in the total feature set. Based on the total feature set, the data analysis network element can obtain an index of the feature set corresponding to each service.
  • the data analysis network element may also summarize feature sets corresponding to the part of the service type, obtain a total feature set corresponding to the part of the service, and set a feature index for each feature in the feature set.
  • the data analysis network element can combine the feature sets of the video service and the payment service to obtain a total feature set as shown in Table 1.
  • the set of feature indexes corresponding to the video service is ⁇ 1, 2, 4, 5, 6, 7>
  • the set of feature indexes corresponding to the payment service is ⁇ 1, 2, 3, 4, 8, 9 >.
  • the feature index corresponding to a feature may be represented by a binary character
  • the set of feature indexes corresponding to a service type may be represented by a binary string.
  • the index of the feature set corresponding to one service type can be represented by a binary string of n bits. For example, when the i-th bit in the binary string corresponding to a service type is 1, the feature set corresponding to the service type includes the i-th feature in the total feature set.
  • the binary character string corresponding to the set of feature indexes ⁇ 1, 2, 4, 5, 6, 7> of the video service is 110111100, and the feature index of the payment service is set.
  • the corresponding binary string of ⁇ 1, 2, 3, 4, 8, 9> is 111100011.
  • the data analysis network element may also acquire a matching algorithm corresponding to the information of each feature set.
  • the matching algorithm can be obtained by analyzing the training data by the big data. For example, the data analysis network element acquires the training data of a certain service type, and then acquires the information of the feature set corresponding to the training data of the service type and the characteristic parameter (the characteristic parameter). It can be a collection of feature values, such as a feature vector.
  • the data analysis network element obtains a matching algorithm corresponding to the information of the feature set by using a big data analysis method according to the obtained feature parameters.
  • the data analysis network element may also acquire a matching algorithm corresponding to the information of each feature set by a pre-configured method.
  • the matching algorithm can be a mathematical function or a data model.
  • the data analysis network element associates the matching algorithm with a service type or an execution rule corresponding to the information of the feature set. Association can be understood as establishing a mapping relationship.
  • the input to the matching algorithm can be a characteristic parameter of the user plane data.
  • the output of the matching algorithm is whether the feature parameter conforms to the matching algorithm.
  • the service type associated with the user plane data corresponding to the input feature parameter may be determined according to the output result and the service type associated with the matching algorithm.
  • the output result of the matching algorithm is 1, it indicates that the input characteristic parameter conforms to the matching algorithm, and the service type associated with the matching algorithm is the service type associated with the corresponding user plane data; if the output result of the matching algorithm is 0, If the input feature parameter does not match the matching algorithm, the service type associated with the matching algorithm is not the service type associated with the corresponding user plane data.
  • the data analysis network element associates the output result of the matching algorithm with a service type or an execution rule. Association can be understood as establishing a mapping relationship. For example, the output result is 0 corresponding to the first service type or the first execution rule, and the output result is 1 corresponding to the second service type or the second execution rule, and the output result is 2 to 4 corresponding to the third service type or the first Three execution rules.
  • the input of the matching algorithm may be a characteristic parameter of the user plane data, and the service type associated with the user plane data or the execution rule may be determined according to the output result of the matching algorithm. In this scenario, the output of the matching algorithm may be service type information or execution rule information.
  • the data analysis network element sends information of the at least one feature set to the policy control network element.
  • the policy control network element may be the PCF network element in FIG. 1 or other network elements having the policy control function, which is not limited herein.
  • the data analysis network element may send the information of the feature set corresponding to the multiple service types to the policy control network element at the same time, and the data analysis network element may also send the information of the feature set corresponding to the multiple service types to the policy control network element.
  • the data analysis network element may actively send the feature set information to the policy control network element, or may send the feature set information according to the policy control network element request.
  • the data analysis network element may send the information of the feature set to the policy control network element in real time, and may also send the information of the feature set to the policy control network element when the network is idle, which is not limited herein.
  • the data analysis network element may select the information of the partial feature set from the information of the feature set acquired in advance in step 202 according to the partial feature, and then use the information of the selected partial feature set as the at least one.
  • the information of the feature set is sent to the policy control network element.
  • Some features may be some or some of the features in the feature set, or may be features that are not in the feature set but are associated with the feature set. For example, some features may be DNN and/or identification information of UPF network elements corresponding to the data network.
  • the data analysis network element may receive a message from the session management network element, where the message includes the identification information of the DNN and/or the UPF network element, and the data analysis network element selects the DNN and/or from the information of the feature set acquired in advance.
  • the data analysis network element may also send information of the partial features associated with the information of the feature set to the policy control network element.
  • Some features may be some or some of the features in the feature set, or may be features that are not in the feature set but are associated with the feature set.
  • some of the features may be DNN and/or identification information of the UPF network element corresponding to the data network, and some of the features may also be address information of the server where the service is located, part of the feature, or IP quintuple information.
  • the information of the partial feature may be sent to the policy control network element together with the information of the associated feature set, and the information of the partial feature may also be sent to the policy control network element separately from the information of the feature set.
  • the data analysis network element may further send the service type information corresponding to the information of the at least one feature set to the policy control network element.
  • the service type information may be a service type, or may be indication information of a service type, for example, at least one of a service type number, a number corresponding to the information of the feature set, and an output result of the matching algorithm.
  • the service type information may be sent to the policy control network element in the information of the corresponding feature set, and the service type information may also be separately sent to the policy control network element.
  • the data analysis network element may further send execution rule information corresponding to the information of the at least one feature set to the policy control network element.
  • the execution rule information may be the information of the feature related to the execution rule, or may be a specific execution rule, or may be the indication information of the execution rule, for example, the number of the execution rule, which is not limited herein.
  • Table 2 below illustrates the content that the data analysis network element sends to the policy control network element:
  • the information of the feature set is a set of feature indexes, and the set of feature indexes is numbered, and the set of one feature index corresponds to one service type.
  • the service type information is the number of the service type.
  • the execution rule information is a feature related to the execution rule, and the feature related to the execution rule is a wireless channel quality switching threshold of the service.
  • a partial feature associated with a business type or a collection of feature indexes is a network name.
  • the data analysis network element may only send the set of feature indices in Table 2 to the policy control network element.
  • the data analysis network element may further send at least one of a number, a service type, a handover threshold, and a network name to the policy control network element.
  • the policy control network element receives information of the at least one feature set sent by the data analysis network element.
  • the policy control network element sends information of at least part of the feature set in the at least one feature set to the user plane data processing network element.
  • the user plane data processing network element may be the user plane function network element in FIG. 1, or may be an access network device, or may be another network element having a user plane data processing function.
  • the policy control network element may send the information of the at least part of the feature set to the user plane data processing network element via the session management network element.
  • the policy control network element may send the received information of the at least one feature set to the user plane data processing network element.
  • the policy control network element may also select information of the partial feature set from the received information of the at least one feature set according to the partial feature.
  • Some features may be features in a feature set, or features that are not in the feature set but are associated with the feature set.
  • the policy control network element receives some features from the session management network element.
  • the part of the feature may be the identification information of the DNF and/or the UPF network element corresponding to the data network.
  • the policy control network element selects information of the feature set that is consistent with the identifier information of the DNN and/or the UPF network element from the received information of the at least one feature set, and sends the information of the at least part of the feature set to the user plane data processing network element.
  • the policy control network element may receive the set of five feature indexes in the table 2 from the data analysis network element. If the policy control network element receives the session management message from the session management network element, the network name included in the session management message For DNN-1, the policy control network element may send only the set of two feature indexes corresponding to DNN-1 to the user plane data processing network element.
  • the policy control network element can receive service type information from the data analysis network element.
  • the service type information may be included in the information of the corresponding feature set, and the policy control network element may also receive the information of the feature set and the service type information corresponding to the information of the feature set, respectively.
  • the policy control network element may also not receive service type information from the data analysis network element.
  • the policy control network element may analyze the service type corresponding to the information of the feature set according to some features in the feature set. For example, the information of the service type corresponding to the information of the feature set is determined according to the IP quintuple information.
  • the policy control network element may send the service type information corresponding to the information of the at least part of the feature set to the user plane data processing network element.
  • the service type information may be sent together with the information of the at least part of the feature set, or may be separately sent with the information of the at least part of the feature set.
  • the policy control network element may generate an execution rule for the information of the at least part of the feature set according to the acquired service type information, that is, generate an execution rule for the service type corresponding to the information of the at least part of the feature set.
  • the policy control network element may also receive an execution rule from the data analysis network element, the execution rule corresponding to the information of the at least part of the feature set.
  • the policy control network element may further receive an execution rule related feature from the data analysis network element, and determine an execution rule corresponding to the information of the at least part of the feature set according to the received feature.
  • the policy control network element may receive the handover threshold in Table 2 from the data analysis network element, and generate an execution rule for the set of feature indexes corresponding to the handover threshold according to the received handover threshold.
  • the policy control network element may simultaneously obtain the feature set information and the execution rule information corresponding to the feature set information from the data analysis network element, and the policy control network element may also receive the feature set information and the feature set from the data analysis network element respectively. Execution rule information corresponding to the information.
  • the policy control network element may send the execution rule information corresponding to the information of the at least part of the feature set to the user plane processing network element.
  • the execution rule information may be sent together with the information of the at least part of the feature set, or may be sent separately from the information of the at least part of the feature set, which is not limited herein.
  • the user plane data processing network element receives information from the at least part of the feature set of the policy control network element.
  • the user plane data processing network element may store information of the received feature set.
  • the user plane data processing network element receives the feature index set
  • the user plane data processing network element acquires the corresponding feature set according to the set of feature indexes.
  • the user plane data processing network element when the user plane data processing network element cannot directly acquire some features in the information of the feature set, the user plane data processing network element can convert some features that cannot be directly obtained into directly acquireable features. .
  • the user plane data received by the user plane data processing network element does not necessarily include the user identifier information. If the feature set includes the user identifier information, the user plane data processing network element can query the IP address pool, obtain the IP address corresponding to the user identifier information, and then associate the IP address with the user identifier information. Therefore, the user plane data processing network element can match the IP address in the received user plane data with the IP address in the feature set.
  • the user plane data processing network element may receive service type information from the policy control network element, where the service type information corresponds to the information of the feature set received by the user plane data processing network element.
  • the service type information corresponds to the information of the feature set received by the user plane data processing network element.
  • the user plane data processing network element may further receive execution rule information from the policy control network element, where the execution rule information corresponds to information of the feature set received by the user plane data processing network element.
  • the form of the execution rule information and the receiving manner are shown in step 204, and are not described here.
  • the data analysis network element and the user plane data processing network element can obtain the total feature set. Therefore, after the user plane data processing network element receives the index of the feature set, the user plane data processing network element acquires the feature set corresponding to the index of the feature set according to the total feature set and the index of the feature set.
  • This method can further reduce the data throughput between network elements and avoid the need to pass the definition of the characteristics of the service association between the network elements each time.
  • the data analysis network element may synchronize the total feature set with the user plane data processing network element. For example, the data analysis network element sends the total feature set to the user plane data processing network element.
  • the data analysis network element can synchronize the total feature set with the user plane data processing network element in real time, or it can be a periodic synchronization total feature set.
  • the data analysis network element and the user plane data processing network element can also obtain the total feature set in a pre-configured manner.
  • Steps 203 and 204 are optional steps.
  • the data analysis network element also sends at least one feature set information, service type information, and feature set corresponding to the feature set information to the user plane data processing network element through the service management interface. At least one of the execution rule information corresponding to the information.
  • the user plane data processing network element receives the user plane data, and obtains the feature parameter of the user plane data according to the received information of at least part of the feature set.
  • the user plane data may be a data packet.
  • the user plane data processing network element receives the feature set, after the user plane data processing network element receives the user plane data, the user plane data processing network element acquires the feature parameter of the user plane data according to the received feature set. If the user plane data processing network element receives the feature index set, after the user plane data processing network element receives the user plane data, the user plane data processing network element determines the feature corresponding to the index according to the feature index, and then obtains the user plane. The data corresponds to the feature parameter of the feature. Further, the user plane data processing network element can acquire the feature parameter of the user plane data, and the feature parameter corresponds to the received feature index set.
  • the user plane data processing network element may acquire multiple feature parameters of the user plane data, where the multiple feature parameters and the received information of the multiple feature sets are respectively correspond.
  • the information of the partial feature set may be selected from the information of the received feature set according to the partial features.
  • the user plane data processing network element acquires a feature parameter of the user plane data, and the feature parameter corresponds to the information of the selected partial feature set.
  • Some features may be features in a feature set, or features that are not in the feature set but are associated with the feature set. For example, some features can be IP quintuple information.
  • the user plane data processing network element obtains the information of the feature set from the data analysis network element and the IP quintuple information corresponding to the information of the feature set. Specifically, some of the features may be an IP address and/or a port number.
  • the user plane data processing network element selects, from the information of the received feature set, information of a feature set whose IP address and/or port number match the IP address and/or port number corresponding to the user plane data.
  • the user plane data processing network element may sequentially acquire the feature parameters of the user plane data corresponding to the feature set.
  • the user plane data processing network element may also acquire the feature parameters of the partial features in the feature set. If the feature parameter does not meet the preset condition, the user plane data processing network element stops acquiring the feature parameters of the user plane data corresponding to the feature set.
  • the feature set includes the following features: user plane data size, interval, entropy, IP address, and port number.
  • the user plane data processing network element may first obtain the IP address and port number of the user plane data in the IP quintuple. If the IP address and port number of the user plane data in the IP quintuple does not meet the preset condition, the user plane data processing network element stops acquiring the feature parameters of the feature set.
  • the user plane data processing network element may be sent to other network elements (for example, other user plane data processing network elements or control planes).
  • the network element acquires the feature parameter corresponding to the feature.
  • the feature parameter corresponding to the feature that the user plane data processing network element cannot obtain is called the information associated with the user plane data.
  • the user plane data processing network element obtains the location information of the terminal device from the access network device.
  • the location information of the terminal device is an associated information of the user plane data.
  • the user plane data processing network element may request the access network device to send the location information of the terminal device, and the user plane data processing network element may also subscribe the location information of the terminal device to the access network device, of course, the user plane data processing network element.
  • the location information of the terminal device may also be obtained from the session management network element.
  • the user plane data processing network element aggregates the feature parameters acquired by itself and the feature parameters obtained from other network elements, so that the feature parameters of the user plane data can be obtained.
  • the feature parameter may be an actual value corresponding to the feature of the user plane data, or may be a processed value, such as a normalized value or a quantized value.
  • the feature parameter may also be the specific content of the feature, and is not necessarily a specific value, such as: DNN or type of terminal device.
  • the feature parameters of the user plane data may be embodied in the form of a feature vector.
  • the user plane data processing network element sends the feature parameter of the user plane data to the data analysis network element.
  • the user plane data processing network element sends the feature parameters obtained in step 206 to the data analysis network element.
  • the user plane data processing network element may send the obtained feature parameter to the data analysis network element via the session management network element and the policy control network element.
  • the user plane data processing network element may also send the acquired feature parameter to the data analysis network element via the session management network element and the service interface between the session management network element and the data analysis network element.
  • the user plane data processing network element may further send association information of the feature parameter to the data analysis network element, for example, the number of the information of the feature set.
  • the data analysis network element determines, according to the feature parameter, service type information and/or execution rule information associated with the user plane data.
  • the data analysis network element After receiving the feature parameter of the user plane data, the data analysis network element determines the service type information or the execution rule information associated with the user plane data according to the feature parameter and the matching algorithm obtained in step 202.
  • the service type information may be a specific service type or an indication of the service type.
  • the execution rule information may be a specific execution rule or a feature related to the execution rule. For details, refer to step 203.
  • each feature parameter may be input into a corresponding matching algorithm, and then the service type associated with the user plane data is determined according to the output result of the matching algorithm. Information or execution rule information.
  • the matching algorithm corresponding to the feature parameter is a matching algorithm corresponding to the information of the feature set corresponding to the feature parameter.
  • the matching algorithm may be determined by using the association information of the feature parameter.
  • the association information of the feature parameter may be a number corresponding to the information of the feature set (for example, the number in Table 2).
  • each feature parameter may also be input into multiple matching algorithms pre-acquired by the data analysis network element, and then according to the matching algorithm.
  • the output determines the service type information or execution rule information associated with the user plane data.
  • the data analysis network element may receive the feature vector 1 and the feature vector 2 corresponding to the set of feature indices numbered 1 and 2 from the user plane data processing network element.
  • the data analysis network element inputs the feature vector 1 to the matching algorithm 1 corresponding to the set of feature indexes numbered 1, and the obtained output result is 0.
  • the data analysis network element inputs the feature vector 2 to the matching algorithm 2 corresponding to the set of feature indexes numbered 2, and the obtained output result is 1.
  • the data analysis network element may use the service type 2 associated with the matching algorithm 2 as the service type associated with the user plane data.
  • the data analysis network element may also determine execution rule information corresponding to the user plane data according to the service type information associated with the user plane data, as shown in step 202.
  • the data analysis network element sends a response result of the feature parameter to the user plane data processing network element.
  • the response result may include service type information and/or execution rule information of the user plane data.
  • the data analysis network element may send the response result of the feature parameter to the user plane data processing network element via the policy control network element and the session management network element.
  • the data analysis network element may also send the response result of the feature parameter to the user plane data processing network element through the service management interface through the session management network element.
  • the response result may include a service type associated with the user plane data, for example, the service type associated with the user plane data determined by the data analysis network element in step 208; or the response result may include indication information of the service type associated with the user plane data, For example, the number of the set of feature indexes (eg, number 2 in step 208) or the number of the service type; or the result of the response may include a feature parameter corresponding to the service type associated with the user plane data, for example, the feature in step 208 Vector 2.
  • the response result may further include whether the feature parameter meets the determination result of the corresponding matching algorithm, for example, the result of the matching algorithm output 0 or 1 in step 208.
  • the policy control network element may generate corresponding execution rule information for the service type associated with the user plane data according to the response result, as shown in step 203. Then, the policy control network element sends the generated execution rule information to the user plane data processing network element.
  • the response result may also include execution rule information corresponding to the user plane data.
  • the data analysis network element may determine the execution rule information corresponding to the user plane data according to the service type information associated with the user plane data. For details, refer to step 202.
  • the user plane data processing network element obtains a response result of the feature parameter from the data analysis network element, and obtains a service type associated with the user plane data and/or an execution rule associated with the user plane data according to the response result.
  • the user plane data processing network element may receive the response result of the feature parameter from the data analysis network element via the policy control network element and the session management network element. If the service type is included in the response, the received service type is determined as the service type associated with the user plane data; if the response result includes the service type indication information, the service type associated with the user plane data is obtained according to the indication information.
  • the user plane data processing network element may obtain the execution rule associated with the user plane data according to the service type information associated with the household data; the user plane data processing network element may also obtain the execution rule of the user plane data association from the policy control network element according to the response result. For example, the user plane data processing network element obtains the execution rule from the policy control network element according to the service type information included in the response result; the user plane data processing network element can also directly obtain the execution rule associated with the user plane data from the response result.
  • the user plane data processing network element processes the user plane data according to the service type associated with the user plane data and/or the execution rule associated with the user plane data.
  • the user plane data processing network element processes the user plane data according to the service type associated with the user plane data. For example, the user plane data processing network element adds the label information of the service type to the user plane data according to the service type associated with the user plane data; or the user plane data processing network element forwards the user plane data according to the priority information of the service; or, the user plane The data processing network element adds scheduling priority information to the user plane data according to the service type associated with the user plane data; or the user plane data processing network element determines that the destination address of the user plane data is the address of the terminal device and the terminal device is in an idle state, The user plane data processing network element sends the paging priority information of the terminal device to the session management network element according to the service type associated with the user plane data; or the user plane data processing network element performs the user plane data according to the service type associated with the user plane data. Billing statistics.
  • the user plane data processing network element processes the user plane data according to an execution rule associated with the user plane data. For example, the user plane data processing network element forwards the user plane data according to the service priority information indicated in the execution rule; or the user plane data processing network element adds the label information of the service type to the user plane data according to the execution rule; or, the user plane data
  • the processing network element adds scheduling priority information to the user plane data according to the execution rule; or the user plane data processing network element determines that the destination address of the user plane data is the address of the terminal device and the terminal device is in an idle state, and the user plane data processing network element is configured according to
  • the execution rule sends the paging priority information of the terminal device to the session management network element; or the user plane data processing network element performs charging statistics on the user plane data according to the execution rule associated with the user plane data.
  • the user plane data processing network element obtains information from at least one feature set of the data analysis network element.
  • the user plane data processing network element acquires the service type associated with the user plane data or the execution rule associated with the user plane data according to the received information of the feature set, thereby implementing data analysis using the data analysis network element in the communication network.
  • the big data analysis method is used to obtain the service type or execution rule associated with the user plane data, and the deep packet inspection (DPI) of the user plane data is not required, and the method in this embodiment is simpler.
  • the user plane data processing network element acquires the service type associated with the user plane data or the execution rule associated with the user plane data according to the received information of the feature set, thereby preventing the data analysis network element from being sent to the user.
  • the surface data processing network element sends a matching algorithm to prevent network congestion caused by frequent update of the matching algorithm.
  • FIG. 3 is a schematic flowchart of a data analysis method according to a second embodiment of the present application.
  • the data analysis network element is an NWDA network element
  • the user plane data processing network element is a UPF network element
  • the third party server is an OTT server
  • the feature set information is a feature index set
  • the feature parameter is a feature vector.
  • the NWDA network element sends the information of the feature set to the UPF network element by using a packet data (PDU) session establishment/modification process initiated by the UE.
  • PDU packet data
  • the same portions of the present embodiment as the second embodiment can be referred to the description of the second embodiment.
  • the data analysis method of this embodiment includes:
  • the NWDA network element acquires training data.
  • step 201 For the method for obtaining the training data by the NWDA network element, refer to step 201, which is not described here.
  • the NWDA network element analyzes the training data to obtain a set of feature indexes.
  • step 202 For details, refer to step 202, which is not described here.
  • the terminal device requests the SMF network element to initiate a PDU session establishment/modification process by using the AMF network element.
  • step refer to the process of the terminal device initiating a PDU session establishment/modification request in the prior art.
  • the SMF network element sends a request message to the PCF network element to request an execution rule related to the PDU session.
  • the execution rule related to the PDU session may be at least one of an execution policy, a control policy, a charging policy, and a policy and charging control.
  • step refer to the PDU connectivity access network (PDU-CAN) session establishment/modification procedure initiated by the SMF in the prior art.
  • PDU-CAN PDU connectivity access network
  • the request message includes a name of a data network (DNN) where the service requested by the terminal device is located, and/or a user plane function identifier (UPF ID) associated with the data network.
  • DNN data network
  • UPF ID user plane function identifier
  • the PCF network element sends a request message to the NWDA network element to request a set of feature indexes.
  • the set of the requested feature index is a set of feature indexes corresponding to the service type requested by the terminal device.
  • the request message includes a name of the data network where the service requested by the terminal device is located, and/or a user plane function identifier associated with the data network, that is, the DNN and/or the UPF ID received by the PCF network element in step 304.
  • the NWDA network element After receiving the request message sent by the PCF network element, the NWDA network element selects, from the set of feature indexes obtained in step 302, a set of feature indexes that are consistent with the received DNN and/or UPF ID.
  • the feature set corresponding to the set of feature indexes includes a DNN and/or a UPF ID.
  • the feature set corresponding to the set of feature indexes does not include the DNN and/or the UPF ID, but the DNN and/or the UPF ID are associated with the set of feature indexes.
  • Step 306 is an optional step. After receiving the request message sent by the PCF network element, the NWDA network element may directly perform step 307 without performing step 306.
  • the NWDA network element sends a set of feature indexes to the PCF network element.
  • the NWDA network element When the NWDA network element performs step 306, the NWDA sends the set of feature indexes selected by the NWDA network element in step 306. When the NWDA network element does not perform step 306, the NWDA network element sends a preset set of feature indexes.
  • the preset feature index set may be a set of all feature indexes acquired by the NWDA network element, or may be a set of feature indexes determined by the NWDA network element according to the current network condition.
  • the NWDA network element sends the set of feature indexes to the PCF network element, refer to the description of step 203.
  • the NWDA network element may further send the service type information corresponding to the set of the feature index to the PCF network element.
  • the NWDA network element may also send a feature associated with the set of feature indices to the PCF network element, such as the DNN in Table 2.
  • the NWDA network element sends a response message of the request message in step 305 to the PCF network element, where the response message includes the content sent by the NWDA network element to the PCF network element.
  • the PCF network element generates an execution rule according to the service type information.
  • the PCF network element generates an execution rule (for example, a quality of service policy) corresponding to the service type according to the received service type information (for example, a service type).
  • an execution rule for example, a quality of service policy
  • the PCF network element may further select an execution rule from the generated execution rule and the execution rule received from the NWDA network element.
  • the PCF network element sends a set of feature indexes to the SMF network element.
  • the set of feature indexes sent by the PCF network element to the SMF network element may be a set of feature indexes received by the PCF network element from the NWDA network element.
  • the set of feature indexes sent by the PCF network element to the SMF network element may also be a set of feature indexes corresponding to the service type requested by the terminal device.
  • the PCF network element may further send, to the SMF network element, at least one of a feature associated with the set of feature indexes, a service type information corresponding to the set of feature indexes, and an execution rule.
  • the PCF network element sends a response message of the request message in step 304 to the SMF network element, where the response message includes the content sent by the PCF network element to the SMF network element.
  • the response message can be referred to the PDU-CAN session establishment/modification procedure in the prior art.
  • the SMF network element sends a set of feature indexes to the UPF network element.
  • the set of feature indexes sent by the SMF network element to the UPF network element may be a set of feature indexes received by the SMF network element.
