OCNWDAF Analytics

7 OCNWDAF Analytics

The OCNWDAF assists in collecting and analysing data in a 5G network. An OCNWDAF consumer can avail analytics information or reports for various events in the network. The OCNWDAF allows the Communications Service providers (CSPs) to efficiently monitor, manage, automate, and optimise their network operations by analysing the data collected across the network.

The consumers can subscribe (or unsubscribe) to the OCNWDAF to obtain specific analytics reports as a one-time event or periodically get notified when a defined event is detected. The analytics information provided by the OCNWDAF is either statistical information on past events or predictive information which can be used to balance the resources in the network. The OCNWDAF currently supports NFs as data producers but the data consumers can be not only be 5G NFs but AFs and OAM can also be consumers of analytics information.

Listed below are the type of analytics reports that OCNWDAF can provide:

Historical analytics
Future analytics
Reports when a thresholds are crossed

The Analytics Subscription Service and Analytics Information Service are used for obtaining different analytics reports. The Analytics Subscription Service obtains periodic reports based on future or current events, threshold and abnormal event reports. The Analytics Information Service obtains historic statistical reports and prediction reports.

Analytics Request

A consumer analytics request (subscription or information request) to the OCNWDAF contains the following information:

Analytics ID(s): Analytics ID identifies the requested type of analytics.

Analytics Filter Information (optional)
Target of Analytics Reporting: The target indicates the object(s) for which Analytics information is requested. It includes entities such as a specific UE or a group of UE(s) or all UEs.

Analytics Reporting Information with the following parameters:
- In the case of Analytics Subscription requests, event reporting parameters are defined as per 3GPP TS 29.520.
- In the case of Analytics Subscription requests, Reporting Thresholds are defined, which are the conditions on the level of each requested analytics. When the threshold is crossed, the OCNWDAF notifies the consumer. The specified conditions may include rules like "ASCENDING", "DESCENDING", or "CROSSED". The default rule is "CROSSED" if no matching rule is provided.
Analytics target period: Time interval which includes both start and end time (in UTC format). The target period specified can be either the past time or future time.
- Analytics target period specified in past time indicates an analytics statistics request.
- Analytics target period specified in the future time indicates an analytics prediction request.
By setting start time and end time to the same value, the consumer of the analytics can request analytics or subscribe to analytics for a specific time rather than for a time interval.

Preferred level of accuracy of the analytics (for example, low or high).
For Analytics Information requests, the time when analytics information is required can be specified. If the time is reached and no analytics are received the consumer does not wait for the analytics information any longer and the OCNWDAF sends an error response to the consumer.
The maximum number of objects (optional). This specifies the number of objects in a list of analytics for each request.
Maximum number of SUPIs is an optional parameter that specifies the number of relevant SUPIs in the analytics object to be returned in the analytics response. When this parameter is not specified, the OCNWDAF returns the stipulated number of relevant SUPIs in the analytics object.

Analytics Report

The OCNWDAF provides the following analytics information to the consumer:

Analytics information based on specified Analytics ID and target time.
Notification Correlation Information for Analytics Subscription requests.
The OCNWDAF provides the following additional information:
- Timestamp of analytics generation. The analytics consumer can determine the relevance of analytics information based on the timestamp received.
- Validity period, the time until which the analytics information is valid.
- Probability assertion: This indicates confidence in the prediction. The confidence is expressed as a value based on definition of parameters "Preferred level of accuracy" and "Analytics Target Period" in the consumer request and the data availability with OCNWDAF. The OCNWDAF returns a value of zero confidence if sufficient data is not collected to match the requested accuracy level within the analytics target time. If the analytics target time is not specified, the OCNWDAF waits till adequate data is collected then provides a response or notification.
  
  Note:
  Statistical analytics does not contain this parameter as confidence in prediction is not applicable for this type of analytics.

The OCNWDAF provides the following analytics:

Slice Load level analytics
UE Mobility analytics
UE Abnormal Behaviour analytics (Unexpected UE Location)
NF Load Level analytics
Network Performance analytics
User Data Congestion analytics (only statistical analytics)
QoS Sustainability analytics
Capex Optimization analytics

7.1 QoS (Quality of Service) Sustainability Analytics

Overview

Consumers can derive QoS (Quality of Service) Sustainability Analytics from the OCNWDAF. This analytics information includes QoS change statistics in a target area (for a specified period). OCNWDAF provides both statistical and predictive analytics. Predictive analytics includes information about anticipated QoS changes at a future time in a target area. Consumers can obtain analytics information for a slice by specifying the S-NSSAI of the slice in the analytics request. Users can subscribe for a one-time analytics report or opt for continuous analytics reports from the OCNWDAF.

Note:

QoS Analytics subscribers must include the QoSSustainability flag in their subscriptions and analytics requests.

QoS Sustainability Analytics Workflow

The following diagram depicts the QoS Sustainability Analytics workflow:

Figure 7-1 QoS Sustainability Analytics Workflow

The following procedure describes the workflow:

The NF consumer requests (Nnwdaf_Analyticsinfo_request) or subscribes to (Nnwdaf_AnalyticsSubscription_Subscribe) the OCNWDAF for QoS sustainability analytics information.
The OCNWDAF collects the required data from Operation, Administration, and Maintenance (OAM).The OCNWDAF subscribes to the OAM and continuously collects data from it (irrespective of an analytics request being present or absent). This data is stored in the Analytics DB.
The OCNWDAF derives the requested analytics from the CAP4C based on the conditions specified in the consumer request.
The OCNWDAF provides the QoS sustainability analytics to the consumer NF.

Consumer Request to OCNWDAF

A consumer request for QoS sustainability analytics contains the following information:

Analytics ID set to "QoS Sustainability”
Target of Analytics Reporting set to “any UE”
The following analytics filter information:
- QoS requirements (mandatory): Either "5QI (5G QoS Identifier)" or "QoS Resource Type."
  - 5QI: For standardized or preconfigured characteristics specified in 3GPP TS 23.501 (table 5.7.4-1), including Packet Delay Budget (PDB) and Packet Error Rate (PER) values.
- Location information (mandatory): Specified as AoI (Area of Interest,that is, a list of Tracking Areas (TAs) or cells) or a path of interest.
- S-NSSAI (optional)
The analytics target period that indicates the time over which the statistics are requested.
A reporting threshold (optional, and applicable only for analytics subscriptions). A matching direction can be provided along with the reporting threshold such as crossed, below, or above the default value.
A subscription request also contains the Notification Correlation ID and the Notification Target Address.

Data Collection from OAM

The OCNWDAF collects data the following data from the OAM to generate QoS sustainability analytics:

Table 7-1 Data Collection from OAM

Information	Source	Description
RAN UE Throughput	OAM (3GPP TS 28.554 [10])	Average UE bitrate in the cell (for transfers restricted by the air interface the payload data volume on RLC level per elapsed time unit on the air interface), per timeslot, per cell, per 5QI, and per S-NSSAI.
QoS Flow Retainability	OAM (3GPP TS 28.554 [10])	Number of abnormally released QoS flows during the time the QoS Flows were used per timeslot, per cell, per 5QI, and per S-NSSAI.

Note:

For a complete list of request parameters, see QoS Sustainability Analytics APIs.

Note:

The data types supported by OCNWDAF comply with the 3GPP specifications. For more information about the 3GPP data types, see 3GPP Technical Specification 29.520, Release 17, Network Data Analytics Services.

7.2 User Data Congestion Analytics

Overview

The OCNWDAF provides User Data Congestion Analytics (statistical and threshold) to consumers. It includes information about the congestion experienced during user data transfer over the user plane. The congestion information reported is specific to an Area of Interest (AoI) or an User Equipment (UE). If the consumer provides only the UEs ID, the OCNWDAF determines the AoI in which congestion analytics is required based on the ID. The OCNWDAF also provides congestion analytics for a slice based on the Single-Network Slice Selection Assistance Information (S-NSSAI). User can specify a threshold for the congestion level and the OCNWDAF reports congestion analytics when the threshold is breached. User can subscribe for one-time or continuous analytics reporting. Predictive analytics is not supported.

User Data Congestion Analytics Workflow

User can retrieve analytics information for a specific location or a UE.

The following diagram depicts how User Data Congestion analytics is retrieved:

Figure 7-2 User Data Congestion Analytics Workflow

The following procedure describes the workflow:

The NF consumer requests (Nnwdaf_Analyticsinfo_Request) or subscribes to (Nnwdaf_AnalyticsSubscription_Request) the OCNWDAF for User Data Congestion analytics information.
The OCNWDAF collects the required data from Operation, Administration, and Maintenance (OAM).The OCNWDAF subscribes to the OAM and continuously collects data from it (irrespective of an analytics request being present or absent). The collected data is stored in the Analytics DB.
The OCNWDAF derives the requested analytics from the CAP4C based on the conditions specified in the consumer request.
The OCNWDAF provides the User Data Congestion sustainability analytics to the consumer NF. (Nnwdaf_Analyticsinfo_Requestreponse or Nnwdaf_AnalyticsSubscription_Notify)

Consumer Request to OCNWDAF

A consumer request for user data congestion analytics contains the following information:

Analytics ID set to "User Data Congestion".
Target of Analytics Reporting containing single UE (SUPI) or "any UE".
Analytics Filter Information containing:
- Traffic type: The allowed options for "Traffic type" are Overall, GBR, or delay-critical GBR. This parameter is mandatory if the target is "any UE". For all other types of "Target of Analytics Reporting", this is an optional parameter.
- Area of Interest: It consists of a list of Tracking Areas (TAs) or Cells. This is a mandatory parameter if the "Target of Analytics Reporting" is set to "any UE". For all other types of "Target of Analytics Reporting", this is an optional parameter.
- Optional list of analytics subsets: This includes Area subset (a specific TA or cell ID within the AoI) and Analytics target period.
- The S-NSSAI to obtain congestion analytics in a network slice. This is an optional parameter.
Reporting Threshold (optional and applicable only for Analytics Subscriptions).
Analytics target period indicates the time period over which the statistics are requested.
The Notification Correlation ID and Notification Target Address are included in an analytics subscription.

Data Collection for User Data Congestion Analytics

The OCNWDAF collects data from AMF, OAM, UPF, and AF to generate user data congestion analytics.

Table 7-2 Data Collection from NF and OAM

Information Collected	Source	Description
UE Location	AMF	The UE location information that OCNWDAF uses to determine the Area of Interest.
Measurements	OAM	Performance Measurements that the OCNWDAF uses to determine congestion levels. Performance Measurements are related to information transfer over the user plane.
Throughput UL/DL	OAM	Average throughput Uplink (UL)/ (Downlink) DL over the measurement period.
Throughput UL/DL (peak)	OAM	Peak throughput UL/DL over the measurement period.

Request/Response Parameters

Listed below are the parameters in the user data congestion analytics both in request and response:

Table 7-3 Request Parameters

Parameter	Data Type	Presence	Description
eventSubscriptions	array(EventSubscription)	M	Specifies the subscribed event. The event parameter within the subscription is set to "USER DATA CONGESTION".
evtReq	ReportingInformation	O	Represents the reporting requirements of the event subscription. If this parameter is not provided, the default values for ReportingInformation is used. The notification method "notifMethod" within ReportingInformation takes precedence over the notificationMethod within EventSubscription.
notificationURI	Uri	C	The URI to which OCNWDAF sends the reports
eventNotifications	array(EventNotification)	C	Notifications about individual events This parameter is present if the immediate reporting indication in the "immRep" attribute within the "evtReq" attribute is set to true in the event subscription and the reports are available. This parameter is provided in response to the OCNWDAF subscription if immediatereport is true and the reports are available.
failEventReports	array(FailureEventInfo)	O	This parameter is provided by the OCNWDAF. When this parameter is available, it contains the event(s) for which the subscription is not successful, it also includes the failure reason(s). This parameter is populated and returned in the response for the error case.
consNfInfo	ConsumerNfInformation	O	Represents the analytics consumer NF information
notifCorrId	String	M	Notification correlation identifier. It is generated after a successful subscription and is returned in the response.
supportedFeatures	SupportedFeatures	C	This parameter is provided by the NF service consumer in the POST request for creating an OCNWDAF Event Subscription resource, and it is also provided in response to the corresponding request.

Note:

For a complete list of request parameters, see User Data Congestion Analytics APIs.

Note:

Output User Data Congestion Analytics

The following analytics information is obtained:

User Data Congestion Statistics

Table 7-4 User Data Congestion Analytics

Parameter	Data Type	Presence	Description
event	NwdafEvent	M	Event that is notified.
start	DateTime	O	UTC time indicating the start time of the observation period. The absence of this attribute means subscription at present time.
expiry	DateTime	O	It defines the expiration time after which the analytics information becomes invalid.
timeStampGen	DateTime	O	It defines the timestamp of analytics generation.
userDataCongInfos	Table 7-5	C	The user data congestion information.

UserDataCongestionInfo

Table 7-5 UserDataCongestionInfo

Attribute Name	Data Type	P	Description
networkArea	NetworkAreaInfo	M	The network area to which the subscription applies.
congestionInfo	CongestionInfo	M	The congestion information of the specific location.
snssai	Snssai	C	Identifies an S-NSSAI. Is present if the "snssais" was provided within the EventSubscription during the subscription for event notification procedure.

Type CongestionInfo

Table 7-6 CongestionInfo

Attribute Name	Data type	P	Description
congType	CongestionType	M	Identifies the congestion analytics type.
timeIntev	TimeWindow	M	Represents a start time and a stop time requested for the congestion information.
nsi	ThresholdLevel	M	Network status indication.
congestionDirection	String	M	Identifies if the report being generated is for Uplink or Downlink congestion.

Enumeration CongestionType

Table 7-7 CongestionType

Enumeration Value	Description
USER_PLANE	The congestion analytics type is user plane.

Note:

7.3 Network Performance Analytics

Overview

The OCNWDAF provides network performance analytics to the consumer. It includes information about resource consumption by the gNodeB (gNB), a key component in the 5G Radio Access Network (RAN), as well as key mobility performance indicators in the Area of Interest (AoI). The gNodeB is a base station in the 5G network. The network performance analytics provides information such as gNB status, gNB resource usage, communication, and mobility performance in the Area of Interest (AoI). It also provides statistics or predictions on the number of UEs in that AoI. The OCNWDAF can provide predictive or statistical analytics based on consumer requests.

The OCNWDAF GUI is enhanced to display the following information:

All available cells in a selected Tracking Area
For each cell, the user can view the following information:
- GNB Computing, Memory, and Disk Usage
- Session Success Ratio
- Hand Over (HO) Success Ratio
The user can set a threshold value, select a tracking area and a time interval in the GUI. All the available cells in the selected tracking area within the selected time interval and specified threshold value are displayed on the screen.

For more information, see Monitoring page on the OCNWDAF GUI.