  • the SMF network element may further send, to the UPF network element, at least one of a feature associated with the set of feature indexes, a service type information corresponding to the set of feature indexes, and an execution rule.
  • the SMF network element in a process of session establishment/modification initiated by the SMF network element to the UPF network element, the SMF network element sends a set of feature indexes to the UPF network element.
  • Step 309 and step 310 can be referred to step 204, which will not be described in detail herein.
  • the UPF network element receives the user plane data, and obtains a feature vector of the set of the user plane data corresponding to the feature index according to the received feature index set.
  • step 205 For details, refer to step 205 and step 206, which are not described here.
  • the UPF network element sends a feature vector to the SMF network element.
  • the feature vector sent by the UPF network element to the SMF network element may be the feature vector obtained in step 311.
  • the UPF network element may further send the set number of the feature index corresponding to the feature vector to the SMF network element.
  • the UPF network element in a flow of a session establishment/modification response sent by the UPF network element to the SMF network element, the UPF network element sends the feature vector to the SMF network element.
  • the SMF network element sends the received feature vector to the NWDA network element via the PCF network element.
  • the SMF network element may further send the set number of the feature index corresponding to the feature vector to the NWDA network element.
  • step 207 This step can be seen in step 207, which is not described here.
  • the NWDA network element determines the service type information associated with the user plane data according to the received feature vector and the corresponding matching algorithm. For details, refer to the description in step 208.
  • the NWDA network element sends the response result of the feature vector to the UPF network element by using the PCF network element and the SMF network element.
  • the response result includes service type information and/or execution rule information that may include user plane data.
  • the response result may include a result that the feature vector conforms to the corresponding matching algorithm, and the response result may also include whether all the feature vectors conform to the result of the corresponding matching algorithm.
  • the response result also includes a set number of the feature index corresponding to the feature vector.
  • the UPF network element processes the corresponding user plane data according to the response result.
  • FIG. 4 is a schematic flowchart of a data analysis method according to a third embodiment of the present application.
  • the data analysis method of this embodiment includes:
  • the NWDA network element sends a set of feature indexes to the PCF network element.
  • the set of feature indices is a set of feature indices obtained by the NWDA network element based on the training data in step 402.
  • the PCF network element After receiving the request message sent by the SMF, the PCF network element selects, from the set of feature indexes received in step 403, a set of feature indexes that are consistent with the DNN and/or UPF ID of the current PDU session sent by the SMF network element.
  • a method of selecting a feature index of a PCF network element refer to the method for selecting a feature index of the NWDA network element in step 306.
  • step 407 can be directly performed.
  • the PCF network element generates an execution rule according to the service type information.
  • the PCF network element sends a set of feature indexes to the SMF network element.
  • the PCF network element when the PCF network element performs step 406, the PCF network element sends the set of feature indexes selected by the PCF network element in step 406 to the SMF network element.
  • the PCF network element does not perform step 406, the PCF network element sends a set of feature indexes preset by the PCF network element to the SMF network element.
  • the preset feature index set may be a set of all feature indexes received by the PCF network element from the NWDA network element, or may be a set of feature indexes determined by the PCF network element according to the current network condition.
  • the set of feature indexes sent by the PCF network element to the SMF network element may also be a set of feature indexes corresponding to the service type requested by the terminal device.
  • the PCF network element may further send, to the SMF network element, at least one of a feature associated with the set of feature indexes, a service type information corresponding to the set of feature indexes, and an execution rule.
  • the PCF network element sends a response message of the request message in step 405 to the SMF network element, where the response message includes the content sent by the PCF network element to the SMF network element.
  • FIG. 5 is a schematic flowchart of a data analysis method according to a fourth embodiment of the present application.
  • the data analysis method of this embodiment includes:
  • the NWDA network element sends a response result of the feature vector to the PCF network element.
  • the response result includes service type information that can include user plane data.
  • the response result may include a result that the feature vector conforms to the corresponding matching algorithm, and the response result may also include whether all the feature vectors conform to the result of the corresponding matching algorithm.
  • the response result also includes a set number of the feature index corresponding to the feature vector.
  • the PCF network element generates an execution rule according to the response result.
  • the PCF network element generates an execution rule (for example, a quality of service policy) corresponding to the service type according to the received service type information (for example, a service type).
  • an execution rule for example, a quality of service policy
  • the PCF network element may further select an execution rule from the generated execution rule and the execution rule received from the NWDA network element.
  • the PCF network element sends a response result to the UPF network element via the SMF network element.
  • the response result includes service type information and/or execution rules of the user plane data.
  • the UPF network element processes the corresponding user plane data according to the response result.
  • step 316 See the description of step 316 for details.
  • FIG. 6 is a schematic flowchart of a data analysis method according to a fifth embodiment of the present application.
  • the data analysis method of this embodiment includes:
  • 607-612 Refer to the description of steps 408-413 for details.
  • the NWDA network element sends a response result of the feature vector to the PCF network element.
  • the response result includes service type information that can include user plane data.
  • the response result may include a result that the feature vector conforms to the corresponding matching algorithm, and the response result may also include whether all the feature vectors conform to the result of the corresponding matching algorithm.
  • the response result also includes a set number of the feature index corresponding to the feature vector.
  • the PCF network element generates an execution rule according to the response result.
  • the PCF network element generates an execution rule (for example, a quality of service policy) corresponding to the service type according to the received service type information (for example, a service type).
  • an execution rule for example, a quality of service policy
  • the PCF network element may further select an execution rule from the generated execution rule and the execution rule received from the NWDA network element.
  • the PCF network element sends a response result to the UPF network element by using the SMF network element.
  • the response result includes service type information and/or execution rules of the user plane data.
  • the UPF network element processes the corresponding user plane data according to the response result.
  • a feature extraction unit may be disposed in the user plane data processing network element and the data analysis network element.
  • the feature extraction unit in the data analysis network element acquires the information of the feature set of the service type according to the training data of a certain service type, and further acquires the feature parameter corresponding to the training data (for example, the feature vector).
  • the feature extraction unit in the user plane data processing network element acquires the information of the user plane data and the feature set sent by the household data processing network element, and acquires the feature parameter corresponding to the household data according to the acquired information of the feature set (eg, the feature vector) .
  • the information of the feature set may be information of at least part of the feature set of the information of the at least one feature set received by the user plane data processing network element.
  • the feature extraction unit transmits the obtained feature parameters to the user plane data processing network element.
  • the feature extraction unit in the user plane data processing network element may further obtain information related to the user plane data, for example, a cell identifier of the terminal device.
  • the feature extraction unit may also acquire feature parameters corresponding to the information associated with the user plane data.
  • only one feature extraction unit may be provided.
  • the feature extraction unit may be a single network element, and the feature extraction unit may also be disposed in the user plane data processing network element or other network element, which is not limited herein.
  • FIG. 7 is a schematic flowchart of a data analysis method according to a sixth embodiment of the present application. In the flowchart of this embodiment, only the user plane data processing network element, the feature extraction network element, and the data analysis network element are shown. For the process interaction of the user plane data processing network element, the feature extraction network element, the data analysis network element and other network elements, refer to the embodiment of FIG. 2 to FIG. 6, which will not be described here.
  • the data analysis network element obtains training data.
  • step 201 See the description of step 201 for details.
  • the data analysis network element sends the training data to the feature extraction unit.
  • the training data sent by the data analysis network element to the feature extraction unit may be training data corresponding to multiple service types, or may be training data of a service type.
  • the data analysis network element may actively send the training data to the feature extraction unit, or may send the training data according to the request of the feature extraction unit.
  • the feature extraction unit and the data analysis network element may exchange data through the policy control network element and the session management network element.
  • the feature extraction unit analyzes the training data, and obtains a set of feature indexes corresponding to a certain service and a feature vector of the set of the feature index corresponding to the training data of the service type.
  • step 202 See step 202 for details.
  • the feature extraction unit sends the feature vector in step 703 to the data analysis network element.
  • the feature extraction unit may further send the set of feature indexes in the 703 to the data analysis network element.
  • the feature extraction unit may number the set of feature indexes and then send the number to the data analysis network element. In another possible implementation manner, the feature extraction unit may also send the service type corresponding to the set of feature indexes to the data analysis network element.
  • the data analysis network element obtains a matching algorithm corresponding to the service according to a feature vector of a certain service.
  • the data analysis network element can obtain the matching algorithm by using the big data analysis method in the prior art. For details related to the matching algorithm, refer to step 202.
  • the data analysis network element sends the received feature index set to the user plane data processing network element.
  • the set of feature indexes obtained by the feature extraction unit may also be sent to the user plane data processing network element in other manners.
  • the feature extraction unit directly sends the obtained feature index set to the user plane data processing network element.
  • the user plane data processing network element After receiving the user plane data, the user plane data processing network element sends the user plane data to the feature extraction unit.
  • the user plane data processing network element may also send the set of feature indexes corresponding to the user plane data to the feature extraction unit.
  • the feature extraction unit acquires a feature vector of the user plane data according to the received set of user plane data and the feature index.
  • the feature extraction unit sends the obtained feature vector corresponding to the user plane data to the user plane data processing network element.
  • the feature extraction unit may acquire the feature vector of the set of the feature index corresponding to the user plane data.
  • the set of preset feature indexes may be a set of all feature indexes in the feature extraction unit.
  • the user plane data processing network element sends the feature vector corresponding to the user plane data to the data analysis network element.
  • step 207 See the description of step 207 for details.
  • the data analysis network element sends the service type information associated with the user plane data according to the received feature vector.
  • step 208 See the description of step 208 for details.
  • the data analysis network element sends a response result of the feature vector to the user plane data processing network element.
  • the user plane data processing network element processes the user plane data according to the response result.
  • step 210 See the description of step 210 and step 211 for details.
  • the feature extraction unit is deployed in multiple network elements, if you need to maintain the feature project (for example, version upgrade or signature database update), you need to maintain all the NEs that have the feature extraction unit deployed.
  • the method is more complicated to maintain.
  • only one feature lifting unit is provided. When maintenance of the feature engineering is required, only the network element in which the feature extraction unit is deployed is maintained, so that the maintenance is simple and convenient.
  • FIG. 8 is a schematic diagram of a data analysis apparatus according to an embodiment of the present application.
  • the data analysis device includes a transceiver unit 801, a processing unit 802, and a storage unit 803.
  • the transceiver unit 801, the processing unit 802, and the storage unit 803 may be physically separated from each other, or may be integrated into one or more physical units, which is not limited herein.
  • the transceiver unit 801 is configured to implement content interaction between the processing unit 802 and other units or network elements.
  • the transceiver unit 801 can be a communication interface of the data analysis device, a transceiver circuit or a transceiver, or a transceiver.
  • the transceiver unit 801 can also be a communication interface or a transceiver circuit of the processing unit 802.
  • the transceiver unit 801 can be a transceiver chip.
  • the data analysis apparatus may also include a plurality of transceiver units 801 or the transceiver unit 801 includes a plurality of sub-transceiver units.
  • the transceiver unit 801 may further include a transmitting unit and a receiving unit.
  • the processing unit 802 is configured to implement processing of data by the data analysis device.
  • Processing unit 802 can be a processing circuit or a processor.
  • the processor may be a central processing unit (CPU), a network processor (NP) or a combination of a CPU and an NP.
  • the processor may further include a hardware chip.
  • the hardware chip may be an application-specific integrated circuit (ASIC), a programmable logic device (PLD), or a combination thereof.
  • the PLD may be a complex programmable logic device (CPLD), a field-programmable gate array (FPGA), a Generic Array Logic (GAL), or any combination thereof.
  • CPLD complex programmable logic device
  • FPGA field-programmable gate array
  • GAL Generic Array Logic
  • the data analysis device may also include a plurality of processing units or the processing unit 802 includes a plurality of sub-data processing units.
  • the processor may be a single-CPU processor or a multi-core processor.
  • the storage unit 803 is configured to store computer instructions executed by the processing unit 802.
  • the storage unit 803 may be a storage circuit or a memory.
  • the memory can be either volatile memory or non-volatile memory, or can include both volatile and nonvolatile memory.
  • the non-volatile memory may be a read-only memory (ROM), a programmable read only memory (ROMM), an erasable programmable read only memory (erasable PROM, EPROM), or an electrical Erase programmable EPROM (EEPROM) or flash memory.
  • the volatile memory can be a random access memory (RAM) that acts as an external cache.
  • the storage unit 803 may be a unit that is independent of the processing unit 802, or may be a storage unit in the processing unit 802, which is not limited herein. Although only one storage unit 803 is shown in FIG. 8, the data analysis device may include a plurality of storage units 803 or the storage unit 803 includes a plurality of sub storage units.
  • the processing unit 802 can perform content interaction with other network elements through the transceiver unit 801. For example, the processing unit 802 acquires or receives content from other network elements. If the processing unit 802 and the transceiver unit 801 are physically separate components, the processing unit 802 can perform content interaction with other units within the data analysis device without going through the transceiver unit 801.
  • the transceiver unit 801, the processing unit 802, and the storage unit 803 can be connected to each other through a bus.
  • the bus can be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus.
  • PCI peripheral component interconnect
  • EISA extended industry standard architecture
  • the bus can be divided into an address bus, a data bus, a control bus, and the like.
  • the processing unit 802 causes the data analysis device to implement the methods in the first to sixth embodiments of the present application based on the computer instructions stored in the storage unit 803.
  • the data analysis device may be a user plane data processing network element, for example, a UPF network element, a base station.
  • the data analysis device may also be a policy control network element, such as a PCF network element.
  • the data analysis device may also be a data analysis network element, such as an NWDA network element.
  • the transceiver unit 801 is configured to acquire information from at least one feature set of the data analysis network element and to receive user plane data, where the information of the at least one feature set is The information of each feature set corresponds to at least one service type or at least one execution rule.
  • the processing unit 802 is configured to acquire feature parameters of the user plane data according to the information of the at least one feature set.
  • the transceiver unit 801 is further configured to send the feature parameter to the data analysis network element and obtain a response result of the feature parameter from the data analysis network element.
  • the processing unit 802 is further configured to acquire, according to the response result, a service type associated with the user plane data or an execution rule of the user plane data association.
  • the processing unit 802 is specifically configured to acquire the execution rule of the user plane data association from the policy control network element according to the response result, or specifically, obtain the data analysis from the response result according to the response result. Execution rules associated with the user plane data of the network element.
  • the processing unit 802 is further configured to process the user plane data according to a service type associated with the user plane data or an execution rule associated with the user plane data.
  • the processing unit 802 is specifically configured to forward the user plane data according to the service priority information indicated in the execution rule associated with the user plane data; or the processing unit 802 is specifically configured to use the user plane data according to the user plane data.
  • the associated service type or the execution rule associated with the user plane data adds the label information of the service type to the user plane data; or the processing unit 802 is specifically configured to use the service type associated with the user plane data or the The execution rule associated with the user plane data adds scheduling priority information to the user plane data; or the processing unit 802 is specifically configured to perform, according to the service type associated with the user plane data or the execution rule associated with the user plane data.
  • the user plane data performs charging statistics.
  • the processing unit 802 is specifically configured to determine that the destination address of the user plane data is an address of the terminal device, and the terminal device is in an idle state, and perform the association according to the service type associated with the user plane data or the user plane data association.
  • the rule sends the paging priority information of the terminal device to the session management network element.
  • the processing unit 802 is specifically configured to: select, according to a partial feature, information of a partial feature set from information of the at least one feature set; acquire feature parameters of the user plane data, the feature The parameter corresponds to information of the partial feature set.
  • the partial feature includes Internet Protocol IP quintuple information including the user plane data.
  • the transceiver unit 801 is further configured to acquire information related to the user plane data from other user plane data processing network elements or control plane network elements according to the information of the at least one feature set;
  • the processing unit 802 is specifically configured to acquire feature parameters of the user plane data according to the information of the at least one feature set and the associated information.
  • the information of the feature set is a set of feature indexes
  • the feature parameter is a feature vector
  • the transceiver unit 801 is further configured to implement the transceiving operation of the content of the user plane data processing network element and the external network element in the first embodiment to the sixth embodiment of the present application.
  • the processing unit 802 is further configured to implement the processing operation of the internal data or signaling of the user plane data processing network element in the first embodiment to the sixth embodiment of the present application.
  • the processing unit 802 is configured to implement step 311 in the second embodiment. Processing operation in step 316.
  • the processing unit 802 causes the user plane data processing network element to implement the operations performed by the user plane data processing network element in the first to sixth embodiments of the present application, according to the computer instructions stored in the storage unit 803.
  • the processing unit 802 uses the transceiver unit 801 to acquire information from at least one feature set of the data analysis network element, where each feature set of the at least one feature set is The information is corresponding to the at least one service type or the at least one execution rule; the processing unit 802 acquires the feature parameter of the user plane data according to the information of the at least one feature set; the processing unit 802 analyzes the network element by using the transceiver unit 801 Sending the feature parameter; the processing unit 802 uses the transceiver unit 801 to obtain the response result of the feature parameter from the data analysis network element; the processing unit 802 acquires the service type or the association associated with the user plane data according to the response result.
  • the execution rules associated with the user plane data is associated with the user plane data.
  • the processing unit 802 uses the transceiver unit 80 to acquire the execution rule of the user plane data association from the policy control network element according to the response result; or, the processing unit 802 uses the transceiver unit 801 according to the transceiver unit 801.
  • the response result obtains an execution rule of the user plane data association from the data analysis network element.
  • the processing unit 802 processes the user plane data according to the service type associated with the user plane data or the execution rule associated with the user plane data.
  • the processing unit 802 forwards the user plane data according to the service priority information indicated in the execution rule associated with the user plane data; or the processing unit 802 associates the service according to the user plane data.
  • An execution rule associated with the type or the user plane data is used to add label information of the service type to the user plane data; or the execution unit associated with the service type associated with the user plane data or the user plane data is The user plane data is added with scheduling priority information; or the processing unit 802 performs charging statistics on the user plane data according to the service type associated with the user plane data or the execution rule associated with the user plane data.
  • the processing unit 802 determines that the destination address of the user plane data is an address of the terminal device and the terminal device is in an idle state; and the processing unit 802 associates the service type or the location according to the user plane data.
  • the execution rule of the user plane data association uses the transceiver unit 801 to send the paging priority information of the terminal device to the session management network element.
  • the processing unit 802 selects information of a partial feature set from the information of the at least one feature set according to the partial feature; the processing unit 802 acquires a feature parameter of the user plane data, where the feature parameter corresponds to Information about the partial feature set.
  • the processing unit 802 acquires the information associated with the user plane data from other user plane data processing network elements or control plane network elements according to the information of the at least one feature set; the processing unit 802 is configured according to the The information of the at least one feature set and the associated information acquires feature parameters of the user plane data.
  • the transceiver unit 801 is configured to send information of the at least one feature set to the user plane data processing network element and receive the feature parameter of the user plane data from the user plane data processing network element, where And the information of each feature set of the information of the at least one feature set corresponds to at least one service type or at least one execution rule, where the information of the at least one feature set includes information of a feature set corresponding to the feature parameter;
  • the processing unit 802 is configured to determine, according to the feature parameter, service type information associated with the user plane data or execution rule information associated with the user plane data.
  • the processing unit 802 is further configured to select, according to the partial feature, the information of the at least one feature set from the information of the feature set acquired in advance.
  • the part of the feature includes the data network name and/or the identification information of the user plane function network element corresponding to the data network.
  • the processing unit 802 is specifically configured to select, from the information of the pre-acquired feature set, information of a feature set that is consistent with the data network name and/or the identifier information of the user plane function network element corresponding to the data network.
  • the processing unit 802 is specifically configured to determine, according to the feature parameter and a matching algorithm corresponding to the feature parameter, service type information associated with the user plane data or associated with the user plane data. Execute rule information.
  • the service type information associated with the user plane data includes indication information of a service type or a service type associated with the user plane data.
  • the transceiver unit 801 is further configured to send a response result of the feature parameter to the user plane data processing network element, where the response result includes service type information associated with the user plane data. Or execution rule information associated with the user plane data.
  • the processing unit 802 is further configured to acquire information about the at least one feature set according to the training data; or the transceiver unit 801 is further configured to receive the at least the feature extraction network element. Information about a feature set.
  • the information of the feature set is a set of feature indexes
  • the feature parameter is a feature vector
  • the transceiver unit 801 is further configured to implement the transceiving operation of the content of the data analysis network element and the external network element in the first embodiment to the sixth embodiment of the present application.
  • the processing unit 802 is further configured to implement the processing operation of the internal data or signaling of the data analysis network element in the first to sixth embodiments of the present application.
  • the processing unit 802 is used to implement step 302 and step 306 in the second embodiment. And the processing operation in step 314.
  • the processing unit 802 causes the data analysis network element to implement the operations performed by the data analysis network element in the first to sixth embodiments of the present application according to the computer instructions stored in the storage unit 803.
  • the processing unit 802 uses the transceiver unit 801 to send information of at least one feature set to the user plane data processing network element, where each feature set of the at least one feature set information The information corresponds to at least one service type or at least one execution rule; the processing unit 802 uses the transceiver unit 801 to receive feature parameters of the user plane data from the user plane data processing network element, wherein the information of the at least one feature set The information of the feature set corresponding to the feature parameter is included; the processing unit 802 determines the service type information associated with the user plane data or the execution rule information associated with the user plane data according to the feature parameter.
  • the processing unit 802 selects information of the at least one feature set from information of the feature set acquired in advance according to the partial feature.
  • the part of the feature includes the data network name and/or the identification information of the user plane function network element corresponding to the data network; the processing unit 802 selects the information from the pre-acquired feature set information. And the information about the data set name and/or the feature set of the user plane function network element corresponding to the data network.
  • the processing unit 802 determines the service type information associated with the user plane data or the execution rule information associated with the user plane data according to the feature parameter and a matching algorithm corresponding to the feature parameter.
  • the processing unit 802 sends, by using the transceiver unit 801, a response result of the feature parameter to the user plane data processing network element, where the response result includes service type information associated with the user plane data or Execution rule information associated with the user plane data.
  • the processing unit 802 is configured to acquire, by the transceiver unit 801, information from at least one feature set of the data analysis network element, where the information of the at least one feature set is The information of each feature set corresponds to at least one service type or at least one execution rule; the processing unit 802 is further configured to send, by using the transceiver unit 801, the information of the at least one feature set to the user plane data processing network element. Information on at least part of the feature set.
  • the processing unit 802 is further configured to select information of the at least part of the feature set from the information of the at least one feature set according to the partial feature.
  • the part of the feature includes the data network name and/or the identification information of the user plane function network element corresponding to the data network.
  • the processing unit is configured to select, from the information of the at least one feature set, information of a feature set that is consistent with the data network name and/or the identifier information of the user plane function network element corresponding to the data network.
  • the transceiver unit 801 is further configured to send the service type information corresponding to the information of the at least part of the feature set to the user plane data processing network element.
  • the transceiver unit 801 is further configured to send, to the user plane data processing network element, an execution rule corresponding to the information of the at least part of the feature set.
  • the transceiver unit 801 is further configured to acquire service type information corresponding to the information of the at least part of the feature set of the data analysis network element, where the processing unit 802 is further configured to perform The obtained service type information generates an execution rule corresponding to the information of the at least part of the feature set; the transceiver unit 801 is further configured to send, to the user plane data processing network element, an execution rule corresponding to the information of the at least part of the feature set .
  • the information of the feature set is a set of feature indexes
  • the feature parameter is a feature vector
  • the transceiver unit 801 is further configured to implement the sending and receiving operations of the content of the policy control network element and the external network element in the first to sixth embodiments of the present application.
  • the processing unit 802 is further configured to implement the processing operation of the internal data or signaling of the policy control network element in the first to sixth embodiments of the present application.
  • the processing unit 802 is configured to implement the processing operation of step 308 in the second embodiment. .
  • the processing unit 802 causes the policy control network element to implement the operations performed by the policy control network element in the first to sixth embodiments of the present application, according to the computer instructions stored in the storage unit 803.
  • the processing unit 802 uses the transceiver unit 801 to acquire information from at least one feature set of the data analysis network element, where each feature set of the at least one feature set is The information corresponds to at least one service type or at least one execution rule; the processing unit 802 uses the transceiver unit 801 to transmit information of at least part of the feature set of the at least one feature set to the user plane data processing network element.
  • the processing unit 802 selects information of the at least part of the feature set from the information of the at least one feature set according to the partial feature.
  • the part of the feature includes the data network name and/or the identification information of the user plane function network element corresponding to the data network; the processing unit 802 selects the information from the information of the at least one feature set.
  • the data network name and/or the information of the feature set that is consistent with the identification information of the user plane function network element corresponding to the data network.
  • the processing unit 802 sends, by using the transceiver unit 801, the service type information corresponding to the information of the at least part of the feature set to the user plane data processing network element.
  • the processing unit 802 uses the transceiver unit 801 to send an execution rule corresponding to the information of the at least part of the feature set to the user plane data processing network element.
  • the processing unit 802 uses the transceiver unit 801 to acquire service type information corresponding to the information of the at least part of the feature set of the data analysis network element; and the processing unit 802 is configured according to the acquired service type information. Generating an execution rule corresponding to the information of the at least part of the feature set; the processing unit 802 uses the transceiver unit 801 to send an execution rule corresponding to the information of the at least part of the feature set to the user plane data processing network element.
  • the names of request messages, response messages, and other various messages are employed for convenience of description. However, these messages are merely illustrative of the content to be carried or the functions to be carried.
  • the specific name of the message is not limited to the present application, for example, it may be a first message, a second message, a third message, or the like. These messages can be specific messages and can be some of the fields in the message. These messages can also represent various service operations.
  • the computer program product can include one or more computer instructions.
  • the computer can be a general purpose computer, a special purpose computer, a computer network, or other programmable device.
  • the computer instructions can be stored in a computer readable storage medium or transferred from one computer readable storage medium to another computer readable storage medium, for example, the computer instructions can be from a website site, computer, server or data center Transfer to another website site, computer, server, or data center by wire (eg, coaxial cable, fiber optic, digital subscriber (DSL), or wireless (eg, infrared, wireless, microwave, etc.).
  • the computer readable storage medium can be any available media that can be accessed by a computer or a data storage device such as a server, data center, or the like that includes one or more available media.
  • the usable medium may be a magnetic medium (eg, a floppy disk, a hard disk, a magnetic disk), an optical medium (eg, a DVD), or a semiconductor medium (eg, a solid state disk (SSD)) or the like.
  • a magnetic medium eg, a floppy disk, a hard disk, a magnetic disk
  • an optical medium eg, a DVD
  • a semiconductor medium eg, a solid state disk (SSD)
  • the disclosed systems, devices, and methods may be implemented in other manners.
  • the device embodiments described above are merely illustrative.
  • the division of the unit is only a logical function division.
  • there may be another division manner for example, multiple units or components may be combined or Can be integrated into another system, or some features can be ignored or not executed.
  • the mutual coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interface, device or unit, and may be in an electrical, mechanical or other form.
  • the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, may be located in one place, or may be distributed to multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of the embodiment.
  • 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.
  • the functions may be stored in a computer readable storage medium if implemented in the form of a software functional unit and sold or used as a standalone product.
  • the technical solution of the present application which is essential or contributes to the prior art, or a part of the technical solution, may be embodied in the form of a software product, which is stored in a storage medium, including
  • the instructions are used to cause a computer device (which may be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods described in various embodiments of the present application.
  • the foregoing storage medium includes various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read only memory, a random access memory, a magnetic disk, or an optical disk.

Landscapes

  • Engineering & Computer Science (AREA)
  • Signal Processing (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Data Mining & Analysis (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Artificial Intelligence (AREA)
  • Evolutionary Biology (AREA)
  • Evolutionary Computation (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)
  • Mobile Radio Communication Systems (AREA)

Abstract

本申请实施例提供一种数据分析方法和数据分析装置。该方法，包括：用户面数据处理网元获取来自数据分析网元的至少一个特征集合的信息，其中，所述至少一个特征集合的信息中的每个特征集合的信息对应至少一种业务类型或至少一种执行规则；所述用户面数据处理网元根据所述至少一个特征集合的信息获取用户面数据的特征参数；所述用户面数据处理网元向所述数据分析网元发送所述特征参数；所述用户面数据处理网元获取来自所述数据分析网元的所述特征参数的响应结果；所述用户面数据处理网元根据所述响应结果获取所述用户面数据关联的业务类型或所述用户面数据关联的执行规则。本申请实施例实现了在通信网络中利用数据分析网元对数据进行分析。

Description

数据分析方法和装置
本申请要求在2017年09月30日提交中华人民共和国知识产权局、申请号为201710915784.4、发明名称为“数据分析方法和装置”的中国专利申请的优先权，其全部内容通过引用结合在本申请中
技术领域
本申请实施例涉及通信领域，更具体地，涉及一种获取特征参数的方法和装置。
背景技术
在第五代(5rd generation，5G)通信网络中，引入了网络数据分析(network data analytics,NWDA)网元。NWDA网元采用大数据分析方法进行模型训练，并使用训练出的模型对数据进行分析。然而，现有技术并没有提供在通信网络中如何利用NWDA网元进行数据分析的具体方法。
发明内容
本申请实施例提供一种数据分析方法和数据分析装置，能够实现在通信网络中利用数据分析网元进行数据分析。
本申请的第一方面，提供了一种数据分析方法，包括：用户面数据处理网元获取来自数据分析网元的至少一个特征集合的信息，其中，所述至少一个特征集合的信息中的每个特征集合的信息对应至少一种业务类型或至少一种执行规则；所述用户面数据处理网元接收用户面数据；所述用户面数据处理网元根据所述至少一个特征集合的信息获取所述用户面数据的特征参数；所述用户面数据处理网元向所述数据分析网元发送所述特征参数；所述用户面数据处理网元获取来自所述数据分析网元的所述特征参数的响应结果；所述用户面数据处理网元根据所述响应结果获取所述用户面数据关联的业务类型或所述用户面数据关联的执行规则。本申请实施例实现了在通信网络中利用数据分析网元对数据进行分析。
在第一方面的第一种可能的实现方式中，所述用户面数据处理网元根据所述响应结果获取所述用户面数据关联的业务类型或所述用户面数据关联的执行规则，包括：所述用户面数据处理网元根据所述响应结果获取来自策略控制网元的所述用户面数据关联的所述执行规则；或者，所述用户面数据处理网元根据所述响应结果获取来自所述数据分析网元的所述用户面数据关联的执行规则。
在第一方面的第二种可能的实现方式中，所述方法还包括：所述用户面数据处理网元根据所述用户面数据关联的业务类型或所述用户面数据关联的执行规则处理所述用户面数据。
结合第一方面的第二种可能的实现方式，在第一方面的第三种可能的实现方式中，所述用户面数据处理网元根据所述用户面数据关联的业务类型或所述用户面数据关联的执行规则处理所述用户面数据，包括：所述用户面数据处理网元根据所述用户面数据关联的执行规则中指示的业务优先级信息转发所述用户面数据；或者，所述用户面数据处理网元 根据所述用户面数据关联的业务类型或所述用户面数据关联的执行规则为所述用户面数据增加业务类型的标签信息；或者，所述用户面数据处理网元根据所述用户面数据关联的业务类型或所述用户面数据关联的执行规则为所述用户面数据增加调度优先级信息；或者，所述用户面数据处理网元根据所述用户面数据关联的业务类型或所述用户面数据关联的执行规则对所述用户面数据进行计费统计。
结合第一方面的第二种可能的实现方式，在第一方面的第四种可能的实现方式中，所述用户面数据处理网元根据所述用户面数据关联的业务类型或所述用户面数据关联的执行规则处理所述用户面数据，包括：所述用户面数据处理网元确定所述用户面数据的目的地址为终端设备的地址且所述终端设备处于空闲状态；所述用户面数据处理网元根据所述用户面数据关联的业务类型或所述用户面数据关联的执行规则向会话管理网元发送所述终端设备的寻呼优先级信息。
结合第一方面的任一种可能的实现方式，所述响应结果包括所述用户面数据关联的业务类型信息和/或所述用户面数据关联的执行规则信息。
结合第一方面的任一种可能的实现方式，在第一方面的第五种可能的实现方式中，所述用户面数据处理网元根据所述至少一个特征集合的信息获取所述用户面数据的特征参数，包括：所述用户面数据处理网元根据部分特征从所述至少一个特征集合的信息中选择部分特征集合的信息；所述用户面数据处理网元获取所述用户面数据的特征参数，所述特征参数对应所述部分特征集合的信息。
结合第一方面的第五种可能的实现方式，在第一方面的第六种可能的实现方式中，所述部分特征包括包括所述用户面数据的互联网协议IP五元组信息。
结合第一方面的任一种可能的实现方式，所述方法还包括：所述用户面数据处理网元根据所述至少一个特征集合的信息从其他用户面数据处理网元或控制面网元获取所述用户面数据关联的信息；所述用户面数据处理网元根据所述至少一个特征集合的信息获取所述用户面数据的特征参数包括：所述用户面数据处理网元根据所述至少一个特征集合的信息和所述关联的信息获取所述用户面数据的特征参数。
在第一方面的第七种可能的实现方式中，所述用户面数据处理网元包括特征提取单元，所述用户面数据处理网元根据所述至少一个特征集合的信息获取所述用户面数据的特征参数，包括：所述特征提取单元根据所述至少一个特征集合的信息中的至少部分特征集合的信息获取所述用户面数据的特征参数。
本申请的第二方面，提供了一种数据分析方法，包括：数据分析网元向用户面数据处理网元发送至少一个特征集合的信息，其中，所述至少一个特征集合的信息中的每个特征集合的信息对应至少一种业务类型或者至少一种执行规则；所述数据分析网元接收来自所述用户面数据处理网元的用户面数据的特征参数，其中，所述至少一个特征集合的信息包括所述特征参数对应的特征集合的信息；所述数据分析网元根据所述特征参数确定所述用户面数据关联的业务类型信息或者所述用户面数据关联的执行规则信息。
在第二方面的第一种可能的实现方式中，所述方法还包括：所述数据分析网元根据部分特征从预先获取的特征集合的信息中选择所述至少一个特征集合的信息。
结合第二方面的第一种可能的实现方式，在第一方面的第二种可能的实现方式中，所 述部分特征包括数据网络名称和/或数据网络对应的用户面功能网元的标识信息；所述数据分析网元根据所述部分特征从预先获取的特征集合的信息中选择所述至少一个特征集合的信息，包括：所述数据分析网元从所述预先获取的特征集合的信息中选择与所述数据网络名称和/或所述数据网络对应的用户面功能网元的标识信息一致的特征集合的信息。
结合第二方面的任一种可能的实现方式，在第二方面的第三种可能的实现方式中，所述数据分析网元根据所述特征参数确定所述用户面数据关联的业务类型信息或者所述用户面数据关联的执行规则信息，包括：所述数据分析网元根据所述特征参数及所述特征参数对应的匹配算法确定所述用户面数据关联的业务类型信息或者所述用户面数据关联的执行规则信息。
结合第二方面的第三种可能的实现方式，在第二方面的第四种可能的实现方式中，所述用户面数据关联的业务类型信息包括所述用户面数据关联的业务类型或业务类型的指示信息。
结合第二方面的任一种可能的实现方式，所述方法还包括：所述数据分析网元向所述用户面数据处理网元发送所述特征参数的响应结果，所述响应结果包括所述用户面数据关联的业务类型信息或者所述用户面数据关联的执行规则信息。
本申请的第三方面，提供了一种数据分析方法，包括：策略控制网元获取来自数据分析网元的至少一个特征集合的信息，其中，所述至少一个特征集合的信息中的每个特征集合的信息对应至少一种业务类型或至少一种执行规则；所述策略控制网元向用户面数据处理网元发送所述至少一个特征集合的信息中的至少部分特征集合的信息。
在第三方面的第一种可能的实现方式中，所述方法还包括：所述策略控制网元根据部分特征从所述至少一个特征集合的信息中选择所述至少部分特征集合的信息。
结合第三方面的第一种可能的实现方式，在第三方面的第二种可能的实现方式中，所述部分特征包括数据网络名称和/或数据网络对应的用户面功能网元的标识信息；所述策略控制网元根据所述部分特征从所述至少一个特征集合的信息中选择所述至少部分特征集合的信息，包括：所述策略控制网元从所述至少一个特征集合的信息中选择与所述数据网络名称和/或所述数据网络对应的用户面功能网元的标识信息一致的特征集合的信息。
结合第三方面的任一种可能的实现方式，所述方法还包括：所述策略控制网元向所述用户面数据处理网元发送所述至少部分特征集合的信息对应的业务类型信息。
结合第三方面的任一种可能的实现方式，所述方法还包括：所述策略控制网元向所述用户面数据处理网元发送所述至少部分特征集合的信息对应的执行规则。
结合第三方面的任一种可能的实现方式，所述方法还包括：所述策略控制网元获取来自所述数据分析网元的所述至少部分特征集合的信息对应的业务类型信息；所述策略控制网元根据所述获取的业务类型信息生成所述至少部分特征集合的信息对应的执行规则；所述策略控制网元向所述用户面数据处理网元发送所述至少部分特征集合的信息对应的执行规则。
本申请的第四方面，提供了一种数据分析方法，包括：用户面数据处理网元获取来自数据分析网元的特征集合的信息，其中，所述特征集合的信息对应一种业务类型或一种执行规则；所述用户面数据处理网元接收用户面数据；所述用户面数据处理网元根据所述特 征集合的信息获取所述用户面数据的特征参数；所述用户面数据处理网元向所述数据分析网元发送所述特征参数；所述用户面数据处理网元获取来自所述数据分析网元的所述特征参数的响应结果；所述用户面数据处理网元根据所述响应结果获取所述用户面数据关联的业务类型或所述用户面数据关联的执行规则。
本申请的第四方面还可以包括第一方面中的第一种到第四种可能的实现方式。
本申请的第五方面，提供了一种数据分析方法，包括：数据分析网元向用户面数据处理网元发送特征集合的信息，其中，所述特征集合的信息对应一种业务类型或者一种执行规则；所述数据分析网元接收来自所述用户面数据处理网元的用户面数据的特征参数，其中，所述特征参数对应所述特征集合的信息；所述数据分析网元根据所述特征参数确定所述用户面数据关联的业务类型信息或者所述用户面数据关联的执行规则信息。
本申请的第六方面，提供了一种数据分析方法，包括：策略控制网元获取来自数据分析网元的特征集合的信息，其中，所述特征集合的信息对应一种业务类型或一种执行规则；所述策略控制网元向用户面数据处理网元发送所述特征集合的信息。
本申请的第七方面，提供了一种数据分析的装置，包括：收发单元，用于获取来自数据分析网元的至少一个特征集合的信息以及用于接收用户面数据，其中，所述至少一个特征集合的信息中的每个特征集合的信息对应至少一种业务类型或至少一种执行规则；处理单元，用于根据所述至少一个特征集合的信息获取所述用户面数据的特征参数；所述收发单元还用于向所述数据分析网元发送所述特征参数以及获取来自所述数据分析网元的所述特征参数的响应结果；所述处理单元还用于根据所述响应结果获取所述用户面数据关联的业务类型或所述用户面数据关联的执行规则。
本申请的第八方面，提供了一种数据分析的装置，包括：收发单元，用于向用户面数据处理网元发送至少一个特征集合的信息以及接收来自所述用户面数据处理网元的用户面数据的特征参数，其中，所述至少一个特征集合的信息中的每个特征集合的信息对应至少一种业务类型或者至少一种执行规则，所述至少一个特征集合的信息包括所述特征参数对应的特征集合的信息；处理单元，用于根据所述特征参数确定所述用户面数据关联的业务类型信息或者所述用户面数据关联的执行规则信息。
本申请的第九方面，提供了一种数据分析的装置，包括：包括处理单元和收发单元，所述处理单元用于通过所述收发单元获取来自数据分析网元的至少一个特征集合的信息，其中，所述至少一个特征集合的信息中的每个特征集合的信息对应至少一种业务类型或至少一种执行规则；所述处理单元还用于通过所述收发单元向用户面数据处理网元发送所述至少一个特征集合的信息中的至少部分特征集合的信息。
在上述任何一方面及任何一种可能的实现方式，所述特征集合的信息为特征索引的集合。
在上述任何一方面及任何一种可能的实现方式，所述特征参数为特征向量。
在上述任何一方面及任何一种可能的实现方式，所述特征参数为特征值的集合。
本申请的第十方面，提供了一种数据分析的装置，包括：存储单元，用于存储计算机指令；处理单元，用于根据所述存储单元中存储的计算机指令执行上述第一方面到第六方面以及各种可能的实现方式中的任何一种方法。
本申请的第十一方面，提供了一种计算机存储介质，所述计算机可读存储介质中存储有指令，当其在计算机上运行时，使得计算机执行上述第一方面到第六方面以及各种可能的实现方式中的任何一种方法。
本申请的第十二方面，提供了一种包含指令的计算机程序产品，当其在计算机上运行时，使得计算机执行上述第一方面到第六方面以及各种可能的实现方式中的任何一种方法。
附图说明
图1是实现本申请实施例的一种通信系统的示意图。
图2是本申请第一实施例的数据分析方法的示意性流程图。
图3是本申请第二实施例的数据分析方法的示意性流程图。
图4是本申请第三实施例的数据分析方法的示意性流程图。
图5是本申请第四实施例的数据分析方法的示意性流程图。
图6是本申请第五实施例的数据分析方法的示意性流程图。
图7是本申请第六实施例的数据分析方法的示意性流程图。
图8是本申请实施例的一种数据分析装置的示意图。
具体实施方式
本申请实施例可以用于长期演进网络(long term evolution，LTE)、5G或者下一代网络、固定网络，家庭基站网络，非3GPP(如wifi)接入的移动网络等。在本申请中，以应用于5G网络进行举例说明。
图1是能够实现本申请实施例的一种通信系统的示意图。在该通信系统中，终端设备101通过接入网(access network,AN)设备102接入核心网。
其中，该终端设备101包括但不限于：用户设备(user equipment，UE)、用户单元、用户站、移动站、移动台、远方站、远程终端设备、移动终端设备、用户终端设备、终端设备、无线通信设备、用户代理、用户装置、蜂窝电话、无绳电话、会话启动协议(session initiation protocol，SIP)电话、无线本地环路(wireless local loop，WLL)站、个人数字处理(personal digital assistant，PDA)、具有无线通信功能的手持设备、计算设备、连接到无线调制解调器的处理设备、车载设备、可穿戴设备、物联网中的终端设备设备、家用电器、虚拟现实设备、未来5G网络中的终端设备设备或者未来演进的公共陆地移动网络(public land mobile network，PLMN)中的终端设备设备等。
接入网设备102可以是与终端设备101进行通信的设备。接入网设备可以为特定的地理区域提供通信覆盖，并且可以与位于该覆盖区域(小区)内的终端设备进行通信。接入网设备102可以与任意数目终端设备通信。接入网设备102与终端设备101之间可以有多个空口连接，例如，接入网设备102与终端设备101之间存在两个空口连接，分别用于传输数据流A和数据流B。接入网设备可以支持不同制式的通信协议，或者可以支持不同的通信模式。例如，该接入网设备102以是演进型基站(evolved node B，eNodeB)，或者是无线保真接入点(wireless fidelity access point，WiFi AP)、或者是全球微波接入互操作性基站(worldwide interoperability for microwave access base station，WiMAX BS)，或者是云无线接入网络(cloud radio access network，CRAN)中的无线控制器，或者该网络设备可 以为未来5G网络中的接入网设备或者未来演进PLMN中的接入网设备等。
核心网可以包括：控制面功能(control plane function，CPF)网元、用户面功能(user plane function，UPF)网元103、策略控制功能(policy control function,PCF)网元104以及NWDA网元105。其中，控制面功能网元可以包括接入管理功能(access management function，AMF)网元106和会话管理功能(session management function，SMF)网元107。通过接入网设备102和用户面功能网元103，可以实现终端设备101和数据网络(data network，DN)108之间用户面数据的传输。
PCF网元104具有策略控制决策的功能，为网络提供策略。NWDA网元105用于大数据学习和分析。AMF网元106用于移动性管理、合法监听、或者接入授权以及鉴权等。SMF网元107用于实现会话和承载管理、地址分配等。DN108为用于传输数据的网络，具体的，DN108可以为互联网协议(internet protocol,IP)多媒体子系统(IP multimedia subsystem,IMS)服务器或者分组数据网络(packet data network,PDN)或者应用服务器(application server,App server)。
可以理解的是，在图1所示的通信系统中，各组成网元的功能仅为示例性的，各个组成网元在应用于本申请的实施例中时，并非全部功能都是必需的。
在本申请的实施例中，“和/或”仅仅是一种描述关联对象的关联关系，表示可以存在三种关系。例如，A和/或B，可以表示：单独存在A，同时存在A和B，单独存在B这三种情况。另外，在本申请的描述中，“多个”是指两个或多于两个。
在本申请的实施例中，某一网元(例如：A网元)获取来自另一网元(例如：B网元)的信息，可以指A网元直接从B网元接收信息，也可以指A网元经其他网元(例如：C网元)从B网元接收信息。当A网元经C网元从B网元接收信息时，C网元可以对信息进行透传，也可以将信息进行处理，例如：将信息携带在不同的消息中进行传输或者对信息进行筛选，只发送筛选后的信息给A网元。类似的，在本申请的各实施例中，A网元向B网元发送信息，可以指A网元直接向B网元发送信息，也可以指A网元经其他网元(例如：C网元)向B网元发送信息。
图2是本申请第一实施例的数据分析方法的示意性流程图。该数据分析方法包括：
201：数据分析网元获取训练数据。
数据分析网元可以是图1中的NWDA网元。数据分析网元也可以是其他具有网络数据分析功能的网元，在此不做限定。
数据分析网元可以分别从其他网元获取训练数据，数据分析网元也可以从同一个网元集中获取训练数据。具体的，其他网元可以是电信网络设备，也可以是第三方服务器。其中，电信网络设备可以为如下设备中的至少一种：终端设备、接入网设备、控制面功能网元(例如：AMF网元或SMF网元)、UPF网元、PCF网元、网络管理网元(例如：业务支撑系统(business support system,BSS)或运营支撑系统(operation support system,OSS)或管理支撑系统(management support system,MSS))、统一数据管理(unified data management,UDM)网元和IMS网元。第三方服务器可以为App Server、OTT(over the top)服务器、垂直行业管控中心中的至少一种。
数据分析网元可以通过与其他网元进行数据交互直接获取训练数据。数据分析网元也可以通过其他网元间接的获取训练数据。例如：数据分析网元通过网络开放功能(network  exposure function,NEF)网元获取来自第三方服务器的数据。
数据分析网元可以实时地从其他网元获取训练数据，数据分析网元也可以在数据分析网元和/或其他网元空闲时获取训练数据。
数据分析网元获取的训练数据可以是原始数据。数据分析网元获取的训练数据可也可以是其他网元预处理后的数据。例如：应用服务器出于保护用户隐私的目的，将原始数据中的敏感信息清洗，并将处理后的数据发送给数据分析网元。