Network Performance Analytics Workflow

The following diagram depicts the workflow of an analytics consumer request to OCNWDAF for Network Performance analytics:

Figure 7-3 Network Performance Analytics Workflow

The analytics consumer sends an analytics request to the OCNWDAF for Network Performance analytics. The analytics request is either an analytics subscription request (Nnwdaf_AnalyticsSubscription_Subscribe) or an analytics information request (Nnwdaf_AnalyticsInfo_Request). In the request, the Analytics ID is set to Network Performance. The request also specifies the target of analytics reporting, the filter (Area of Interest), reporting thresholds, and the analytics target period.
The OCNWDAF sends a discovery request to the NRF to obtain information about the AMFs present in the AMF region of the Area of Interest (AoI). The OCNWDAF subscribes to the NRF to get the NF load and status information about these AMFs.
The OCNWDAF subscribes to the OAM and continuously collects data irrespective of whether an analytics request is present or absent. The collected data is stored in the Analytics DB.
The OCNWDAF collects information about the number of UEs in the Area of Interest from the AMF using Namf_EventExposure_Subscribe service and the Target of Event Reporting specified in the analytics request.
The OCNWDAF derives the analytics information from the CAP4C and sends either a Nnwdaf_AnalyticsSubscription_Notify or Nnwdaf_AnalyticsInfo_Request response to the analytics consumer.
The OCNWDAF is notified about any change in the network performance, such as a change in gNB status, gNB resource usage, communication, and mobility performance in the AoI during the specified target period. The OAM detects the changes or changes in the NF load is reported by the NRF.
The OCNWDAF derives analytics from CAP4C based on the most recently collected data.
The OCNWDAF uses the Nnwdaf_AnalyticsSubscription_Notify to notify the analytics consumer about changes in the analytics information during the specified analytics target period or when reporting thresholds are breached.

Consumer Request to OCNWDAF

A consumer request for Network Performance analytics contains the following information:

The analytics ID is set to Network Performance.
The Target of Analytics Reporting: This can be a UE or internal group identifier of a group for which the analytics information is required. The group contains the UEs that are located in the Area of Interest during the time specified in the Analytics target period. The target of analytics reporting is set to "any UE" by default.
Analytics filter information consisting of the following:
- Area of Interest (a list of TA or Cells) specifying the focus area. This parameter is mandatory if the Target of Analytics Reporting parameter is set to "any UE". Providing the filter information is optional if the Target of Analytics Reporting parameter is set to any other value.
- Subset of analytics (optional)
Reporting threshold (optional)
Analytics target period, the period for which the statistics or predictions are required.
Maximum number of objects (optional)
In an analytics subscription request, the Notification Correlation Id and the Notification Target Address are included.

Request/Response Parameters

Listed below are the parameters in the network performance analytics request and response:

Table 7-8 Request/Response Parameters

Parameter	Data Type	Presence	Description
eventSubscriptions	array(EventSubscription)	M	Specifies the subscribed event. The event parameter within the subscription is set to "NETWORK_ PERFORMANCE".
evtReq	ReportingInformation	O	Represents the reporting requirements of the event subscription. If this parameter is not provided, the default values for ReportingInformation is used. The notification method "notifMethod" within ReportingInformation takes precedence over the notificationMethod within EventSubscription.
notificationURI	Uri	C	The URI to which OCNWDAF sends the reports.
eventNotifications	array(EventNotification)	C	Notifications about individual events. This parameter is present if the immediate reporting indication in the "immRep" attribute within the "evtReq" attribute is set to true in the event subscription and the reports are available. This parameter is provided in response to the OCNWDAF subscription if immediatereport is true and the reports are available.
failEventReports	array(FailureEventInfo)	O	This parameter is provided by theOCNWDAF. When this parameter is available, it contains the event(s) for which the subscription is not successful, it also includes the failure reason(s). This parameter is populated and returned in the response for the error case.
consNfInfo	ConsumerNfInformation	O	Represents the analytics consumer NF information.
notifCorrId	String	M	Notification correlation identifier. It is generated after a successful subscription and is returned in the response.
supportedFeatures	SupportedFeatures	C	This parameter is provided by the NF service consumer in the POST request for creating an OCNWDAF Event Subscription resource, and it is also provided in response to the corresponding request.

Note:

For a complete list of request parameters see, Network Performance Analytics APIs.
The data types supported by OCNWDAF comply with the 3GPP specifications. For more information about the 3GPP data types, see 3GPP Technical Specification 29.520, Release 17, Network Data Analytics Services.

Network Performance Data Collection

The OCNWDAF requires resource utilization and RAN status of the gNbs to generate Network Performance analytics. It also requires PDU session and UE Handover information. This information is obtained from OAM. The Network Performance ID also requires information on the number of UEs from the AMF. The OCNWDAF collects the needed information from the following sources to generate Network Performance analytics:

Table 7-9 Data Collection

Information Collected	Source	Description
gNodeB Resource Utilization (CPU, Memory, Storage)	OAM	Radio resource utilization information.
RAN Status	OAM	The status of the RAN.
PDU Session Success and Failure	OAM	Success or failure information of PDU sessions.
UE Handover Success and Failure	OAM	Success or failure information of UE handovers.
Number of UEs	AMF	Number of UEs in Area Report.

For more information see, Data Collection Services

Output Network Performance Analytics

The following types of analytics information is obtained:

Network Performance statistics
Network Performance predictions

The predictive analysis has an additional parameter, "Confidence", which indicates the accuracy of the prediction.

Table 7-10 Network Performance Analytics

Parameter	Data Type	Presence	Description
event	NwdafEvent	M	Event that is notified.
start	DateTime	O	UTC time indicating the start time of the observation period. The absence of this attribute means subscription at present time.
expiry	DateTime	O	It defines the expiration time after which the analytics information becomes invalid.
timeStampGen	DateTime	O	It defines the timestamp of analytics generation.
duration	Duration	C	It defines the time period for which the threshold was in the target state.
nwPerfs	array (NetworkPer fInfo)	C	The network performance information.

Type NetworkPerfInfo

Table 7-11 NetworkPerfInfo

Attribute Name	Data type	P	Description
networkArea	NetworkAreaInfo	M	The network area to which the subscription applies.
nwPerfType	NetworkPerfType	M	The type of network performance.
relativeRatio	SamplingRatio	O	The reported relative ratio expressed in percentage. Either the relativeRatio or absoluteNum is provided.
absoluteNum	Uinteger	O	The reported absolute number.
confidence	Uinteger	C	Indicates the confidence of the prediction. If sufficient data is unavailable during the specified analytics target time interval, the confidence value returned is "0".

Type NetworkAreaInfo

Table 7-12 NetworkAreaInfo

Attribute name	Data type	P	Cardinality	Description
ecgis	Array	O	0 to N	Is the list of E-UTRA cell identities.
ncgis	Array	O	0 to N	Is the list of NR cell identities
gRanNodeIds	Array	O	0 to N	Is the list of list of the NG-RAN nodes.
tais	Array	O	0 to N	Is the list of tracking area identities.

Enumeration NetworkPerfType

Table 7-13 NetworkPerfType

Enumeration value	Description
GNB_ACTIVE_RATIO	Indicates the ratio of active gNBs (up and running) to the total number of gNBs.
GNB_COMPUTING_USAGE	Indicates the gNodeBs computing resource usage.
GNB_MEMORY_USAGE	Indicates gNodeBs memory usage.
GNB_DISK_USAGE	Indicates gNodeBs disk usage.
NUM_OF_UE	Indicates the number of UEs.
SESS_SUCC_RATIO	Indicates the ratio of the successful PDU session setups to the total PDU session setup attempts.
HO_SUCC_RATIO	Indicates the ratio of successful handovers to the total handover attempts.

The analytics report consists of the following information:

Table 7-14 Analytics Report

Information	Description
Area subset	TA or Cell ID within the requested area of interest .
Analytics target period subset	Defined time window within the requested analytics target period.
gNB status information	Average ratio of gNBs that are up and running during the entire analytics target period in the area subset.
gNB resource usage	Average usage of assigned resources.
gNB resource usage for GBR traffic	Average usage of assigned resources (average, peak).
gNB resource usage for Delay-critical GBR traffic	Average usage of assigned resources (average, peak).
Number of UEs	Average number of UEs observed in the area subset.
Communication performance	Average ratio of successful PDU sessions setup.
Mobility performance	Average ratio of successful handovers.
Confidence (only for predictions)	Percentage accuracy of prediction.