数据分析网元获取的训练数据可以是网络数据，例如：终端设备的地址、小区标识(cell ID)、时间信息、或者网络拥塞状况等。数据分析网元获取的训练数据还可以是应用数据，例如：IP五元组、用户面数据大小，用户面数据间隔、业务类型、业务体验、或者扩展字段等。数据分析网元还可以将获取的应用数据和网络数据进行关联，得到关联后的训练数据。例如：数据分析网元根据终端设备的地址和/或时间信息，将应用数据和网络数据进行关联。终端设备的地址可以是IP地址，也可以是以太网地址。
数据分析网元可以以业务类型为粒度获取训练数据。例如：数据分析网元分别获取视频业务的训练数据、支付业务的训练数据、或者基于长期演进的语音(voice over long term evolution,VOLTE)业务的训练数据。
数据分析网元也可以以网元为粒度获取训练数据。例如：从终端设备获取的训练数据可以包括：终端类型、终端设备的地址、操作系统版本、终端设备的温度、终端设备的电量、或者终端设备测量的小区无线信道质量等；从会话管理网元获取的训练数据可以包括：会话管理网元的标识(identifier,ID)、终端设备的地址、或者数据网络名称(data network name,DNN)等；从用户面功能网元获取的训练数据可以包括：用户面功能网元ID、隧道端点标识(tunnel end point identifier,TEID)、拥塞级别、IP五元组、用户面数据大小、或者用户面数据个数等；从接入网设备获取的训练数据可以包括：小区标识、服务质量(quality of service,Qos)参数、实时无线信道质量(例如：参考信号接收功率(reference signal receiving power,RSRP),或者参考信号接收质量(reference singnal received quality,RSRQ),或者信号与干扰加噪声比(signal to interference plus noise ratio,SINR))、业务切换门限、滤波系数、天线倾角、载频、载波、丢包率、保证比特速率(guaranteed bit rate,GBR)、最大比特速率(maximum bit rate,MBR)、或者拥塞级别等；从第三方服务器获取的训练数据可以包括：IP五元组、开始时间、结束时间、或者业务类型等。
202：数据分析网元对训练数据进行分析，获取特征集合的信息。
数据分析网元采用大数据分析方法对步骤201中获得的训练数据进行分析，获取至少一个特征集合的信息。特征集合的信息可以是具体的特征集合，也可以是跟特征集合对应的信息，例如：特征集合的索引。该至少一个特征集合的信息中的每个特征集合的信息对应至少一种业务类型或至少一种执行规则。执行规则可以是执行策略(enforcement policy)、控制策略、计费策略(charging policy)、策略与计费控制(policy&charging control,PCC)规则(rule)中的至少一种。PCC规则可以包括服务质量策略。
数据分析网元对各种业务类型的训练数据分别进行分析，从而获取每种业务类型对应的特征集合的信息。其中，一种业务类型可以唯一的对应一个特征集合的信息；一种业务类型也可以对应多个特征集合的信息；一个特征集合的信息可以唯一的对应一种业务类型，一个特征集合的信息也可以对应多种业务类型。当一个特征集合的信息对应多种业务类型 时，还可以结合该特征集合的信息以外的其他条件或者特征确定对应的业务类型。
数据分析网元可以根据特征集合的信息对应的业务类型，确定该业务类型的执行规则信息，该业务类型的执行规则信息即该特征集合的信息对应的执行规则信息。例如：当业务类型为支付业务时，确定对支付类业务的处理优先级。在本申请中，执行规则信息可以是具体的执行规则，也可以是跟执行规则相关的信息，例如：用于获取执行规则信息。数据分析网元也可以根据特征集合的信息中的内容，确定特征集合的信息对应的执行规则信息。例如：数据分析网元可以根据特征集合中的无线信道质量、拥塞级别、丢包率、切换门限中的至少一个确定特征集合对应的执行规则。数据分析网元可以根据训练数据确定执行规则相关的信息，例如：无线信道质量、拥塞级别、丢包率、或者切换门限等。
数据分析网元可以获取业务粒度的特征。例如：若某种业务的训练数据对应的用户面数据的大小为某特定值或属于某特定范围，则数据分析网元可以将用户面数据的大小作为该种业务的一个特征。又例如：若某种业务的训练数据对应的用户面数据来自某一或某些特性的数据网络，则数据分析网元可以将该数据网络名称作为该种业务的另一个特征。
数据分析网元可以获取终端设备粒度的特征。例如：不同厂商生产的终端设备对应的某种业务的用户面数据特征可能不同。当要区分不同终端设备的该种业务时，数据分析网元可以将终端设备的类型作为该种业务的一个特征。可选的，数据分析网元可以根据国际移动设备身份码(international mobile equipment identity，IMEI)中的类型分配码(type allocation code，TAC)区分终端设备的类型，因此，数据分析网元可以将终端设备的IMEI作为一个特征。又例如：不同操作系统的终端设备对应的某种业务的用户面数据特征可能不同。因此，数据分析网元也可以将终端设备的操作系统的类型作为该种业务的一个特征。
数据分析网元还可以获取用户粒度的特征。例如：某用户为商家用户，则该用户持有的终端设备在固定位置范围内长时间进行某种业务。因此，数据分析网元可以将该终端设备的位置信息作为该种业务的一个特征。可选的，数据分析网元可以从应用服务器获取该终端设备的公网地址信息和端口号信息。然后数据分析网元可以向网络地址转换(network address translation,NAT)网元查询该公网地址对应的内网地址信息。数据分析网元向网管获取该内网地址信息对应的用户标识信息和位置信息等。
以视频业务为例，数据分析网元对训练数据进行分析，可以得到如下特征集合：
<终端设备类型，终端设备位置，时间，数据流(flow)中首个数据包的大小，数据流中所有数据包大小的均值，数据流中所有数据包大小的熵值>。
以支付业务为例，数据分析网元对训练数据进行分析，可以得到如下特征集合：
<终端设备类型，终端设备位置，小区标识，时间，数据流中所有数据包上下行的时间间隔的均值，数据流中所有数据包上下行的时间间隔的熵值>。
需要说明的是，本申请中视频业务和支付业务的特征集合仅仅是为了方便理解而进行的举例说明，本申请中视频业务和支付业务的特征集合还可以是其他内容，本申请并不做限定。
在一种可能的实现方式中，数据分析网元可以将各种业务类型对应的特征集合进行汇总，得到一个总的特征集合，并且给该总的特征集合中的每个特征设置索引。基于总的特征集合，数据分析网元可以得到每种业务对应的特征集合的索引。可选的，数据分析网元也可以将部分业务类型对应的特征集合进行汇总，得到对应该部分业务的总的特征集合， 并且给该特征集合中的每个特征设置特征索引。
例如：数据分析网元可以将视频业务和支付业务的特征集合进行合并，得到如表1所示的总的特征集合。
表1
特征 特征索引
终端设备类型 1
终端设备位置 2
小区标识 3
时间 4
数据流中首个数据包的大小 5
数据流中所有数据包大小的均值 6
数据流中所有数据包大小的熵值 7
数据流中所有数据包上下行的时间间隔的均值 8
数据流中所有数据包上下行的时间间隔的熵值 9
基于表1，可以得到视频业务对应的特征索引的集合是<1,2,4,5,6,7>，支付业务对应的特征索引的集合是<1,2,3,4,8,9>。
在另一种可能的实现方式中，一个特征对应的特征索引可以用一个二进制字符表示，则一种业务类型对应的特征索引的集合可以用一个二进制字符串表示。假设总的特征集合中特征个数为n，则一种业务类型对应的特征集合的索引可以用n位的二进制字符串表示。例如：当一种业务类型对应的二进制字符串中的第i位为1时，表示该业务类型对应的特征集合包括总的特征集合中的第i个特征。基于表1，以上述视频业务和支付业务进行举例说明，视频业务的特征索引的集合<1,2,4,5,6,7>对应的二进制字符串为110111100，支付业务的特征索引的集合<1,2,3,4,8,9>对应的二进制字符串为111100011。
数据分析网元还可以获取每个特征集合的信息对应的匹配算法。该匹配算法可以通过大数据分析训练数据的方法获取，例如：数据分析网元获取某种业务类型的训练数据，然后获取该种业务类型的训练数据对应的特征集合的信息和特征参数(特征参数可以是特征值的集合，例如：特征向量)。数据分析网元根据获得的特征参数，通过大数据分析方法获得该特征集合的信息对应的匹配算法。数据分析网元也可以通过预先配置的方法获取每个特征集合的信息对应的匹配算法。可选的，匹配算法可以是一个数学函数或者数据模型。
在一种可能的实现方式中，数据分析网元将该匹配算法和该特征集合的信息对应的业务类型或者执行规则关联。关联可以理解成建立映射关系。匹配算法的输入可以是用户面数据的特征参数。匹配算法的输出结果是该特征参数是否符合该匹配算法。根据输出结果以及匹配算法关联的业务类型即可确定输入的特征参数对应的用户面数据所关联的业务类型。例如：若匹配算法的输出结果为1，则表示输入的特征参数符合该匹配算法，则该匹配算法关联的业务类型为对应的用户面数据关联的业务类型；若匹配算法的输出结果为0，则表示输入的特征参数不符合该匹配算法，则该匹配算法关联的业务类型不是对应的用户面数据关联的业务类型。
在另一种可能的实现方式中，数据分析网元将该匹配算法的输出结果与业务类型或者执行规则关联。关联可以理解成建立映射关系。例如：输出结果为0对应第一种业务类型或者第一种执行规则，输出结果为1对应第二种业务类型或者第二种执行规则，输出结果为2～4对应第三种业务类型或者第三种执行规则。匹配算法的输入可以是用户面数据的特 征参数，根据匹配算法的输出结果即可确定用户面数据关联的业务类型或者执行规则。在这种场景下，匹配算法的输出结果可以为业务类型信息或者执行规则信息。
203：数据分析网元向策略控制网元发送至少一个特征集合的信息。
策略控制网元可以是图1中的PCF网元，也可以是其他具有策略控制功能的网元在此不做限定。
数据分析网元可以将多个业务类型对应的特征集合的信息同时发给策略控制网元，数据分析网元也可以将多个业务类型对应的特征集合的信息分别发给策略控制网元。数据分析网元可以主动向策略控制网元发送特征集合的信息，也可以根据策略控制网元的请求发送特征集合的信息。数据分析网元可以实时的向策略控制网元发送特征集合的信息，也可以在网络空闲时向策略控制网元发送特征集合的信息，在此不做限定。
在一种可能的实现方式中，数据分析网元可以根据部分特征从步骤202中预先获取的特征集合的信息中选择部分特征集合的信息，然后，将选择的部分特征集合的信息作为上述至少一个特征集合的信息向策略控制网元发送。部分特征可以是特征集合中的某个或者某些特征，也可以是不在特征集合中但跟特征集合关联的特征。例如：部分特征可以是DNN和/或与数据网络对应的UPF网元的标识信息。例如：数据分析网元可以接收来自会话管理网元的消息，该消息中包括DNN和/或UPF网元的标识信息，数据分析网元从预先获取的特征集合的信息中选择与DNN和/或UPF网元的标识信息一致的特征集合的信息。
数据分析网元还可以向策略控制网元发送特征集合的信息关联的部分特征的信息。部分特征可以是特征集合中的某个或者某些特征，也可以是不在特征集合中但跟特征集合关联的特征。例如：部分特征可以是DNN和/或与数据网络对应的UPF网元的标识信息，部分特征也可以是业务所在服务器的地址信息、部分特征还可以是IP五元组信息。部分特征的信息可以同关联的特征集合的信息一起发送给策略控制网元，部分特征的信息也可以跟特征集合的信息分别发送给策略控制网元。
数据分析网元还可以向策略控制网元发送该至少一个特征集合的信息对应的业务类型信息。该业务类型信息可以是业务类型，也可以是业务类型的指示信息，例如：业务类型的编号、特征集合的信息对应的编号、匹配算法的输出结果中的至少一种。该业务类型信息可以包括在对应的特征集合的信息中向策略控制网元发送，该业务类型信息也可以单独向策略控制网元发送。
数据分析网元还可以向策略控制网元发送该至少一个特征集合的信息对应的执行规则信息。其中，执行规则信息可以是与执行规则相关的特征的信息，也可以是具体的执行规则，还可以是执行规则的指示信息，例如：执行规则的编号，在此不做限定。
下面以表2进行举例说明数据分析网元向策略控制网元发送的内容：
表2中，特征集合的信息为特征索引的集合，对特征索引的集合进行编号，一个特征索引的集合对应一种业务类型。业务类型的信息为业务类型的编号。执行规则信息为执行规则相关的特征，执行规则相关的特征为业务的无线信道质量切换门限。业务类型或者特征索引的集合关联的部分特征为网络名称。数据分析网元可以仅向策略控制网元发送表2中的特征索引的集合。数据分析网元还可以向策略控制网元发送编号、业务类型、切换门限、网络名称中的至少一个。
表2
Figure PCTCN2018104115-appb-000001
204：策略控制网元接收数据分析网元发送的至少一个特征集合的信息。策略控制网元向用户面数据处理网元发送所述至少一个特征集合中的至少部分特征集合的信息。
用户面数据处理网元可以是图1中的用户面功能网元，也可以是接入网设备，还可以是其他具有用户面数据处理功能的网元。在本申请中，策略控制网元可以经会话管理网元向用户面数据处理网元发送所述至少部分特征集合的信息。
策略控制网元可以将接收到的至少一个特征集合的信息发送给用户面数据处理网元。策略控制网元也可以根据部分特征从接收到的至少一个特征集合的信息中选择部分特征集合的信息。部分特征可以是特征集合中的特征，也可以是不在特征集合中但跟特征集合关联的特征。例如：策略控制网元接收来自会话管理网元的部分特征。该部分特征可以是DNN和/或数据网络对应的UPF网元的标识信息。策略控制网元从接收到的至少一个特征集合的信息中选择与DNN和/或UPF网元的标识信息一致的特征集合的信息作为所述至少部分特征集合的信息发送给用户面数据处理网元。例如：策略控制网元可以从数据分析网元接收表2中的5个特征索引的集合，若策略控制网元接收到来自会话管理网元的会话管理消息，该会话管理消息中包括的网络名称为DNN-1，则策略控制网元可以仅向用户面数据处理网元发送DNN-1对应的2个特征索引的集合。
策略控制网元可以接收来自数据分析网元的业务类型信息。该业务类型信息可以包括在对应的特征集合的信息中，策略控制网元也可以分别接收特征集合的信息及该特征集合的信息对应的业务类型信息。
策略控制网元也可以不从数据分析网元接收业务类型信息。策略控制网元可以根据特征集合中的部分特征，分析得出该特征集合的信息对应的业务类型。例如：根据IP五元组信息确定该特征集合的信息对应的业务类型的信息。
策略控制网元可以向用户面数据处理网元发送所述至少部分特征集合的信息对应的业务类型信息。该业务类型信息可以与所述至少部分特征集合的信息一起发送，也可以与所述至少部分特征集合的信息分别发送。
策略控制网元可以根据获取的业务类型信息，为所述至少部分特征集合的信息生成执 行规则，即为所述至少部分特征集合的信息对应的业务类型生成执行规则。策略控制网元也可以接收来自数据分析网元的执行规则，该执行规则跟所述至少部分特征集合的信息对应。策略控制网元还可以接收来自数据分析网元的执行规则相关的特征，根据接收到的特征确定所述至少部分特征集合的信息对应的执行规则。例如：策略控制网元可以从数据分析网元接收表2中的切换门限，根据接收到的切换门限，为该切换门限对应的特征索引的集合生成执行规则。策略控制网元可以同时从数据分析网元获取特征集合的信息及该特征集合的信息对应的执行规则信息，策略控制网元也可以分别从数据分析网元接收特征集合的信息及该特征集合的信息对应的执行规则信息。
策略控制网元可以向用户面处理网元发送所述至少部分特征集合的信息对应的执行规则信息。该执行规则信息可以与所述至少部分特征集合的信息一起发送，也可以与所述至少部分特征集合的信息分别发送，在此不做限定。
205：用户面数据处理网元接收来自策略控制网元的所述至少部分特征集合的信息。
在一种可能的实现方式中，用户面数据处理网元可以存储接收到的特征集合的信息。当用户面数据处理网元接收到的是特征索引的集合时，用户面数据处理网元根据特征索引的集合获取对应的特征集合。
在一种可能的实现方式中，当用户面数据处理网元无法直接获取特征集合的信息中的部分特征时，用户面数据处理网元可以将无法直接获取的部分特征转换为能够直接获取的特征。例如：用户面数据处理网元接收到的用户面数据中并不一定包括用户标识信息。若特征集合中包括用户标识信息，用户面数据处理网元可以查询IP地址池，获得用户标识信息对应的IP地址，然后将该IP地址与该用户标识信息关联。因此，用户面数据处理网元可以将接收到的用户面数据中的IP地址与特征集合中的IP地址进行匹配。
用户面数据处理网元可以接收来自策略控制网元的业务类型信息，该业务类型信息与用户面数据处理网元接收到的特征集合的信息对应。该业务类型信息的形式以及接收方式参见步骤203-205，在此不再描述。
用户面数据处理网元还可以接收来自策略控制网元的执行规则信息，该执行规则信息与用户面数据处理网元接收到的特征集合的信息对应。该执行规则信息的形式以及接收方式参见步骤204，在此不再描述。
在一种可能的实现方式中，数据分析网元以及用户面数据处理网元均可以获取总的特征集合。因此，用户面数据处理网元接收到特征集合的索引后，用户面数据处理网元根据总的特征集合以及特征集合的索引获取特征集合的索引对应的特征集合。采用这种方法可以进一步的降低网元间的数据吞吐量，并避免每次都需要在网元间传递业务关联的特征的定义。可选的，数据分析网元可以和用户面数据处理网元同步总的特征集合，例如：数据分析网元向用户面数据处理网元发送总的特征集合。数据分析网元可以实时的和用户面数据处理网元同步总的特征集合，也可是是周期性的同步总的特征集合。数据分析网元和用户面数据处理网元也可以通过预先配置的方式获得总的特征集合。
步骤203和204为可选步骤，数据分析网元也通过服务化接口经会话管理网元向用户面数据处理网元发送至少一个特征集合的信息、特征集合的信息对应的业务类型信息、特征集合的信息对应的执行规则信息中的至少一种。
206：用户面数据处理网元接收用户面数据，根据接收到的至少部分特征集合的信息 获取所述用户面数据的特征参数。
在本申请中，用户面数据可以是数据包。
若用户面数据处理网元接收到的是特征集合，当用户面数据处理网元接收到用户面数据后，根据接收到的特征集合，用户面数据处理网元获取用户面数据的特征参数。若用户面数据处理网元接收到的是特征索引的集合，当用户面数据处理网元接收到用户面数据后，用户面数据处理网元根据特征索引确定该索引对应的特征，然后获取用户面数据对应该特征的特征参数，进而，用户面数据处理网元可以获取用户面数据的特征参数，该特征参数与接收到的特征索引的集合对应。
当用户面数据处理网元接收到多个特征集合的信息时，用户面数据处理网元可以获取用户面数据的多个特征参数，该多个特征参数与接收到的多个特征集合的信息分别对应。在一种可能的实现方式中，当用户面数据处理网元接收到用户面数据后，可以根据部分特征，从接收到的特征集合的信息中选择部分特征集合的信息。然后，用户面数据处理网元获取用户面数据的特征参数，该特征参数与选择出的部分特征集合的信息对应。部分特征可以是特征集合中的特征，也可以是不在特征集合中但跟特征集合关联的特征。例如：部分特征可以为IP五元组信息。用户面数据处理网元从数据分析网元获取特征集合的信息以及该取特征集合的信息对应的IP五元组信息。具体的，部分特征可以为IP地址和/或端口号。用户面数据处理网元从接收到的特征集合的信息中选择IP地址和/或端口号与用户面数据对应的IP地址和/或端口号一致的特征集合的信息。
当用户面数据处理网元获取用户面数据的一个特征参数时，用户面数据处理网元可以对应特征集合依次获取用户面数据的特征参数。用户面数据处理网元也可以先获取特征集合中部分特征的特征参数，若该特征参数不满足预设的条件，则用户面数据处理网元停止获取用户面数据对应该特征集合的特征参数。例如：特征集合中依次包括如下特征：用户面数据大小、间隔、熵、IP地址、端口号时，用户面数据处理网元可以先获取用户面数据在IP五元组中的IP地址和端口号，若用户面数据在IP五元组中的IP地址和端口号不满足预设的条件，则用户面数据处理网元停止获取该特征集合的特征参数。
在一种可能的实现方式中，若特征集合中包括用户面数据处理网元自身无法获取的特征，用户面数据处理网元可以向其他网元(例如：其他用户面数据处理网元或者控制面网元)获取该特征对应的特征参数。在本申请中，用户面数据处理网元自身无法获取的特征对应的特征参数，称为用户面数据关联的信息。例如：若特征集合中包括终端设备的位置信息，则用户面数据处理网元从接入网设备获取该终端设备的位置信息。该终端设备的位置信息即为用户面数据的一种关联的信息。具体的，用户面数据处理网元可以请求接入网设备发送终端设备的位置信息，用户面数据处理网元也可以向接入网设备订阅终端设备的位置信息，当然，用户面数据处理网元还可以向会话管理网元获取终端设备的位置信息。用户面数据处理网元将自身获取的特征参数与从其他网元获取的特征参数汇总，从而可以获得用户面数据的特征参数。