Note:

7.4 Network Function (NF) Load Analytics

The OCNWDAF can provide NF load analytics information to the analytics consumer. The analytics generated by OCNWDAF can be either predictive or statistical, based on the type of consumer analytics request.

NF Load Analytics Workflow

The following workflow depicts an analytics consumer request to OCNWDAF for NF load analytics:

Figure 7-4 NF Load Analytics Workflow

The consumer NF sends an analytics request to the OCNWDAF for NF load analytics of a specific NF. The analytics request is either a Nnwdaf_AnalyticsInfo_Request or a Nnwdaf_AnalyticsSubscription_Subscribe request. The Analytics ID is set to NF load, the target for analytics and the analytics filter are also specified in the request. The NF can request statistics or predictions or both and can also provide a time window during which the analytics are generated and shared.
The request is authorized and then the OCNWDAF subscribes to the NRF to receive notifications about load and status changes of NF instances identified by the NF ids. The Nnrf_NFManagement_NFStatusSubscribe service is used to subscribe for each NF instance.
The NRF notifies the OCNWDAF of any changes in the load and status NF instances by using Nnrf_NFManagement_NFStatusNotify service.
The OCNWDAF derives the requested analytics from CAP4C.
The OCNWDAF provides requested NF load analytics to the consumer NF along with the corresponding Validity Period or Area of Interest, using either the Nnwdaf_AnalyticsInfo_RequestResponse response or Nnwdaf_AnalyticsSubscription_Notify response, based on the consumer NF request.
If the consumer NF request is a subscription request, the OCNWDAF generates new analytics based on the Analytics target period or Reporting Threshold specified in the request. The OCNWDAF receives notifications from the NRF whenever there is a change in the NF load of the target NF, accordingly the OCNWDAF generates new analytics information for the consumer NF.

Oracle Communications Network Analytics Data Director (OCNADD) as Data Source

The Oracle Communications Network Analytics Data Director (OCNADD) can be configured as a data collection source for the OCNWDAF.

Figure 7-5 NF Load Analytics Workflow OCNADD as Data Source

Note:

During OCNADD deployment, configure the parameter Include Message with xDR as HEADERS_DATA. Select this option to include xDRs with complete messages. For more information, see "Select xDR Type Related Info" in the "Creating Correlation Configurations procedure" in the "Correlation Configurations" section of the Oracle Communications Network Analytics Data Director User Guide.

Below is the workflow description when OCNADD is configured as the data source:

The analytics consumer sends an analytics request to the OCNWDAF for NF Load analytics. The analytics request is either an analytics subscription request (Nnwdaf_AnalyticsSubscription_Subscribe) or an analytics information request (Nnwdaf_AnalyticsInfo_Request).
The OCNWDAF collects the required data such as cell location or session information from the Data Director.
The OCNWDAF derives the required analytics from the CAP4C.
The OCNWDAF sends the analytics information to the analytics consumer through a Nnwdaf_AnalyticsSubscription_Notify or Nnwdaf_AnalyticsInfo_Request message.

Configure OCNADD as a Data Source

A specific data set is required from the OCNADD for NF Load analytics, follow the procedure below to configure OCNADD as a data source for NF Load Analytics:

OCNADD Filter Configuration

Configure the OCNADD filter as below to isolate messages sent by the UDM to the NRF by adding the user-agent filter condition:

#Filter Request
curl -k --location --request POST 'https://ocnaddconfiguration:12590/ocnadd-configuration/v2/filter' \
 --header 'Content-Type: application/json' \
 --data-raw '{
     "filterConfigurationDtos": [{
             "userName": "Admin",
             "filterInfoDto": {
                 "filterName": "NRFHeartbeat_UDM",
                 "filterConditionDto": [
                    {
                         "conditionName": "service-name",
                         "conditionValue": [
                             "nnrf-nfm/v1/nf-instances"
                         ]
                    }
                    {
                        "user-agent": "udm"
                    }
                    {
                        "methodType":["PATCH"]
                    }
                 ],
                 "filterRule": ["service-name" AND "user-agent" AND "methodType"],
                 "action": "ALLOW",
                 "status": "ACTIVATE"
             },
             "associationDto": {
                 "associationType": "EGRESS_FILTER",
                 "associationName": [
                     "CAP4C1-FILTERED"
                 ]
             }
         }
     ]
 }'

OCNADD Correlation Configuration

Set the following parameters for the correlation service configuration:

xDR type = SUDR (Single Unit Detailed Record), as transaction correlation is not required.
includeMessageWithxDR = HEADERS_DATA

OCNADD Correlation Filter for NRF Load

Configure the OCNADD Correlation Filter for NRF Load as below:

#Correlation Request
curl -k --location --request POST 'https://ocnaddconfiguration:12590/ocnadd-configuration/v1/correlation' \
--header 'Content-Type: application/json' \
--data '{
    "correlationConfigDto": {
        "configurationName":"NRFHeartBeatPatch",
        "userName":"Admin",
        "dataStreamStartPoint": "LATEST",
        "inboundDataStreamName":"CAP4C1-FILTERED",
        "outboundDataStreamName":"CAP4C2-CORRELATED",
        "includeMessageWithxDR": "HEADERS_DATA",
        "supportedXdrContents": [
                    "path",
                    "methodType"
                ],
        "xdrType": "SUDR"
    }
}

Configure Filters for LCI-OCI Headers

OCNADD filter configuration for capturing the UDM-AMF response messages:

Note:

You can use a default correlation configuration for this filter.

#Filter Request
curl -k --location --request POST 'https://ocnaddconfiguration:12590/ocnadd-configuration/v2/filter' \
 --header 'Content-Type: application/json' \
 --data-raw '{
     "filterConfigurationDtos": [{
             "userName": "Admin",
             "filterInfoDto": {
                 "filterName": "UDM_AMF_Response",
                 "filterConditionDto": [
                    {
                        "user-agent": "amf"
                    }
                    {
                        "message-direction": "RxResponse"
                    }
                 ],
                 "filterRule": ["user-agent" AND "RxResponse"],
                 "action": "ALLOW",
                 "status": "ACTIVATE"
             },
             "associationDto": {
                 "associationType": "EGRESS_FILTER",
                 "associationName": [
                     "CAP4C1-FILTERED"
                 ]
             }
         }
     ]
 }'

N8 Filter Configuration

Configure the N8 filter for NUDM-UECM API calls as below:

#Filter Request
curl -k --location --request POST 'https://ocnaddconfiguration:12590/ocnadd-configuration/v2/filter' \
 --header 'Content-Type: application/json' \
 --data-raw '{
     "filterConfigurationDtos": [{
             "userName": "Admin",
             "filterInfoDto": {
                 "filterName": "N8_AMF2UDM",
                 "filterConditionDto": [
                    {
                         "path": "registrations/amf-3gpp-access"
                    }
                    {
                        "user-agent": "amf"
                    }
                    {
                        "methodType": [
                            "PUT"
                        ]
                    }
                 ],
                 "filterRule": ["path" AND "user-agent" AND "methodType"],
                 "action": "ALLOW",
                 "status": "ACTIVATE"
             },
             "associationDto": {
                 "associationType": "EGRESS_FILTER",
                 "associationName": [
                     "CAP4C1-FILTERED"
                 ]
             }
         }
     ]
 }'

Correlation Configuration for N8 Filter

Below is the correlation configuration for N8 filter:

#Correlation Request
curl -k --location --request POST 'https://ocnaddconfiguration:12590/ocnadd-configuration/v1/correlation' \
--header 'Content-Type: application/json' \
--data '{
    "correlationConfigDto": {
        "configurationName":"N8AMF2UDMUECMCorrelation",
        "userName":"Admin",
        "dataStreamStartPoint": "LATEST",
        "inboundDataStreamName":"CAP4C1-FILTERED",
        "outboundDataStreamName":"CAP4C2-CORRELATED",
        "correlationMode":"CORRELATION_ID",
        "includeMessageWithxDR": "HEADERS",
        "supportedXdrContents": [
                    "path",
                    "supi",
                    "gpsi",
                    "pei",
                    "methodType",
                    "statusCode",
                    "producerNfType",
                    "consumerId",
                    "consumerFqdn",
                    "producerFqdn",
                    "contentType",
                    "ueId",
                    "snssai",
                    "dnn",
                ],
        "xdrType": "TDR"
    }
}'

Enhanced NF Load Analytics with OCNADD as Data Source

The Oracle Communications Network Analytics Data Director (OCNADD) can be configured as a data source for NF load analytics. When OCNADD is configured as a data source, the NF Load Analytics information in the OCNWDAF GUI is enhanced to include information such as Round Trip Time (RTT), Observed Transactions Per Second (OTPS), NF Load information, and Error Rates (only for UDM). A graphical representation of all these NF Load analytics parameters is displayed on the screen. For more information, see NF Load Dashboard.

The NF Load Dashboard is enhanced to display NF Load predictive analytics. You can provide a Start Time and Forecasting Period, and click Start Prediction. The OCNWDAF performs predictive analysis and displays the information on the OCNWDAF GUI. For more information, see NF Load Dashboard.

The OCNWDAF provides a GUI dashboard for selecting, training, and optimizing one or more ML Models for every analytics category. A new ML model, SARIMAX, is introduced for NF Load analytics. When this model is selected, configure the following Model Tuning Parameters:

Enter Moving Average Value - MA(q)
Enter Auto Regressive Value - AR(p)

For more information, see Machine Learning (ML) Model Selector.

Consumer request to OCNWDAF

A typical consumer request for NF load analytics consists of the following:

The analytics ID is set to NF_Load.
The Target of Analytics Reporting : Only "any UE" is currently supported.
Analytics filter information, consisting of the following:
- S-NSSAI (optional)
- Either NF Instance IDs or NF Set IDs or NF types are required. The order of precedence is NF Type, NF Instance ID , and then NF Set ID. If all or any two of them exist in the same request, one with the highest precedence is used.
- List of analytics subsets (optional)
Preferred level of accuracy (optional)
Accuracy level per analytics subset (optional)
Preferred order of results (ascending or descending) (optional)
Reporting Threshold (optional)
Analytics target time
If the consumer has opted for a subscription, the Notification Correlation Id and the Notification Target Address are included.

NF Load Data Collection

The OCNWDAF collects the following NF data to generate NF load analytics:

Table 7-15 Data Collection

Information Collected	Source	Description
NF Load	NRF	The load of specific NF instances in their NF profile.
NF Status	NRF	The status of specific NF instances. The status can be registered, suspended or not discoverable.

Note:

OCNWDAF can request NRF for data related to NF instances.
OCNWDAF can correlate the NF resources configuration with NF resource usage for generating the analytics output.

For more information see Data Collection from NRF.

Output NF Load Analytics

The following analytics information is obtained:

NF load statistics information
NF load prediction information

The NF load statistics include the following information:

Table 7-16 NF Load Statistics

Parameter	Data Type	P	Cardinality	Description
List of resource status	Integer	M	1 up to the maximum value	List of observed load information for each NF instance along with the corresponding NF id or NF Set ID (as applicable).
NF Type	NFType	M	1	Type of the NF instance.
NF Instance ID	NfInstanceId	M	1	Identification of the NF instance.
NF Set ID	NfSetId	O	0 to 1	Identification of the NF instance set
NF Status	NfStatus	O	0 to 1	The availability status of the NF in the Analytics target period. It is expressed as a percentage of time per status value. The possible values are registered, unregistered, or undiscoverable.
NF resource usage	Integer	C	0 to 1	The average usage of assigned resources (CPU, memory, and disk).
NF load	Integer	C	0 to 1	The average load on the NF instance during the analytics target period.
NF peak load	Integer	O	0 to 1	The maximum load on the NF instance during the analytics target period.
NF load (per area of interest)	Integer	C	0 to 1	The average load of the NF instances over the area of interest.
S-NSSAI	Snssai	C	0 to 1	Identifies the S-NSSAI.

The NF load prediction include the following information:

Table 7-17 NF Load Statistics

Parameter	Data Type	P	Cardinality	Description
List of resource status	Integer	M	1 up to the maximum value	List of observed load information for each NF instance along with the corresponding NF id or NF Set ID (as applicable).
NF Type	NFType	M	1	Type of the NF instance.
NF Instance ID	NfInstanceId	M	1	Identification of the NF instance.
NF Set ID	NfSetId	O	0 to 1	Identification of the NF instance set
NF Status	NfStatus	O	0 to 1	The availability status of the NF in the Analytics target period. It is expressed as a percentage of time per status value. The possible values are registered, unregistered, or undiscoverable.
NF resource usage	Integer	C	0 to 1	The average usage of assigned resources (CPU, memory, and disk).
NF load	Integer	C	0 to 1	The average load on the NF instance during the analytics target period.
NF peak load	Integer	O	0 to 1	The maximum load on the NF instance during the analytics target period.
Confidence	Uinteger	C	0 to 1	Indicates the confidence of this prediction.
NF load (per area of interest)	Integer	C	0 to 1	The average load of the NF instances over the area of interest.
S-NSSAI	Snssai	C	0 to 1	Identifies the S-NSSAI.

Enumeration NFType

Table 7-18 NFType

Enumeration Value	Description
"NRF"	Network Function: NRF
"UDM"	Network Function: UDM
"AMF"	Network Function: AMF
"SMF"	Network Function: SMF
"AUSF"	Network Function: AUSF
"NEF"	Network Function: NEF
"PCF"	Network Function: PCF
"SMSF"	Network Function: SMSF
"NSSF"	Network Function: NSSF
"UDR"	Network Function: UDR
"LMF"	Network Function: LMF
"GMLC"	Network Function: GMLC
"5G_EIR"	Network Function: 5G-EIR
"SEPP"	Network Entity: SEPP
"UPF"	Network Function: UPF
"N3IWF"	Network Function and Entity: N3IWF
"AF"	Network Function: AF
"UDSF"	Network Function: UDSF
"BSF"	Network Function: BSF
"CHF"	Network Function: CHF
"NWDAF"	Network Function: NWDAF
"PCSCF"	Network Function: P-CSCF
"CBCF"	Network Function: CBCF
"UCMF"	Network Function: UCMF
"HSS"	Network Function: HSS
"SOR_AF"	Network Function: SOR-AF
"SPAF"	Network Function: SP-AF
"MME"	Network Function: MME
"SCSAS"	Network Function: SCS/AS
"SCEF"	Network Function: SCEF
"SCP"	Network Entity: SCP
"NSSAAF"	Network Function: NSSAAF
"ICSCF"	Network Function: I-CSCF
"SCSCF"	Network Function: S-CSCF
"DRA"	Network Function: DRA
"IMS_AS"	Network Function: IMS-AS
"AANF"	Network Function: AAnF
"5G_DDNMF"	Network Function: 5G DDNMF
"NSACF"	Network Function: NSACF
"MFAF"	Network Function: MFAF
"EASDF"	Network Function: EASDF
"DCCF"	Network Function: DCCF
"MB_SMF"	Network Function: MB-SMF
"TSCTSF"	Network Function: TSCTSF
"ADRF"	Network Function: ADRF
"GBA_BSF"	Network Function: GBA BSF
"CEF"	Network Function: CEF
"MB_UPF"	Network Function: MB-UPF
"NSWOF"	Network Function: NSWOF
"PKMF"	Network Function: PKMF
"MNPF"	Network Function: MNPF
"SMS_GMSC"	Network Function: SMS-GMSC
"SMS_IWMSC"	Network Function: SMS-IWMSC
"MBSF"	Network Function: MBSF
"MBSTF"	Network Function: MBSTF
"PANF"	Network Function: PANF

Type NfInstanceId

Table 7-19 NfInstanceId

Type Name	Type Definition	Description
NfInstanceId	String	String uniquely identifying a NF instance. The format of the NF Instance ID shall be a Universally Unique Identifier (UUID) version 4, as described in IETF RFC 4122 [15].