特征参数可以是用户面数据对应该特征的实际值，也可以是处理后的值，例如：归一化后的值，或量化后的值。在本申请中，特征参数也可以是特征的具体内容，并不一定是一个具体的取值，例如：DNN或终端设备的类型。在一种可能的实现方式中，用户面数据的特征参数可以以特征向量的形式体现。
207：用户面数据处理网元向数据分析网元发送用户面数据的特征参数。
用户面数据处理网元将步骤206中获得的特征参数向数据分析网元发送。具体的，用户面数据处理网元可以经会话管理网元和策略控制网元向数据分析网元发送获得的特征参数。可选的，用户面数据处理网元也可以经会话管理网元以及会话管理网元与数据分析网元之间的服务化接口向数据分析网元发送获取的特征参数。
可选的，用户面数据处理网元还可以向数据分析网元发送特征参数的关联信息，例如：特征集合的信息的编号。
208：数据分析网元根据特征参数确定用户面数据关联的业务类型信息和/或执行规则信息。
数据分析网元接收到用户面数据的特征参数后，根据该特征参数和步骤202中获得的匹配算法确定该用户面数据关联的业务类型信息或执行规则信息。业务类型信息可以是具体的业务类型，也可以是业务类型的指示信息，具体参见步骤203。执行规则信息可以是具体的执行规则，也可以是执行规则相关的特征，具体参见步骤203。在一种可能的实现方式中，若数据分析网元接收到多个特征参数，可以将每个特征参数输入到对应的匹配算法中，然后根据匹配算法的输出结果确定用户面数据关联的业务类型信息或执行规则信息。在本申请的实施例中，特征参数对应的匹配算法即为特征参数对应的特征集合的信息对应的匹配算法。可选的，可以通过特征参数的关联信息确定对应的匹配算法，例如：特征参数的关联信息可以是特征集合的信息对应的编号(例如：表2中的编号)。在另一种可能的实现方式中，若数据分析网元接收到多个特征参数，也可以将每个特征参数均输入到数据分析网元预先获取的多个匹配算法中，然后根据匹配算法的输出结果确定用户面数据关联的业务类型信息或执行规则信息。
下面结合表2进行举例说明：
数据分析网元可以接收来自用户面数据处理网元的编号为1和2的特征索引的集合对应的特征向量1和特征向量2。数据分析网元将特征向量1输入到编号1为的特征索引的集合对应的匹配算法1，得到的输出结果为0。数据分析网元将特征向量2输入到编号2为的特征索引的集合对应的匹配算法2，得到的输出结果为1。则数据分析网元可以将匹配算法2关联的业务类型2作为用户面数据关联的业务类型。
可选的，数据分析网元也可以根据用户面数据关联的业务类型信息确定用户面数据对应的执行规则信息，参见步骤202。
209：数据分析网元向用户面数据处理网元发送特征参数的响应结果。
响应结果可以包括用户面数据的业务类型信息和/或执行规则信息。
数据分析网元可以经策略控制网元和会话管理网元向用户面数据处理网元发送特征参数的响应结果。数据分析网元也可以通过服务化接口经会话管理网元向用户面数据处理网元发送特征参数的响应结果。
该响应结果可以包括用户面数据关联的业务类型，例如：步骤208中数据分析网元确定的用户面数据关联的业务类型；或者，该响应结果可以包括用户面数据关联的业务类型的指示信息，例如：特征索引的集合的编号(如：步骤208中的编号2)或者业务类型的编号；或者，该响应结果可以包括用户面数据关联的业务类型对应的特征参数，例如：步骤208中的特征向量2。可选的，响应结果还可以包括特征参数是否符合对应的匹配算法的确定 结果，例如：步骤208中匹配算法输出的结果0或1。
可选的，策略控制网元接收到响应结果后，可以根据该响应结果为用户面数据关联的业务类型生成对应的执行规则信息，参见步骤203。然后，策略控制网元将生成的执行规则信息发送给用户面数据处理网元。
可选的，该响应结果也可以包括用户面数据对应的执行规则信息。数据分析网元可以根据用户面数据关联的业务类型信息确定用户面数据对应的执行规则信息，具体参见步骤202。
210：用户面数据处理网元获取来自数据分析网元的特征参数的响应结果，根据响应结果获取用户面数据关联的业务类型和/或用户面数据关联的执行规则。
用户面数据处理网元可以经策略控制网元和会话管理网元接收来自数据分析网元的特征参数的响应结果。若响应结果中包括业务类型，则将接收到的业务类型确定为用户面数据关联的业务类型；若响应结果中包括业务类型的指示信息，则根据指示信息获取用户面数据关联的业务类型。
用户面数据处理网元可以根据户面数据关联的业务类型信息获取用户面数据关联的执行规则；用户面数据处理网元也可以根据响应结果获取来自策略控制网元的用户面数据关联的执行规则，例如：用户面数据处理网元根据响应结果中包括的业务类型信息获取来自策略控制网元的执行规则；用户面数据处理网元还可以直接从响应结果中获取用户面数据关联的执行规则。
211：用户面数据处理网元根据用户面数据关联的业务类型和/或所述用户面数据关联的执行规则处理用户面数据。
在一种可能的实现方式中，用户面数据处理网元根据用户面数据关联的业务类型对该用户面数据进行处理。例如：用户面数据处理网元根据用户面数据关联的业务类型为用户面数据增加业务类型的标签信息；或者，用户面数据处理网元按照业务的优先级信息转发用户面数据；或者，用户面数据处理网元根据用户面数据关联的业务类型为用户面数据增加调度优先级信息；或者，用户面数据处理网元确定用户面数据的目的地址为终端设备的地址且该终端设备处于空闲状态，用户面数据处理网元根据用户面数据关联的业务类型向会话管理网元发送该终端设备的寻呼优先级信息；或者用户面数据处理网元根据用户面数据关联的业务类型对用户面数据进行计费统计。
在另一种可能的实现方式中，用户面数据处理网元根据用户面数据关联的执行规则对该用户面数据进行处理。例如：用户面数据处理网元根据执行规则中指示的业务优先级信息转发用户面数据；或者，用户面数据处理网元根据执行规则为用户面数据增加业务类型的标签信息；或者，用户面数据处理网元根据执行规则为用户面数据增加调度优先级信息；或者，用户面数据处理网元确定用户面数据的目的地址为终端设备的地址且终端设备处于空闲状态，用户面数据处理网元根据执行规则向会话管理网元发送所述终端设备的寻呼优先级信息；或者，用户面数据处理网元根据用户面数据关联的执行规则对用户面数据进行计费统计。
在本实施例中，用户面数据处理网元获取来自数据分析网元的至少一个特征集合的信息。用户面数据处理网元根据接收到的特征集合的信息获取用户面数据关联的业务类型或用户面数据关联的执行规则，从而实现了在通信网络中利用数据分析网元对数据进行分析。 进一步的，采用大数据分析方法获取用户面数据关联的业务类型或执行规则，并不需要对用户面数据进行深度包检测(deep packet inspection,DPI)，本实施例的方法更加简单。更进一步的，在本实施例中，用户面数据处理网元根据接收到的特征集合的信息获取用户面数据关联的业务类型或用户面数据关联的执行规则，从而可以避免数据分析网元向用户面数据处理网元发送匹配算法，可以防止因匹配算法的频繁更新而引起的网络拥塞。
请参阅图3，图3是本申请第二实施例的数据分析方法的示意性流程图。在本实施例中，以数据分析网元为NWDA网元，用户面数据处理网元为UPF网元，第三方服务器为OTT服务器，特征集合的信息为特征索引的集合、特征参数为特征向量，进行举例说明。在本实施例中，NWDA网元通过UE发起的分组数据单元(packet unit data,PDU)会话建立/修改过程向UPF网元发送特征集合的信息。本实施与第二实施例相同的部分可以参见第二实施例的描述。本实施例的数据分析方法包括：
301：NWDA网元获取训练数据。
NWDA网元获取训练数据的方法可以参见步骤201，在此不再描述。
302：NWDA网元对训练数据进行分析，获取特征索引的集合。
该步骤具体可以参见步骤202，在此不再描述。
303：终端设备经AMF网元向SMF网元请求发起PDU会话建立/修改流程。
该步骤可以参见现有技术中终端设备发起PDU会话建立/修改请求的过程。
304：SMF网元向PCF网元发送请求消息，用于请求PDU会话相关的执行规则。PDU会话相关的执行规则可以是执行策略、控制策略、计费策略、策略与计费控制中的至少一种等。
该步骤可以参见现有技术中SMF发起发起的PDU连接访问网络(PDU connectivity access network,PDU-CAN)会话建立/修改流程。
可选的，该请求消息包括终端设备请求的业务所在的数据网络的名称(DNN)和/或数据网络关联的用户面功能标识(UPF ID)。
305：PCF网元向NWDA网元发送请求消息，用于请求特征索引的集合。
可选的，请求的特征索引的集合为终端设备请求的业务类型对应的特征索引的集合。
可选的，该请求消息包括终端设备请求的业务所在的数据网络的名称和/或数据网络关联的用户面功能标识，即步骤304中PCF网元接收到的DNN和/或UPF ID。
306：NWDA网元接收到PCF网元发送的请求消息后，从步骤302获取的特征索引的集合中选择与接收到的DNN和/或UPF ID一致的特征索引的集合。
在一种可能的实现方式中，特征索引的集合对应的特征集合包括DNN和/或UPF ID。在另一种可能的实现方式中，特征索引的集合对应的特征集合不包括DNN和/或UPF ID，但DNN和/或UPF ID与特征索引的集合关联。
步骤306是可选的步骤。NWDA网元接收到PCF网元发送的请求消息后，也可以不执行步骤306而直接执行步骤307。
307：NWDA网元向PCF网元发送特征索引的集合。
当NWDA网元执行步骤306时，NWDA发送步骤306中NWDA网元选择的特征索引的集合。当NWDA网元没有执行步骤306时，NWDA网元发送预设的特征索引的集合。可选的，该预设的特征索引的集合可以是NWDA网元获取的全部特征索引的集合，也可 以是NWDA网元根据当前的网络情况确定的特征索引的集合。NWDA网元向PCF网元发送特征索引的集合的方式具体参见步骤203的描述。
在步骤307中，NWDA网元还可以向PCF网元发送特征索引的集合对应的业务类型信息，具体参见步骤203。NWDA网元还可以向PCF网元发送特征索引的集合关联的特征，例如，表2中的DNN。
在一种可能的实现方式中，NWDA网元向PCF网元发送步骤305中请求消息的响应消息，该响应消息包括上述NWDA网元向PCF网元发送的内容。
308：PCF网元根据业务类型信息生成执行规则。
PCF网元根据接收到的业务类型信息(例如：业务类型)生成该业务类型对应的执行规则(例如：服务质量策略)。
可选的，PCF网元还可以从生成的执行规则和从NWDA网元接收到的执行规则中，选择一个执行规则。
309：PCF网元向SMF网元发送特征索引的集合。
PCF网元向SMF网元发送的特征索引的集合可以为PCF网元接收到的来自NWDA网元的特征索引的集合。
PCF网元向SMF网元发送的特征索引的集合还可以是终端设备请求的业务类型对应的特征索引的集合。
可选的，PCF网元还可以向SMF网元发送特征索引的集合关联的特征、特征索引的集合对应的业务类型信息、执行规则中的至少一个。
在一种可能的实现方式中，PCF网元向SMF网元发送步骤304中的请求消息的响应消息，该响应消息包括上述PCF网元向SMF网元发送的内容。
响应消息可以参见现有技术中的PDU-CAN会话建立/修改流程。
310：SMF网元向UPF网元发送特征索引的集合。
SMF网元向UPF网元发送的特征索引的集合可以为SMF网元接收到的特征索引的集合。
可选的，SMF网元还可以向UPF网元发送特征索引的集合关联的特征、特征索引的集合对应的业务类型信息、执行规则中的至少一个。
在一种可能的实现方式中，在SMF网元向UPF网元发起的会话建立/修改的流程中，SMF网元向UPF网元发送特征索引的集合。
步骤309和步骤310可以参见步骤204，在此不再详述。
311：UPF网元接收用户面数据，并根据接收到的特征索引的集合获取所述用户面数据对应该特征索引的集合的特征向量。
具体参见步骤205和步骤206，在此不再描述。
312：UPF网元向SMF网元发送特征向量。
UPF网元向SMF网元发送的特征向量可以是步骤311中获得的特征向量。
可选的，UPF网元还可以向SMF网元发送特征向量对应的特征索引的集合编号。
在一种可能的实现方式中，在UPF网元向SMF网元发送的会话建立/修改响应的流程中，UPF网元向SMF网元发送特征向量。
313：SMF网元经PCF网元向NWDA网元发送接收到的特征向量。可选的，SMF网 元还可以向NWDA网元发送特征向量对应的特征索引的集合编号。
该步骤可以参见步骤207，在此不再描述。
314：NWDA网元根据接收到的特征向量和对应的匹配算法确定用户面数据关联的业务类型信息，具体参见步骤208的描述。
315：NWDA网元经PCF网元和SMF网元向UPF网元发送特征向量的响应结果。
响应结果包括可以包括用户面数据的业务类型信息和/或执行规则信息。
在一种可能的实施方式中，响应结果可以包括特征向量符合对应的匹配算法的结果，响应结果也可以包括全部特征向量是否符合对应的匹配算法的结果。响应结果还包括特征向量对应的特征索引的集合编号。
具体参见步骤209的描述。
316：UPF网元根据响应结果，处理对应的用户面数据。
具体参见步骤210和211的描述。
请参阅图4，图4是本申请第三实施例的数据分析方法的示意性流程图。本实施例的数据分析方法包括：
401～402：具体参见步骤301～302的描述。
403：NWDA网元向PCF网元发送特征索引的集合。该特征索引的集合为NWDA网元在步骤402中根据训练数据获得的特征索引的集合。NWDA网元向PCF网元发送特征索引的集合的方式具体参见步骤203的描述。
404～405：具体参见步骤303～304的描述。
406：PCF网元接收SMF发送的请求消息后，从步骤403中接收的特征索引的集合中选择与SMF网元发送的当前PDU会话的DNN和/或UPF ID一致的特征索引的集合。PCF网元如何选择特征索引的集合可以参见步骤306中NWDA网元选择特征索引的集合的方法。
本步骤为可选步骤，当不执行本步骤时，可以直接执行步骤407。
407：PCF网元根据业务类型信息生成执行规则。
具体参见步骤308的描述。
408：PCF网元向SMF网元发送特征索引的集合。
在一种可能的实现方式中，当PCF网元执行步骤406时，PCF网元向SMF网元发送步骤406中PCF网元选择的特征索引的集合。当PCF网元没有执行步骤406时，PCF网元向SMF网元发送PCF网元预设的特征索引的集合。可选的，该预设的特征索引的集合可以是PCF网元从NWDA网元接收的全部特征索引的集合，也可以是PCF网元根据当前的网络情况确定的特征索引的集合。PCF网元向SMF网元发送的特征索引的集合还可以是终端设备请求的业务类型对应的特征索引的集合。
可选的，PCF网元还可以向SMF网元发送特征索引的集合关联的特征、特征索引的集合对应的业务类型信息、执行规则中的至少一个。
在一种可能的实现方式中，PCF网元向SMF网元发送步骤405中的请求消息的响应消息，该响应消息包括上述PCF网元向SMF网元发送的内容。
409～415：具体参见步骤310～316的描述。
请参阅图5，图5是本申请第四实施例的数据分析方法的示意性流程图。本实施例的 数据分析方法包括：
501～507：具体参见步骤301～307的描述。
508～513：具体参见步骤309～314的描述。
514：NWDA网元向PCF网元发送特征向量的响应结果。
响应结果包括可以包括用户面数据的业务类型信息。
在一种可能的实施方式中，响应结果可以包括特征向量符合对应的匹配算法的结果，响应结果也可以包括全部特征向量是否符合对应的匹配算法的结果。响应结果还包括特征向量对应的特征索引的集合编号。
515：PCF网元根据响应结果生成执行规则。
PCF网元根据接收到的业务类型信息(例如：业务类型)生成该业务类型对应的执行规则(例如：服务质量策略)。
可选的，PCF网元还可以从生成的执行规则和从NWDA网元接收到的执行规则中，选择一个执行规则。
516：PCF网元经SMF网元向UPF网元发送响应结果。
该响应结果包括用户面数据的业务类型信息和/或执行规则。
517：UPF网元根据响应结果，处理对应的用户面数据。
具体参见步骤316的描述。
请参阅图6，图6是本申请第五实施例的数据分析方法的示意性流程图。本实施例的数据分析方法包括：
601～606：具体参见步骤401～406的描述。
607～612：具体参见步骤408～413的描述。
613：NWDA网元向PCF网元发送特征向量的响应结果。
响应结果包括可以包括用户面数据的业务类型信息。
在一种可能的实施方式中，响应结果可以包括特征向量符合对应的匹配算法的结果，响应结果也可以包括全部特征向量是否符合对应的匹配算法的结果。响应结果还包括特征向量对应的特征索引的集合编号。
614：PCF网元根据响应结果生成执行规则。
PCF网元根据接收到的业务类型信息(例如：业务类型)生成该业务类型对应的执行规则(例如：服务质量策略)。
可选的，PCF网元还可以从生成的执行规则和从NWDA网元接收到的执行规则中，选择一个执行规则。
615：PCF网元经SMF网元向UPF网元发送响应结果。
该响应结果包括用户面数据的业务类型信息和/或执行规则。
616：UPF网元根据响应结果，处理对应的用户面数据。
具体参见步骤415的描述。
在本申请一种可能的实现方式中，用户面数据处理网元和数据分析网元中可以设置特征提取单元。数据分析网元中的特征提取单元根据某种业务类型的训练数据获取该种业务类型的特征集合的信息，进而获取训练数据对应的特征参数(例如：特征向量)。用户面数据处理网元中的特征提取单元获取户面数据处理网元发送的用户面数据和特征集合的信 息，根据获取的特征集合的信息获取户面数据对应的特征参数(例如：特征向量)。其中，特征集合的信息可以是用户面数据处理网元接收到的至少一个特征集合的信息中的至少部分特征集合的信息。特征提取单元向用户面数据处理网元发送获得的特征参数。可选的，用户面数据处理网元中的特征提取单元还可以获取用户面数据关联的信息，例如：终端设备的小区标识。特征提取单元还可以获取该用户面数据关联的信息对应的特征参数。
在本申请另一种可能的实现方式中，可以只设置一个特征提取单元。该特征提取单元可以是一个单独的网元，该特征提取单元也可以设置在用户面数据处理网元或者其他网元中，在此不做限定。
下面以特征集合的信息为特征索引的集合，特征提取单元为单独的一个网元进行举例说明。图7是本申请第六实施例的数据分析方法的示意性流程图。在本实施例的流程图中，仅仅示出了用户面数据处理网元、特征提取网元和数据分析网元。用户面数据处理网元、特征提取网元、数据分析网元和其他网元的流程交互可以参见图2-图6的实施例，在此不再描述。
701：数据分析网元获取训练数据。
具体参见步骤201的描述。
702：数据分析网元向特征提取单元发送训练数据。
数据分析网元向特征提取单元发送的训练数据可以是多种业务类型对应的训练数据，也可以是一种业务类型的训练数据。数据分析网元可以主动向特征提取单元发送训练数据，也可以根据特征提取单元的请求发送训练数据。在本实施例中，若特征提取单元设置在用户面数据处理网元中，则特征提取单元和数据分析网元可以经策略控制网元和会话管理网元交互数据。
703：特征提取单元对训练数据进行分析，获取某种业务对应的特征索引的集合以及该种业务类型的训练数据对应该特征索引的集合的特征向量。
具体参见步骤202。
704：特征提取单元向数据分析网元发送步骤703中的特征向量。
可选的，特征提取单元还可以向数据分析网元发送703中的特征索引的集合。
在一种可能的实施方式中，特征提取单元可以对特征索引的集合进行编号，然后将该编号发送给数据分析网元。在另一种可能的实施方式中，特征提取单元也可以向数据分析网元发送特征索引的集合对应的业务类型。
705：数据分析网元根据某种业务的特征向量获取该种业务对应的匹配算法。
数据分析网元可以采用现有技术中的大数据分析方法获取匹配算法。匹配算法相关的内容具体参见步骤202。
706：数据分析网元向用户面数据处理网元发送接收到的特征索引的集合。
可选的，特征提取单元获得的特征索引的集合也可以采用其他方式发送给用户面数据处理网元，例如：特征提取单元直接向用户面数据处理网元发送获得的特征索引的集合。
该步骤具体参见步骤203～205。
707：用户面数据处理网元接收用户面数据后，将该用户面数据发送给特征提取单元。
可选的，用户面数据处理网元也可以将该用户面数据对应的特征索引的集合发送给特征提取单元。
708：特征提取单元根据接收到的用户面数据和特征索引的集合，获取该用户面数据的特征向量。特征提取单元将获得的用户面数据对应的特征向量发送给用户面数据处理网元。
具体的，若特征提取单元没有接收到该用户面数据对应的特征索引的集合，则特征提取单元可以获取用户面数据对应预设的特征索引的集合的特征向量。例如：预设的特征索引的集合可以是特征提取单元中的全部特征索引的集合。
709：用户面数据处理网元向数据分析网元发送用户面数据对应的特征向量。
具体参见步骤207的描述。
710：数据分析网元发根据接收到的特征向量确定用户面数据关联的业务类型信息。
具体参见步骤208的描述。
711：数据分析网元向用户面数据处理网元发送特征向量的响应结果。
具体参见步骤209的描述。