Type NfSetId

Table 7-20 NfSetId

Type Name Type Definition Description

NfSetId

String

Type Name	Type Definition	Description
NfSetId	String	NF Set Identifier. Formatted as below: `set<SetID>.<nftype>set.5gc.mnc<MNC>.mcc<MCC>` or `set<SetID>.<NFType>set.5gc.nid<NID>.mnc<MNC>.mcc<MCC>` Example: `setxyz.smfset.5gc.mnc012.mcc345`

NF Set Identifier. Formatted as below:

set<SetID>.<nftype>set.5gc.mnc<MNC>.mcc<MCC>

set<SetID>.<NFType>set.5gc.nid<NID>.mnc<MNC>.mcc<MCC>

Example:

setxyz.smfset.5gc.mnc012.mcc345

Type NfStatus

Table 7-21 NfStatus

Attribute name	Data type	P	Cardinality	Description
statusRegistered	SamplingRatio	C	0 to 1	Percentage of time with status “registered”
statusUnregistered	SamplingRatio	C	0 to 1	Percentage of time with status “unregistered”
statusUndiscoverable	SamplingRatio	C	0 to 1	Percentage of time with status “undiscoverable”

Note:

The total of status values should be equal or lower than 100%. At least one value is provided.

Type Snssai

Table 7-22 Snssai

Attribute name	Data type	P	Cardinality	Description
sst	Uinteger	M	1	Represents the Slice or Service Type. It indicates the expected Network Slice behavior in terms of features and services. The allowed range is 0 to 255.
sd	String	O	0 to 1	Represents the Slice Differentiator in the 3-octet string format.

7.5 Slice Load Level Analytics

The OCNWDAF provides Slice Load Level analytics to the consumers at the Network Slice level. To generate this analytics report, the OCNWDAF need not have information about the subscribers in the slice, the slice load level analytics information is not subscriber specific. The OCNWDAF notifies slice specific network status analytics information to the subscribed consumers. The Analytics Subscription service and Analytics Information service expose the slice load level analytics to the consumers.

The Analytics ID used for the slice load level information is “SLICE_LOAD_LEVEL”(for Analytics Subscription Service) or "LOAD_LEVEL_INFORMATION” (for Analytics Information Service).

Note:

In the current release, only one network instance per slice is supported.

Figure 7-6 Slice Load Level Workflow

A consumer subscribes to an OCNWDAF (Nnwdaf_AnalyticsSubscription_Subscribe) or sends a request to an OCNWDAF (Nnwdaf__AnalyticsInfo_Request) with the Analytics ID and optional filters such as S-NSSAI, Area of Interest and Load level threshold value.
The OCNWDAF collects input data from the NRF to derive resource usage statistics and predictions for a Network Slice instance. It sends a Nnrf_NFDiscovery_Request to the NRF and receives a Nnrf_NFDiscovery_Response from the NRF.
The OCNWDAF invokes a Nnrf_NFManagement_NFstatus_Subscribe request to get the NFs status information, the NRF responds with a Nnrf_NFManagement_NFstatus_Subscribe_Notify with requested information.
The OCNWDAF subscribes to the SMF's event exposure service to collect data on the number of PDU sessions currently registered on a specific Network Slice (if the slice is available). The PDU session establishment or release event information is collected.
The OCNWDAF derives the slice load analytics from the CAP4C.
The OCNWDAF delivers the analytics information to the consumers by invoking OCNWDAF_AnalyticsSubscription_Notify or OCNWDAF_AnalyticsInfo_Request response.

Oracle Communications Network Analytics Data Director (OCNADD) as Data Source

The Oracle Communications Network Analytics Data Director (OCNADD) can be configured as a data collection source for the OCNWDAF.

Figure 7-7 Slice Load Level Workflow OCNADD as Data Source

Note:

Below is the workflow description when OCNADD is configured as the data source:

The analytics consumer sends an analytics request to the OCNWDAF for Slice Load Level analytics. The analytics request is either an analytics subscription request (Nnwdaf_AnalyticsSubscription_Subscribe) or an analytics information request (Nnwdaf_AnalyticsInfo_Request).
The OCNWDAF collects the required data such as cell location or session information from the Data Director.
The OCNWDAF derives the required analytics from the CAP4C.
The OCNWDAF sends the analytics information to the analytics consumer through a Nnwdaf_AnalyticsSubscription_Notify or Nnwdaf_AnalyticsInfo_Request message.

Consumer request to OCNWDAF

A consumer request for Slice Load Level analytics contains:

Analytics ID: “SLICE_LOAD_LEVEL” (for Analytics Subscription Service) or LOAD_LEVEL_INFORMATION” (for Analytics Information Service).

Analytics filter information:
- S-NSSAI: The S-NSSAI is the network slice identifier.
- Area of Interest (AoI) (Optional)
- Load level threshold value
- List of NF types (Optional)
- List of analytics subsets (Optional)
- Maximum number of objects (Optional)

The OCNWDAF reports when the load level of the network slice crosses the threshold value specified in the network analytics subscription service.

The analytics is generated on specified event detection (Network Analytics Information Service) or if the defined threshold is reached (Network Analysis Subscription Service). In this case, the event and threshold are related to the load level in the slice.

Slice Load Level Analytics Report

The following analytics information is obtained:

Network Slice load statistics
Network Slice load predictions

The Network Slice load statistics include the following information:

Table 7-23 Network Slice Instance Load Statistics

Parameter	Data type	P	Cardinality	Description
Load Level Information	loadLevelInformation	M	1 up to the maximum value	Load level information of the network slice calculated per time period. Load level is based on the maximum number of UEs or sessions that the slice can support and is limited by the maximum configured UEs or sessions. The load level value is the either the value of "Percent UE" or "Percent Session", whichever is the highest.
S-NSSAI	String	M	0 to 1	The S-NSSAI is the network slice identifier. This information is obtained from the analytics request.
Number of UE Registrations	Integer	M	1 up to the maximum value	The number of UE registrations within the Network Slice.
Percent UE	Integer	M	1 up to the maximum value	Is a proprietary value obtained by the percentage of Number of UE registrations in the slice and the configured value of maximum UEs per S-NSSAI.
Number of PDU Sessions	Integer	M	1 up to the maximum value	The number of PDU Sessions established within the Network Slice.
Percent Sessions	Integer	M	1 up to the maximum value	Is a proprietary value obtained by the percentage of Number of PDU Sessions in the slice and the configured value of maximum sessions per S-NSSAI.
Exceed Load Level Threshold	Boolean	M	True or False	Is true when the load level threshold is crossed within the time period of the analytics report.