712：用户面数据处理网元根据响应结果处理用户面数据。
具体参见步骤210和步骤211的描述。
当多个网元中均部署有特征提取单元时，若需要对特征工程进行维护(例如：版本升级或特征库更新)，则需要对部署有特征提取单元的网元全部进行维护，这种部署方法维护起来比较复杂。在本实施例中，只设置一个特征提起单元，当需要对特征工程进行进行维护时，只需维护部署有特征提取单元的网元，因此维护起来简单方便。
图8是本申请实施例提供的一种数据分析装置的示意图。该数据分析装置包括收发单元801、处理单元802以及存储单元803。收发单元801、处理单元802以及存储单元803可以是在物理上相互分离的单元，也可以是集成到一个或者多个物理单元中，在此不做限定。
收发单元801用于实现处理单元802与其他单元或者网元的内容交互。具体的，收发单元801可以是该数据分析装置的通信接口，也可以是收发电路或者收发器，还可以是收发信机。收发单元801还可以是处理单元802的通信接口或者收发电路。可选的，收发单元801可以是一个收发芯片。
虽然图8中仅仅示出了一个收发单元801，数据分析装置也可以包括多个收发单元801或者收发单元801包括多个子收发单元。收发单元801还可以包括发送单元和接收单元。
处理单元802用于实现数据分析装置对数据的处理。处理单元802可以是处理电路，也可以是处理器。其中，处理器可以是中央处理器(central processing unit，CPU)，网络处理器(network processor，NP)或者CPU和NP的组合。处理器还可以进一步包括硬件芯片。上述硬件芯片可以是专用集成电路(application-specific integrated circuit，ASIC)，可编程逻辑器件(programmable logic device，PLD)或其组合。上述PLD可以是复杂可编程逻辑器件(complex programmable logic device,CPLD)，现场可编程逻辑门阵列(field-programmable gate array，FPGA)，通用阵列逻辑(Generic Array Logic，GAL)或其任意组合。
虽然图8中仅仅示出了一个处理单元802，数据分析装置也可以包括多个处理单元或者处理单元802包括多个子数据处理单元。具体的，处理器可以是一个单核(single-CPU)处理器，也可以是一个多核(multi-CPU)处理器。
存储单元803用于存储处理单元802执行的计算机指令。存储单元803可以是存储电 路也可以是存储器。存储器可以是易失性存储器或非易失性存储器，或可包括易失性和非易失性存储器两者。其中，非易失性存储器可以是只读存储器(read-only memory，ROM)、可编程只读存储器(programmable ROM，PROM)、可擦除可编程只读存储器(erasable PROM，EPROM)、电可擦除可编程只读存储器(electrically EPROM，EEPROM)或闪存。易失性存储器可以是随机存取存储器(random access memory，RAM)，其用作外部高速缓存。
存储单元803可以是独立于处理单元802的单元，也可以是处理单元802中的存储单元，在此不做限定。虽然图8中仅仅示出了一个存储单元803，数据分析装置也可以包括多个存储单元803或者存储单元803包括多个子存储单元。
在本申请的各实施例中，处理单元802可以通过收发单元801与其他网元进行内容交互，例如：处理单元802获取或者接收来自其他网元的内容。若处理单元802与收发单元801是物理上分离的两个部件，处理单元802可以不经过收发单元801与数据分析装置内部的其他单元进行内容交互。
一种可能的实现方式中，收发单元801、处理单元802以及存储单元803可以通过总线相互连接。总线可以是外设部件互连标准(peripheral component interconnect，PCI)总线或扩展工业标准结构(extended industry standard architecture，EISA)总线等。所述总线可以分为地址总线、数据总线、控制总线等。
在本申请的实施例中，处理单元802根据存储单元803中存储的计算机指令，使得实现数据分析装置实现本申请第一实施例到第六实施例中的方法。
具体的，数据分析装置可以是用户面数据处理网元，例如：UPF网元，基站。数据分析装置也可以是策略控制网元，例如：PCF网元。数据分析装置还可以是数据分析网元，例如：NWDA网元。
当数据分析装置为用户面数据处理网元时，收发单元801用于获取来自数据分析网元的至少一个特征集合的信息以及用于接收用户面数据，其中，所述至少一个特征集合的信息中的每个特征集合的信息对应至少一种业务类型或至少一种执行规则。处理单元802用于根据所述至少一个特征集合的信息获取所述用户面数据的特征参数。所述收发单元801还用于向所述数据分析网元发送所述特征参数以及获取来自所述数据分析网元的所述特征参数的响应结果。所述处理单元802还用于根据所述响应结果获取所述用户面数据关联的业务类型或所述用户面数据关联的执行规则。
其中，所述处理单元802具体用于根据所述响应结果获取来自策略控制网元的所述用户面数据关联的所述执行规则，或者，具体用于根据所述响应结果获取来自所述数据分析网元的所述用户面数据关联的执行规则。
在一种可能的实现方式中，所述处理单元802还用于根据所述用户面数据关联的业务类型或所述用户面数据关联的执行规则处理所述用户面数据。
其中，所述处理单元802具体用于根据所述用户面数据关联的执行规则中指示的业务优先级信息转发所述用户面数据；或者，所述处理单元802具体用于根据所述用户面数据关联的业务类型或所述用户面数据关联的执行规则为所述用户面数据增加业务类型的标签信息；或者，所述处理单元802具体用于根据所述用户面数据关联的业务类型或所述用户面数据关联的执行规则为所述用户面数据增加调度优先级信息；或者，所述处理单元802 具体用于根据所述用户面数据关联的业务类型或所述用户面数据关联的执行规则对所述用户面数据进行计费统计。
所述处理单元802具体用于确定所述用户面数据的目的地址为终端设备的地址且所述终端设备处于空闲状态，根据所述用户面数据关联的业务类型或所述用户面数据关联的执行规则向会话管理网元发送所述终端设备的寻呼优先级信息。
在一种可能的实现方式中，所述处理单元802具体用于根据部分特征从所述至少一个特征集合的信息中选择部分特征集合的信息；获取所述用户面数据的特征参数，所述特征参数对应所述部分特征集合的信息。
其中，所述部分特征包括包括所述用户面数据的互联网协议IP五元组信息。
在一种可能的实现方式中，所述收发单元801还用于根据所述至少一个特征集合的信息从其他用户面数据处理网元或控制面网元获取所述用户面数据关联的信息；所述处理单元802具体用于根据所述至少一个特征集合的信息和所述关联的信息获取所述用户面数据的特征参数。
在一种可能的实现方式中，所述特征集合的信息为特征索引的集合，所述特征参数为特征向量。
在本实施例中，收发单元801还用于实现本申请第一实施例到第六实施例中用户面数据处理网元与外部网元的内容的收发操作。处理单元802还用于实现本申请第一实施例到第六实施例中用户面数据处理网元内部数据或者信令的处理操作,例如，处理单元802用于实现第二实施例中步骤311和步骤316中的处理操作。在本实施例中，处理单元802根据存储单元803中存储的计算机指令，使得用户面数据处理网元实现本申请第一实施例到第六实施例中用户面数据处理网元执行的操作。
具体的，在一种可能的实现方式中，处理单元802利用收发单元801获取来自数据分析网元的至少一个特征集合的信息，其中，所述至少一个特征集合的信息中的每个特征集合的信息对应至少一种业务类型或至少一种执行规则；处理单元802根据所述至少一个特征集合的信息获取所述用户面数据的特征参数；处理单元802利用收发单元801向所述数据分析网元发送所述特征参数；处理单元802利用收发单元801获取来自所述数据分析网元的所述特征参数的响应结果；处理单元802根据所述响应结果获取所述用户面数据关联的业务类型或所述用户面数据关联的执行规则。
在一种可能的实现方式中，处理单元802利用收发单元80根据所述响应结果获取来自策略控制网元的所述用户面数据关联的所述执行规则；或者，处理单元802利用收发单元801根据所述响应结果获取来自所述数据分析网元的所述用户面数据关联的执行规则。
在一种可能的实现方式中，处理单元802根据所述用户面数据关联的业务类型或所述用户面数据关联的执行规则处理所述用户面数据。
在一种可能的实现方式中，处理单元802根据所述用户面数据关联的执行规则中指示的业务优先级信息转发所述用户面数据；或者，处理单元802根据所述用户面数据关联的业务类型或所述用户面数据关联的执行规则为所述用户面数据增加业务类型的标签信息；或者，处理单元802根据所述用户面数据关联的业务类型或所述用户面数据关联的执行规则为所述用户面数据增加调度优先级信息；或者，处理单元802根据所述用户面数据关联的业务类型或所述用户面数据关联的执行规则对所述用户面数据进行计费统计。
在一种可能的实现方式中，处理单元802确定所述用户面数据的目的地址为终端设备的地址且所述终端设备处于空闲状态；处理单元802根据所述用户面数据关联的业务类型或所述用户面数据关联的执行规则利用收发单元801向会话管理网元发送所述终端设备的寻呼优先级信息。
在一种可能的实现方式中，处理单元802根据部分特征从所述至少一个特征集合的信息中选择部分特征集合的信息；处理单元802获取所述用户面数据的特征参数，所述特征参数对应所述部分特征集合的信息。
在一种可能的实现方式中，处理单元802根据所述至少一个特征集合的信息从其他用户面数据处理网元或控制面网元获取所述用户面数据关联的信息；处理单元802根据所述至少一个特征集合的信息和所述关联的信息获取所述用户面数据的特征参数。
当数据分析装置为数据分析网元时，收发单元801用于向用户面数据处理网元发送至少一个特征集合的信息以及接收来自所述用户面数据处理网元的用户面数据的特征参数，其中，所述至少一个特征集合的信息中的每个特征集合的信息对应至少一种业务类型或者至少一种执行规则，所述至少一个特征集合的信息包括所述特征参数对应的特征集合的信息；处理单元802用于根据所述特征参数确定所述用户面数据关联的业务类型信息或者所述用户面数据关联的执行规则信息。
在一种可能的实现方式中，所述处理单元802还用于根据部分特征从预先获取的特征集合的信息中选择所述至少一个特征集合的信息。
其中，所述部分特征包括数据网络名称和/或数据网络对应的用户面功能网元的标识信息。所述处理单元802具体用于从所述预先获取的特征集合的信息中选择与所述数据网络名称和/或所述数据网络对应的用户面功能网元的标识信息一致的特征集合的信息。
在一种可能的实现方式中，所述处理单元802具体用于根据所述特征参数及所述特征参数对应的匹配算法确定所述用户面数据关联的业务类型信息或者所述用户面数据关联的执行规则信息。
其中，所述用户面数据关联的业务类型信息包括所述用户面数据关联的业务类型或业务类型的指示信息。
在一种可能的实现方式中，所述收发单元801还用于向所述用户面数据处理网元发送所述特征参数的响应结果，所述响应结果包括所述用户面数据关联的业务类型信息或者所述用户面数据关联的执行规则信息。
在一种可能的实现方式中，所述处理单元802还用于根据训练数据获取所述至少一个特征集合的信息；或，所述收发单元801还用于接收来自特征提取网元的所述至少一个特征集合的信息。
在一种可能的实现方式中，所述特征集合的信息为特征索引的集合，所述特征参数为特征向量。
在本实施例中，收发单元801还用于实现本申请第一实施例到第六实施例中数据分析网元与外部网元的内容的收发操作。处理单元802还用于实现本申请第一实施例到第六实施例中数据分析网元内部数据或者信令的处理操作,例如，处理单元802用于实现第二实施例中步骤302、步骤306和步骤314中的处理操作。
在本实施例中，处理单元802根据存储单元803中存储的计算机指令，使得数据分析 网元实现本申请第一实施例到第六实施例中数据分析网元执行的操作。
具体的，在一种可能的实现方式中，处理单元802利用收发单元801向用户面数据处理网元发送至少一个特征集合的信息，其中，所述至少一个特征集合的信息中的每个特征集合的信息对应至少一种业务类型或者至少一种执行规则；处理单元802利用收发单元801接收来自所述用户面数据处理网元的用户面数据的特征参数，其中，所述至少一个特征集合的信息包括所述特征参数对应的特征集合的信息；处理单元802根据所述特征参数确定所述用户面数据关联的业务类型信息或者所述用户面数据关联的执行规则信息。
在一种可能的实现方式中，处理单元802根据部分特征从预先获取的特征集合的信息中选择所述至少一个特征集合的信息。
在一种可能的实现方式中，所述部分特征包括数据网络名称和/或数据网络对应的用户面功能网元的标识信息；处理单元802从所述预先获取的特征集合的信息中选择与所述数据网络名称和/或所述数据网络对应的用户面功能网元的标识信息一致的特征集合的信息。
在一种可能的实现方式中，处理单元802根据所述特征参数及所述特征参数对应的匹配算法确定所述用户面数据关联的业务类型信息或者所述用户面数据关联的执行规则信息。
在一种可能的实现方式中，处理单元802利用收发单元801向所述用户面数据处理网元发送所述特征参数的响应结果，所述响应结果包括所述用户面数据关联的业务类型信息或者所述用户面数据关联的执行规则信息。
当数据分析装置为策略控制网元时，所述处理单元802用于通过所述收发单元801获取来自数据分析网元的至少一个特征集合的信息，其中，所述至少一个特征集合的信息中的每个特征集合的信息对应至少一种业务类型或至少一种执行规则；所述处理单元802还用于通过所述收发单元801向用户面数据处理网元发送所述至少一个特征集合的信息中的至少部分特征集合的信息。
在一种可能的实现方式中，所述处理单元802还用于根据部分特征从所述至少一个特征集合的信息中选择所述至少部分特征集合的信息。
其中，所述部分特征包括数据网络名称和/或数据网络对应的用户面功能网元的标识信息。所述处理单元具体用于从所述至少一个特征集合的信息中选择与所述数据网络名称和/或所述数据网络对应的用户面功能网元的标识信息一致的特征集合的信息。
在一种可能的实现方式中，所述收发单元801还用于向所述用户面数据处理网元发送所述至少部分特征集合的信息对应的业务类型信息。
在一种可能的实现方式中，所述收发单元801还用于向所述用户面数据处理网元发送所述至少部分特征集合的信息对应的执行规则。
在一种可能的实现方式中，所述收发单元801还用于获取来自所述数据分析网元的所述至少部分特征集合的信息对应的业务类型信息；所述处理单元802还用于根据所述获取的业务类型信息生成所述至少部分特征集合的信息对应的执行规则；所述收发单元801还用于向所述用户面数据处理网元发送所述至少部分特征集合的信息对应的执行规则。
在一种可能的实现方式中，所述特征集合的信息为特征索引的集合，所述特征参数为特征向量。
在本实施例中，收发单元801还用于实现本申请第一实施例到第六实施例中策略控制网元与外部网元的内容的收发操作。处理单元802还用于实现本申请第一实施例到第六实施例中策略控制网元内部数据或者信令的处理操作,例如，处理单元802用于实现第二实施例中步骤308的处理操作。
在本实施例中，处理单元802根据存储单元803中存储的计算机指令，使得策略控制网元实现本申请第一实施例到第六实施例中策略控制网元执行的操作。
具体的，在一种可能的实现方式中，处理单元802利用收发单元801获取来自数据分析网元的至少一个特征集合的信息，其中，所述至少一个特征集合的信息中的每个特征集合的信息对应至少一种业务类型或至少一种执行规则；处理单元802利用收发单元801向用户面数据处理网元发送所述至少一个特征集合的信息中的至少部分特征集合的信息。
在一种可能的实现方式中，处理单元802根据部分特征从所述至少一个特征集合的信息中选择所述至少部分特征集合的信息。
在一种可能的实现方式中，所述部分特征包括数据网络名称和/或数据网络对应的用户面功能网元的标识信息；处理单元802从所述至少一个特征集合的信息中选择与所述数据网络名称和/或所述数据网络对应的用户面功能网元的标识信息一致的特征集合的信息。
在一种可能的实现方式中，处理单元802利用收发单元801向所述用户面数据处理网元发送所述至少部分特征集合的信息对应的业务类型信息。
在一种可能的实现方式中，处理单元802利用收发单元801向所述用户面数据处理网元发送所述至少部分特征集合的信息对应的执行规则。
在一种可能的实现方式中，处理单元802利用收发单元801获取来自所述数据分析网元的所述至少部分特征集合的信息对应的业务类型信息；处理单元802根据所述获取的业务类型信息生成所述至少部分特征集合的信息对应的执行规则；处理单元802利用收发单元801向所述用户面数据处理网元发送所述至少部分特征集合的信息对应的执行规则。
在本申请的各实施例中，为了方面理解，进行了多种举例说明。然而，这些例子仅仅是一些举例，并不意味着是实现本申请的最佳实现方式。
在本申请的各实施例中，为了方便的描述，采用了请求消息，响应消息以及其他各种消息的名称。然而，这些消息仅仅是以举例方式说明需要携带的内容或者实现的功能，消息的具体名称并不对本申请的做出限定，例如：还可以是第一消息，第二消息，第三消息等。这些消息可以是具体的一些消息，可以是消息中的某些字段。这些消息还可以代表各种服务化操作。
上述实施例中，可以全部或部分地通过软件、硬件、固件或者其任意组合来实现。当使用软件实现时，可以全部或部分地以计算机程序产品的形式实现。所述计算机程序产品可以包括一个或多个计算机指令。在计算机上加载和执行所述计算机程序指令时，全部或部分地产生按照本申请实施例所述的流程或功能。所述计算机可以是通用计算机、专用计算机、计算机网络或者其他可编程装置。所述计算机指令可以存储在计算机可读存储介质中，或者从一个计算机可读存储介质向另一个计算机可读存储介质传输，例如，所述计算机指令可以从一个网站站点、计算机、服务器或数据中心通过有线(例如同轴电缆、光纤、数字用户(DSL))或无线(例如红外、无线、微波等)方式向另一个网站站点、计算机、 服务器或数据中心进行传输。所述计算机可读存储介质可以是计算机能够存取的任何可用介质或者是包含一个或多个可用介质集成的服务器、数据中心等数据存储设备。所述可用介质可以是磁性介质(例如，软盘、硬盘、磁盘)、光介质(例如，DVD)、或者半导体介质(例如，固态硬盘(solid state disk，SSD))等。
本领域普通技术人员可以意识到，结合本文中所公开的实施例描述的各示例的单元及算法步骤，能够以电子硬件、或者计算机软件和电子硬件的结合来实现。这些功能究竟以硬件还是软件方式来执行，取决于技术方案的特定应用和设计约束条件。专业技术人员可以对每个特定的应用来使用不同方法来实现所描述的功能，但是这种实现不应认为超出本申请的范围。
所属领域的技术人员可以清楚地了解到，为描述的方便和简洁，上述描述的系统、装置和单元的具体工作过程，可以参考前述方法实施例中的对应过程，在此不做限定。
在本申请所提供的几个实施例中，应该理解到，所揭露的系统、装置和方法，可以通过其它的方式实现。例如，以上所描述的装置实施例仅仅是示意性的，例如，所述单元的划分，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式，例如多个单元或组件可以结合或者可以集成到另一个系统，或一些特征可以忽略，或不执行。另一点，所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口，装置或单元的间接耦合或通信连接，可以是电性，机械或其它的形式。
所述作为分离部件说明的单元可以是或者也可以不是物理上分开的，作为单元显示的部件可以是或者也可以不是物理单元，即可以位于一个地方，或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。
另外，在本申请各个实施例中的各功能单元可以集成在一个处理单元中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个单元中。
所述功能如果以软件功能单元的形式实现并作为独立的产品销售或使用时，可以存储在一个计算机可读取存储介质中。基于这样的理解，本申请的技术方案本质上或者说对现有技术做出贡献的部分或者该技术方案的部分可以以软件产品的形式体现出来，该计算机软件产品存储在一个存储介质中，包括若干指令用以使得一台计算机设备(可以是个人计算机，服务器，或者网络设备等)执行本申请各个实施例所述方法的全部或部分步骤。而前述的存储介质包括：U盘、移动硬盘、只读存储器、随机存取存储器、磁碟或者光盘等各种可以存储程序代码的介质。
以上所述，仅为本申请的具体实施方式，但本申请的保护范围并不局限于此，任何熟悉本技术领域的技术人员在本申请揭露的技术范围内，可轻易想到变化或替换，都应涵盖在本申请的保护范围之内。因此，本申请的保护范围应所述以权利要求的保护范围为准。

Claims (27)

  1. 一种数据分析方法，其特征在于，包括：
    用户面数据处理网元获取来自数据分析网元的至少一个特征集合的信息，其中，所述至少一个特征集合的信息中的每个特征集合的信息对应至少一种业务类型或至少一种执行规则；
    所述用户面数据处理网元根据所述至少一个特征集合的信息获取用户面数据的特征参数；
    所述用户面数据处理网元向所述数据分析网元发送所述特征参数；
    所述用户面数据处理网元获取来自所述数据分析网元的所述特征参数的响应结果；
    所述用户面数据处理网元根据所述响应结果获取所述用户面数据关联的业务类型或所述用户面数据关联的执行规则。
  2. 根据权利要求1所述的方法，其特征在于，所述用户面数据处理网元根据所述响应结果获取所述用户面数据关联的业务类型或所述用户面数据关联的执行规则，包括：
    所述用户面数据处理网元根据所述响应结果获取来自策略控制网元的所述用户面数据关联的所述执行规则；或者
    所述用户面数据处理网元根据所述响应结果获取来自所述数据分析网元的所述用户面数据关联的执行规则。
  3. 根据权利要求1或2所述的方法，其特征在于，所述方法还包括：
    所述用户面数据处理网元根据所述用户面数据关联的业务类型或所述用户面数据关联的执行规则处理所述用户面数据。
  4. 根据权利要求3所述的方法，其特征在于，所述用户面数据处理网元根据所述用户面数据关联的业务类型或所述用户面数据关联的执行规则处理所述用户面数据，包括：
    所述用户面数据处理网元根据所述用户面数据关联的执行规则中指示的业务优先级信息转发所述用户面数据；或者
    所述用户面数据处理网元根据所述用户面数据关联的业务类型或所述用户面数据关联的执行规则为所述用户面数据增加业务类型的标签信息；或者
    所述用户面数据处理网元根据所述用户面数据关联的业务类型或所述用户面数据关联的执行规则为所述用户面数据增加调度优先级信息；或者，
    所述用户面数据处理网元根据所述用户面数据关联的业务类型或所述用户面数据关联的执行规则对所述用户面数据进行计费统计。
  5. 根据权利要求3所述的方法，其特征在于，所述用户面数据处理网元根据所述用户面数据关联的业务类型或所述用户面数据关联的执行规则处理所述用户面数据，包括：
    所述用户面数据处理网元确定所述用户面数据的目的地址为终端设备的地址且所述终端设备处于空闲状态；
    所述用户面数据处理网元根据所述用户面数据关联的业务类型或所述用户面数据关联的执行规则向会话管理网元发送所述终端设备的寻呼优先级信息。
  6. 根据权利要求1或3或4或5所述的方法，其特征在于，所述响应结果包括所述用户面数据关联的业务类型信息和/或所述用户面数据关联的执行规则信息。
  7. 根据权利要求1-6中任意一项所述的方法，其特征在于，所述用户面数据处理网元 根据所述至少一个特征集合的信息获取所述用户面数据的特征，包括：
    所述用户面数据处理网元根据部分特征从所述至少一个特征集合的信息中选择部分特征集合的信息；
    所述用户面数据处理网元获取所述用户面数据对应所述部分特征集合的信息的特征参数。
  8. 根据权利要求7所述的方法，其特征在于，所述部分特征包括包括所述用户面数据的互联网协议IP五元组信息。
  9. 根据权利要求1-8中任意一项所述的方法，其特征在于，所述方法还包括：
    所述用户面数据处理网元根据所述至少一个特征集合的信息从其他用户面数据处理网元或控制面网元获取所述用户面数据关联的信息；
    所述用户面数据处理网元根据所述至少一个特征集合的信息获取所述用户面数据的特征参数包括：
    所述用户面数据处理网元根据所述至少一个特征集合的信息和所述关联的信息获取所述用户面数据的特征参数。
  10. 根据权利要求1中任意一项所述的方法，所述用户面数据处理网元包括特征提取单元，
    所述用户面数据处理网元根据所述至少一个特征集合的信息获取所述用户面数据的特征参数，包括：
    所述特征提取单元根据所述至少一个特征集合的信息中的至少部分特征集合的信息获取所述用户面数据的特征参数。
  11. 根据权利要求1-10中任意一项所述的方法，其特征在于，所述特征集合的信息为特征索引的集合。
  12. 根据权利要求1-11中任意一项所述的方法，其特征在于，所述特征参数为特征值的集合。
  13. 根据权利要求12所述的方法，其特征在于，所述特征参数为特征向量。
  14. 一种数据分析的装置，其特征在于，包括：
    收发单元，用于获取来自数据分析网元的至少一个特征集合的信息，其中，所述至少一个特征集合的信息中的每个特征集合的信息对应至少一种业务类型或至少一种执行规则；
    处理单元，用于根据所述至少一个特征集合的信息获取用户面数据的特征参数；
    所述收发单元还用于向所述数据分析网元发送所述特征参数以及获取来自所述数据分析网元的所述特征参数的响应结果；
    所述处理单元还用于根据所述响应结果获取所述用户面数据关联的业务类型或所述用户面数据关联的执行规则。
  15. 根据权利要求14所述的装置，其特征在于，所述处理单元还用于根据所述响应结果获取所述用户面数据关联的业务类型或所述用户面数据关联的执行规则，包括：
    所述处理单元用于根据所述响应结果获取来自策略控制网元的所述用户面数据关联的所述执行规则，或者，用于根据所述响应结果获取来自所述数据分析网元的所述用户面数据关联的执行规则。
  16. 根据权利要求14或15所述的装置，其特征在于，
    所述处理单元还用于根据所述用户面数据关联的业务类型或所述用户面数据关联的执行规则处理所述用户面数据。
  17. 根据权利要求16所述的装置，其特征在于，所述处理单元用于根据所述用户面数据关联的业务类型或所述用户面数据关联的执行规则处理所述用户面数据，包括：
    所述处理单元用于根据所述用户面数据关联的执行规则中指示的业务优先级信息转发所述用户面数据；或者
    所述处理单元用于根据所述用户面数据关联的业务类型或所述用户面数据关联的执行规则为所述用户面数据增加业务类型的标签信息；或者
    所述处理单元用于根据所述用户面数据关联的业务类型或所述用户面数据关联的执行规则为所述用户面数据增加调度优先级信息；或者，
    所述处理单元用于根据所述用户面数据关联的业务类型或所述用户面数据关联的执行规则对所述用户面数据进行计费统计。
  18. 根据权利要求16所述的装置，其特征在于，所述处理单元用于根据所述用户面数据关联的业务类型或所述用户面数据关联的执行规则处理所述用户面数据，包括：
    所述处理单元用于确定所述用户面数据的目的地址为终端设备的地址且所述终端设备处于空闲状态，根据所述用户面数据关联的业务类型或所述用户面数据关联的执行规则向会话管理网元发送所述终端设备的寻呼优先级信息。
  19. 根据权利要求14或16或17或18所述的装置，其特征在于，所述响应结果包括所述用户面数据关联的业务类型信息和/或所述用户面数据关联的执行规则信息。
  20. 根据权利要求14-19中任意一项所述的装置，其特征在于，所述处理单元用于根据所述至少一个特征集合的信息获取所述用户面数据的特征参数，包括：
    所述处理单元用于根据部分特征从所述至少一个特征集合的信息中选择部分特征集合的信息；
    所述处理单元用于获取所述用户面数据对应所述部分特征集合的信息的特征参数。
  21. 根据权利要求20所述的装置，其特征在于，所述部分特征包括包括所述用户面数据的互联网协议IP五元组信息。
  22. 根据权利要求14-21中任意一项所述的装置，其特征在于，
    所述收发单元还用于根据所述至少一个特征集合的信息从其他用户面数据处理网元或控制面网元获取所述用户面数据关联的信息；
    所述处理单元用于根据所述至少一个特征集合的信息获取所述用户面数据的参数，包括：
    所述处理单元用于根据所述至少一个特征集合的信息和所述关联的信息获取所述用户面数据的特征参数。
  23. 根据权利要求14-22中任意一项所述的装置，其特征在于，所述特征集合的信息为特征索引的集合。
  24. 根据权利要求14-23中任意一项所述的装置，其特征在于，所述特征参数为特征值的集合。
  25. 根据权利要求24所述的方法，其特征在于，所述特征参数为特征向量。
  26. 一种数据分析的装置，其特征在于，包括处理器、存储器和收发器：
    所述存储器，用于存储计算机指令；
    所述处理器，用于根据所述存储单元中存储的计算机指令执行如下操作：
    利用所述收发器获取来自数据分析网元的至少一个特征集合的信息，其中，所述至少一个特征集合的信息中的每个特征集合的信息对应至少一种业务类型或至少一种执行规则；
    根据所述至少一个特征集合的信息获取用户面数据的特征参数；
    利用所述收发器向所述数据分析网元发送所述特征参数；
    利用所述收发器获取来自所述数据分析网元的所述特征参数的响应结果；
    根据所述响应结果获取所述用户面数据关联的业务类型或所述用户面数据关联的执行规则。
  27. 一种计算机存储介质，所述计算机可读存储介质中存储有指令，当其在计算机上运行时，使得计算机执行所述权利要求1-13中任意一项所述的方法。
PCT/CN2018/104115 2017-09-30 2018-09-05 数据分析方法和装置 Ceased WO2019062497A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
EP18860414.4A EP3684139B1 (en) 2017-09-30 2018-09-05 Data analysis method and device
US16/833,195 US11552856B2 (en) 2017-09-30 2020-03-27 Data analytics method and apparatus