Table 7-24 Type LoadLevelInformation

Type Name	Type Definition	Description
LoadLevelInformation	Integer	Load level information of the network slice

The Network Slice load predictions include the following information:

Table 7-25 Network Slice Instance Load Predictions

Parameter	Data type	P	Cardinality	Description
Load Level Information	loadLevelInformation	M	1 up to the maximum value	Load level information of the network slice calculated per time period. Load level is based on the maximum number of UEs or sessions that the slice can support and is limited by the maximum configured UEs or sessions. The load level value is the either the value of "Percent UE" or "Percent Session", whichever is the highest.
S-NSSAI	String	M	0 to 1	The S-NSSAI is a network slice identifier.
Number of UE Registrations	Integer	M	1 up to the maximum value	The number of UE registrations in the Network Slice.
Percent UE	Integer	M	1 up to the maximum value	Is a proprietary value obtained by the percentage of Number of UE registrations in the slice and the configured value of maximum UEs per S-NSSAI.
Number of PDU Sessions	Integer	M	1 up to the maximum value	The number of PDU Session established in the Network Slice.
Percent Sessions	Integer	M	1 up to the maximum value	Is a proprietary value obtained by the percentage of Number of PDU Sessions in the slice and the configured value of maximum sessions per S-NSSAI.
Exceed Load Level Threshold	Boolean	M	True or False	Is true when the load level threshold is crossed within the time period of the analytics report.
Confidence	Uinteger	C	0 to 1	Indicates the confidence of this prediction. It has a value range from Minimum "0" up to Maximum "100".

7.6 UE Related Analytics

OCNWDAF provides the following UE related analytics:

UE mobility analytics. For more information, see UE Mobility Analytics.
UE abnormal behavior analytics (unexpected UE location and proprietary geofencing feature). For more information, see User Equipment (UE) Abnormal Behavior Analytics.

7.6.1 UE Mobility Analytics

OCNWDAF provides UE mobility statistical or predictive analytics to allow consumers to do the following:

Collect UE mobility related information from 5G NFs (such as AMF).
Perform data analytics on the collected information to obtain UE mobility descriptive or predictive analytics.

Note:

The detailed information collected by the OCNWDAF can be network data from 5GCs.

UE mobility related network data collected from 5GC is UE location information.

UE Mobility Workflow

Figure 7-8 UE Mobility Workflow

The consumer sends a request to OCNWDAF for analytics on a specific UE or a group of UEs, using either the Nnwdaf_AnalyticsInfo or the Nnwdaf_AnalyticsSubscription service. The consumer can request statistics or predictions or both. The type of analytics is set to UE mobility information. The NF provides the UE ID or Internal Group ID in the Target of Analytics Reporting.
If the request is authorized, and in order to provide the requested analytics, OCNWDAF may subscribe to events with all the serving AMFs for notification of location changes.

Note:
This step may get skipped when OCNWDAF has the requested analytics available already.

Note:
OCNWDAF determines the AMF serving the UE or the group of UEs as described in 3GPP 23.288 6.2.2.1.
OCNWDAF derives the requested analytics from CAP4C.
OCNWDAF provides the requested UE mobility analytics to the consumer, using either Nnwdaf_AnalyticsInfo_Request response or Nnwdaf_AnalyticsSubscription_Notify.
If the consumer has subscribed to receive notifications for UE mobility analytics, after receiving event notification from the AMFs subscribed by OCNWDAF , OCNWDAF generates new analytics and provides them to the NF through a Nnwdaf_AnalyticsSubscription_Notify response.

Oracle Communications Network Analytics Data Director (OCNADD) as Data Source

The Oracle Communications Network Analytics Data Director (OCNADD) can be configured as a data collection source for the OCNWDAF.

Figure 7-9 UE Mobility Workflow with OCNADD as Data Source

Below is the workflow description when OCNADD is configured as the data source:

The analytics consumer sends an analytics request to the OCNWDAF for UE Mobility analytics. The analytics request is either an analytics subscription request (Nnwdaf_AnalyticsSubscription_Subscribe) or an analytics information request (Nnwdaf_AnalyticsInfo_Request).
The OCNWDAF collects the required data such as cell location or session information from the Data Director.
The OCNWDAF derives the required analytics from the CAP4C.
The OCNWDAF sends the analytics information to the analytics consumer through a Nnwdaf_AnalyticsSubscription_Notify or Nnwdaf_AnalyticsInfo_Request message.

Consumer Request to OCNWDAF

A consumer request for this analytics information contains:

Analytics ID: “UE Mobility”
The following filters can be specified in the subscription request:
- A single UE or a group of UEs (the Target of analytics reporting)
- Analytics target period indicating the time period over which the statistics or predictions are requested
- Preferred level of accuracy of the analytics (low or high)
- A Notification Correlation ID and Notification Target Address in a subscription

Output UE Mobility Analytics

The following UE mobility analytics information is obtained by OCNWDAF:

UE mobility descriptive analytics
UE mobility predictive analytics

The following table lists the UE mobility descriptive analytics:

Table 7-26 UE mobility descriptive analytics

Parameter	Data type	P	Cardinality	Description
Time slot entry	DateTime	O	0 to 1	This attribute identifies the timestamp when the UE arrives at the location.
Time slot start	ScheduledCommunicationTime	O	0 to 1	Identifies time of the day and day of the week which are valid within the observation period when the UE moves.
Duration	DurationSec	M	1	This attribute identifies the time duration the UE stays in the location. If the analytics result applies for a group of UEs, it indicates the average duration for the group of UEs.
UE location	UserLocation	M	1	This attribute contains the detailed location, the ueLocationTimestamp attribute in the 3GPP access type of UserLocation data type shall not be provided.
UE location Ratio	SamplingRatio	C	0 to 1	Indicates the percentage of UEs in the group (in case of a UE group)

The following table lists the UE mobility predictive analytics:

Table 7-27 UE mobility predictive analytics

Parameter	Data type	P	Cardinality	Description
Time slot entry	DateTime	O	0 to 1	This attribute identifies the timestamp when the UE arrives at the location.
Time slot start	ScheduledCommunicationTime	O	0 to 1	Identifies time of the day and day of the week which are valid within the observation period when the UE moves.
Duration	DurationSec	M	1	This attribute identifies the time duration the UE stays in the location. If the analytics result applies for a group of UEs, it indicates the average duration for the group of UEs.
UE location	UserLocation	M	1	Indicates the predicted location during the analytics target period.
Confidence	Uinteger	C	0 to 1	Indicates the confidence of a prediction
Ratio	SamplingRatio	C	0 to 1	Indicates the percentage of UEs in the group (in case of a UE group)

Note:

When the target of analytics reporting is an individual UE, for example, one UE ID (SUPI) is included, OCNWDAF provides the analytics mobility result (list of (predicted) time slots) to the service consumer(s) for the UE.
The results for UE groups address the group globally. The ratio is the proportion of UEs in the group at a given location at a given time.
The time slots are provided by order of time, possibly overlapping. The locations are provided by decreasing value of ratio for a given time slot. The sum of all ratios on a given time slot must be equal or less than 100%. Depending on the list size limitation, the least probable locations on a given analytics target period may not be provided.

7.6.2 User Equipment (UE) Abnormal Behavior Analytics

OCNWDAF provides UE abnormal behavior analytics that allow consumers to identify a specific UE or a group of UEs with abnormal behavior.

The UE abnormal behavior analytics consumer subscribes analytics about abnormal behavior from OCNWDAF based on the UE subscription, network configuration, or application layer request.

OCNWDAF performs data analytics on abnormal behavior provided there is a related subscription.

OCNWDAF provides the following types of UE abnormal behavior analytics:

Unexpected UE location: The UE location is monitored to determine if the UE is an unexpected location.
Geofencing: Geofences are virtual fences or perimeters around physical locations, they can be used to trigger events when a UE enters or exits the region.

UE Abnormal Behavior Workflow

Figure 7-10 UE Abnormal Behavior Workflow

A consumer NF subscribes to OCNWDAF using Nnwdaf_AnalyticsSubscription_Subscribe (Analytics ID set to "Abnormal behavior", Target of Analytics Reporting set to Internal-Group-Identifier, any UE or SUPI, Analytics Filter Information).