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201710915784.4 2017-09-30
CN201710915784.4A CN109600243B (zh) 2017-09-30 2017-09-30 数据分析方法和装置

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US16/833,195 Continuation US11552856B2 (en) 2017-09-30 2020-03-27 Data analytics method and apparatus

Publications (1)

Publication Number Publication Date
WO2019062497A1 true WO2019062497A1 (zh) 2019-04-04

Family

ID=65900571

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2018/104115 Ceased WO2019062497A1 (zh) 2017-09-30 2018-09-05 数据分析方法和装置

Country Status (4)

Country Link
US (1) US11552856B2 (zh)
EP (1) EP3684139B1 (zh)
CN (1) CN109600243B (zh)
WO (1) WO2019062497A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US11012323B2 (en) * 2017-09-30 2021-05-18 Huawei Technologies Co., Ltd. Feature parameter obtaining method and apparatus

Families Citing this family (18)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111901135B (zh) * 2019-05-05 2021-10-15 华为技术有限公司 一种数据分析方法及装置
GB2583771B (en) * 2019-05-10 2022-06-15 Samsung Electronics Co Ltd Improvements in and relating to data analytics in a telecommunication network
CN112104469B (zh) 2019-06-17 2022-07-29 华为技术有限公司 数据处理方法及装置
CN110445793B (zh) * 2019-08-13 2021-04-06 四川长虹电器股份有限公司 一种拥有节点线程级别无冗余计算的分析引擎的分析方法
KR102447806B1 (ko) * 2019-09-06 2022-09-27 삼성전자주식회사 무선 통신 시스템에서 네트워크 자동화를 위한 장치 및 방법
CN110569288A (zh) * 2019-09-11 2019-12-13 中兴通讯股份有限公司 一种数据分析方法、装置、设备和存储介质
CN112584417B (zh) * 2019-09-27 2023-05-09 中国移动通信有限公司研究院 无线业务质量确定方法、装置、计算机设备和存储介质
EP4027584B1 (en) * 2019-09-30 2025-12-10 Huawei Technologies Co., Ltd. Device information determination method and apparatus, and system
US12363558B2 (en) * 2020-01-02 2025-07-15 Telefonaktiebolaget Lm Ericsson (Publ) Association with a network data analytics function
CN115004653B (zh) * 2020-02-07 2025-05-06 华为技术有限公司 一种数据分析方法、装置及系统
CN117320034A (zh) * 2020-04-29 2023-12-29 华为技术有限公司 一种通信方法、装置及系统
US12388877B2 (en) 2020-07-15 2025-08-12 Telefonaktiebolaget Lm Ericsson (Publ) Packet flow descriptor provisioning
CN114124724A (zh) * 2020-08-11 2022-03-01 中国电信股份有限公司 用户行为分析方法、装置、nwdaf以及存储介质
CN112016967B (zh) * 2020-08-28 2023-09-05 中国银联股份有限公司 一种交易数据处理方法与装置
CN114826867B (zh) 2021-01-28 2023-11-17 华为技术有限公司 处理数据的方法、装置、系统及存储介质
CN112954057A (zh) * 2021-02-08 2021-06-11 四川奥诚科技有限责任公司 基于VoLTE网元的数据解析方法、装置、设备及介质
CN116684918A (zh) * 2022-02-18 2023-09-01 大唐移动通信设备有限公司 一种数据传输方法及装置
CN117114123A (zh) * 2022-05-11 2023-11-24 中国电信股份有限公司 网络数据的分析方法、装置、电子设备以及存储介质

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101854652A (zh) * 2010-06-23 2010-10-06 天元莱博(北京)科技有限公司 一种电信网络业务性能监控系统
CN103490849A (zh) * 2012-06-13 2014-01-01 华为技术有限公司 分析信令流量的方法及装置
US20160099857A1 (en) * 2014-10-07 2016-04-07 At&T Intellectual Property I, Lp Method and System to Monitor a Network

Family Cites Families (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9392463B2 (en) * 2012-12-20 2016-07-12 Tarun Anand System and method for detecting anomaly in a handheld device
CN103473291B (zh) * 2013-09-02 2017-01-18 中国科学院软件研究所 一种基于隐语义概率模型的个性化服务推荐系统及方法
US11166334B2 (en) * 2016-07-01 2021-11-02 Idac Holdings, Inc. Methods for supporting session continuity on per-session basis
US10856265B2 (en) * 2016-07-05 2020-12-01 Lg Electronics Inc. Method for selecting resource operation preferred by user in wireless communication system and device for same
WO2018070436A1 (en) * 2016-10-11 2018-04-19 Nec Corporation Method, session management function node, user plane function node, and user equipment for session management parameters maintenance and computer readable recording medium therein
KR102355678B1 (ko) * 2017-05-08 2022-01-26 삼성전자 주식회사 이동 통신 시스템에서의 QoS(Quality Of Service) Flow의 설정 방법 및 장치
US10779254B2 (en) * 2017-08-16 2020-09-15 Electronics And Telecommunications Research Institute Service request method for 5G local service

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101854652A (zh) * 2010-06-23 2010-10-06 天元莱博(北京)科技有限公司 一种电信网络业务性能监控系统
CN103490849A (zh) * 2012-06-13 2014-01-01 华为技术有限公司 分析信令流量的方法及装置
US20160099857A1 (en) * 2014-10-07 2016-04-07 At&T Intellectual Property I, Lp Method and System to Monitor a Network

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
HUAWEI: "Discussion about Big Data Driven Network Architecture", 3GPP SA WG2 MEETING #121 S 2-173192, 19 May 2017 (2017-05-19), XP051281694 *
See also references of EP3684139A4

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US11012323B2 (en) * 2017-09-30 2021-05-18 Huawei Technologies Co., Ltd. Feature parameter obtaining method and apparatus

Also Published As

Publication number Publication date
EP3684139A4 (en) 2020-08-19
EP3684139B1 (en) 2022-01-19
US11552856B2 (en) 2023-01-10
CN109600243B (zh) 2021-09-03
US20200228413A1 (en) 2020-07-16
EP3684139A1 (en) 2020-07-22
CN109600243A (zh) 2019-04-09

Similar Documents

Publication Publication Date Title
WO2019062497A1 (zh) 数据分析方法和装置
US11277324B2 (en) Determining an execution policy of a service
US11012323B2 (en) Feature parameter obtaining method and apparatus
CN109600241B (zh) 数据分析方法和数据分析装置
US12224905B2 (en) Communication method and communication apparatus
CN111436057B (zh) 一种会话管理方法及装置
WO2021135295A1 (zh) 建立连接和获取中继服务代码的方法和通信装置
WO2019158102A1 (zh) 一种确定QoS描述信息的方法和装置
CN109756423A (zh) 策略更新的方法和装置
WO2015043395A1 (zh) 业务接入方法、用户设备和无线控制器
US12238154B2 (en) Multicast session establishment method and network device
CN112087777A (zh) 一种mdbv的确定方法、装置及系统
US12603846B2 (en) Systems and methods for edge-to-edge quality of service flow control in network slices
WO2023011407A1 (zh) 通信方法和装置
US20250380191A1 (en) Systems and methods for supporting network slice services using transport devices
WO2022237857A1 (zh) 确定安全保护开启方式的方法、通信方法及通信装置
WO2018120246A1 (zh) 一种数据传输方法及相关网元
CN109525979B (zh) 一种上下文信息管理方法及装置

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 18860414

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

ENP Entry into the national phase

Ref document number: 2018860414

Country of ref document: EP

Effective date: 20200414