A consumer NF subscribes to abnormal behavior notification from the OCNWDAF for a group of UEs, any UE, or a specific UE. The Analytics ID indicates OCNWDAF to identify misused or hijacked UEs through abnormal behavior analytics.
The OCNWDAF sends a Namf_EventExposure_Subscribe message to the AMF.
OCNWDAF sends subscription requests to the related AMF to collect UE behavioral information if it has not subscribed such data.

Note:
OCNWDAF determines the related AMF(s). The AMF sends event reports to OCNWDAF based on the report requirements contained in the subscription request received from OCNWDAF.
If requested by OCNWDAF through Event Filter(s), the AMF checks whether the UE's behavior matches its expected UE behavioral information. In this case, the AMF sends event reports to OCNWDAF only when it detects that the UE's behavior deviated from its expected UE behavior.
OCNWDAF derives the analytics information from CAP4C.
OCNWDAF sends Nnwdaf_AnalyticsSubscription_Notify (Analytics ID, Exception ID, Internal-Group-Identifier or SUPI, Exception level) to the consumer NF.

Oracle Communications Network Analytics Data Director (OCNADD) as Data Source

The Oracle Communications Network Analytics Data Director (OCNADD) can be configured as a data collection source for the OCNWDAF.

Figure 7-11 UE Abnormal Behavior Workflow with OCNADD as Data Source

Below is the workflow description when OCNADD is configured as the data source:

The analytics consumer sends an analytics request to the OCNWDAF for UE Abnormal Behavior analytics. The analytics request is either an analytics subscription request (Nnwdaf_AnalyticsSubscription_Subscribe) or an analytics information request (Nnwdaf_AnalyticsInfo_Request).
The OCNWDAF collects the required data such as cell location or session information from the Data Director.
The OCNWDAF derives the required analytics from the CAP4C.
The OCNWDAF sends the analytics information to the analytics consumer through a Nnwdaf_AnalyticsSubscription_Notify or Nnwdaf_AnalyticsInfo_Request message.

Consumer request to OCNWDAF

A consumer request for this analytics information contains:

Analytics ID: “Abnormal Behavior”
The target of analytics reporting which can be a single "UE" or "any UE"
Target time period of the request
Expected UE behavior parameters, these parameters are used to define the geofencing area.
Expected analytics type

List of exception IDs: List of exception IDs indicates the specific analytics requested by the consumer.

Note:

Expected analytics type and list of Exception IDs are mutually exclusive, only one of them are specified in the request. The following table provides information about the correlation between expected analytics type and exception IDs:

Table 7-28 Expected analytics type and Exception IDs

Expected analytics type	Exception IDs
Mobility	Unexpected UE location Geofencing

The analytics filter should at least include S-NSSAI (if the target of analytics reporting is any UE)
Area of interest
Application ID
Optionally, maximum number of objects and maximum number of SUPIs
Notification Correlation ID and Target Address in the subscription.

UE Geofencing Analytics

Geofences are virtual fences around physical locations. Geofences can be used to trigger events when a UE enters or exits the defined region. Geofences are defined by using the ExpectedUeBehaviourData parameter in the analytics request. A proprietary exception ID "UE_GEOFENCE" is defined to obtain geofencing analytics information. On receiving an analytics request from a consumer, the OCNWDAF detects if the UE location is within the fence or outside the fence or if the UE location is unknown, and accordingly the OCNWDAF provides the analytics report to the consumer.

When OCNWDAF detects UEs that deviate from the expected behavior, for example, in case an unexpected UE location or an abnormal traffic pattern is displayed, the OCNWDAF notifies the consumer with UE Abnormal Behavior Analytics.

UE Abnormal Behavior Analytics

The following UE abnormal behavior analytics information is obtained by OCNWDAF:

Table 7-29 UE abnormal behavior analytics

Parameter	Data type	Mandatory or Optional	Cardinality	Description
excep	Exception	M	1	Contains the exception information.
Exception ID	ExceptionId	M	1	Indicates the Exception ID. For analytics related to geofencing, the exception ID is "UE_GEOFENCE"
Exception Level	Integer	O	0 to 1	Measured level, compared to the threshold
Exception trend	ExceptionTrend	O	0 to 1	The supported values are "IN_FENCE", "OUT_FENCE" or "UNKNOWN", based on the UE location with respect to the geofencing area.
GroupId	Array	O	1 up to the maximum value	Indicates the internal group identifier
SUPI list	Array	C	1 up to the maximum value	Indicates the SUPIs of the UEs affected with Exception
Ratio	SamplingRatio	C	0 to 1	Indicates the estimated percentage of the UEs affected by Exception within the target of analytics reporting
sampRatio	SamplingRatio	O	0 to 1	Indicates the estimated number of UEs affected by Exception (applicable when Target of Analytics Reporting = "any UE")
dnn	Dnn	C	0 to 1	Identifies DNN, a full DNN with both the Network Identifier and Operator Identifier, or a DNN with the Network Identifier only. Shall be present if the "dnns" was provided within EventSubscription during the subscription for event notification procedure.
snssai	Snssai	C	0 to 1	Identifies the network slice information. Shall be present if the "snssais" was provided within EventSubscription during the subscription for event notification procedure.
confidence	Uinteger	C	0 to 1	Indicates the confidence of the prediction.

Enumeration ExceptionId

Table 7-30 ExceptionId

Enumeration Value	Description
UNEXPECTED_UE_LOCATION	Unexpected UE location
UE_GEOFENCE	Geofencing

Enumeration ExceptionTrend

Table 7-31 ExceptionTrend

Enumeration Value	Description
IN_FENCE	Indicates the UE is within the defined geofence area.
OUT_FENCE	Indicates the UE is outside the defined geofence area.
UNKNOWN	The location of the UE is unknown.

Type NetworkAreaInfo

Table 7-32 NetworkAreaInfo

Attribute name	Data type	P	Cardinality	Description
ecgis	Array	O	0 to N	Is the list of E-UTRA cell identities.
ncgis	Array	O	0 to N	Is the list of NR cell identities
gRanNodeIds	Array	O	0 to N	Is the list of list of the NG-RAN nodes.
tais	Array	O	0 to N	Is the list of tracking area identities.

Type EventReportingRequirement

Table 7-33 EventReportingRequirement

Attribute name	Data type	P	Cardinality	Description
accuracy	Accuracy	O	0 to 1	Preferred level of accuracy of the analytics.
startTs	DateTime	O	0 to 1	UTC time indicating the start time of the observation period. The absence of this attribute means subscription at the present time.
endTs	DateTime	O	0 to 1	UTC time indicating the end time of the observation period. The absence of this attribute means subscription at the present time. If provided, it shall not be less than the start time.
sampRatio	SamplingRatio	O	0 to 1	Percentage of sampling (1 up to 100%) among impacted UEs.
maxSupiNbr	Uinteger		0 to 1	Represents the maximum number of SUPIs expected in an object. Applicable for the event(s) providing a list of SUPIs during the analyticis response.
maxObjectNbr	Uinteger		0 to 1	Maximum number of objects expected for an analytics report. It’s only applicable for the event(s) which may provide more than one entries or objects during event notification.
timeAnaNeeded	DateTime		0 to 1	UTC time indicating the time when analytics information is needed.

Enumeration NotificationMethod

Table 7-34 NotificationMethod

Enumeration Value	Description
THRESHOLD	OCNWDAF subscription is notified when the defined threshold exceeds.

Enumeration Accuracy

Table 7-35 Accuracy

Enumeration Value	Description
LOW	Low accuracy
HIGH	High accuracy

Type SamplingRatio

Table 7-36 SamplingRatio

Name	Definition	Description
SamplingRatio	Integer	Unsigned integer indicating Sampling Ratio , expressed as percentage value from 1 up to 100.