Distributed RAN Information Collection, Consolidation And RAN-Analytics

Abstract

Control Plane and User plane packet data are collected within the Radio Access Network using a plurality of network devices. Consolidation and summarization of this information is then performed to present a unified picture of RAN through abstract APIs to management and analytics applications. The invention identifies methods of retaining the collected network data, such as control and application protocol headers at the collection points, and consolidation and exporting this network data to management/reporting/analytics application using application driven rules for consolidation and summarization. Real-time statistical analysis tools, which may be used to predict failure and degradation trends and proactively control the underlying causes, are also disclosed.

What is claimed is: 
     
         1 . A method of generating consolidated information regarding a Radio Access Network (RAN), comprising:
 correlating, using a device disposed in the RAN, a plurality of multi-layer control plane protocols and user plane protocols from a plurality of logical interfaces based on packets using the protocols;   wherein the packets are intercepted by the device or are received by the device from monitoring or mirroring devices disposed in the RAN; and   exporting the consolidated information to other core or RAN network elements for service orchestration, service selection and mitigation in order to maximize quality of experience for a number of users and priority services.   
     
     
         2 . The method of  claim 1 , further comprising performing data reduction by tracking hierarchy of protocol states, combining information to retain summary information, and associating information from other control plane or user plane protocols. 
     
     
         3 . The method of  claim 2 , wherein the associating comprises associating domain information from one or more protocols in the user plane with TCP, UDP, GTP, or CP flows. 
     
     
         4 . The method of  claim 1 , further comprising using service plane information from other network devices to perform the correlating, wherein the service plane information comprises subscriber plan type, subscriber identity or other subscriber information. 
     
     
         5 . The method of  claim 1 , wherein the exporting further comprises providing an API to query an analysis and reporting node (ARN) or a data collection node (DN) to provide the consolidated information. 
     
     
         6 . The method of  claim 1 , further comprising computing user plane downlink and uplink bandwidth, TCP RTT, TCP initial RTT, determined based on round trip times during TCP connection setup, RAB setup time while establishing User Plane tunnels, TCP and HTTP object down load times. 
     
     
         7 . The method of  claim 1 , wherein the consolidated information comprises key performance indicators (KPIs), and the KPIs are selected from the group consisting of subscriber mobility index (SMI), subscriber quality index (SQI), service area mobility index (SAMI), and Subscriber Application Pattern (SAP). 
     
     
         8 . A method of computing Key Performance Indicators (KPIs) in in a Radio Access Network, comprising:
 correlating a plurality of multi-layer protocol data received from an inline or tap deployment mode where partial or complete copy of protocol packets are received by a device;   computing and estimating trending of one or more KPIs;   detecting an anomaly or a threshold crossings of one or more KPIs; and   varying the amount information collected, analyzed, correlated, saved, and exported so as to include greater amount of information around the anomaly or threshold crossing and reduce the information that is not associated with the anomaly or threshold crossing, so as to reduce the processing time, the amount of information exported, and the storage required for retaining the information.   
     
     
         9 . The method of  claim 8 , wherein the computing and estimating comprises utilizing statistical methods, comparing with historical trends, session and data volumes, estimating user mobility and predicting target sectors and network aggregation points based on network topology. 
     
     
         10 . The method of  claim 8 , wherein the KPIs are collected per aggregation point, and the aggregation points are selected from the group consisting of RAN sector, Service Area, Location Area, Transport Layer Address of RAN or Core Network for user plane tunnels, e-NB, and groups of eNodeBs. 
     
     
         11 . The method of  claim 8 , wherein the detecting is performed by a plurality of Data Collection Nodes (DNs), each DN computing partial metrics and exchanging information with other Data Collection Nodes that contain the remaining information for the same set of users or aggregation points, and further comprising computing completed metrics, in cases where a set of flows that belong to a user or aggregation point is spread across multiple data collection nodes due to user mobility, load balancing or due to network architecture reasons. 
     
     
         12 . The method of  claim 8 , wherein the detecting is performed by an Analysis and Reporting Node (ARN) after receiving partial information from a plurality of Data Collection Nodes, identifying the set of protocols that correspond to a user, or set of users that belong to a logical or physical aggregation points, or a set of locations, and then sending a request to collect or retain or export more detailed information in the vicinity of the anomaly or threshold crossing. 
     
     
         13 . The method of  claim 12 , wherein sending a request comprises requesting real-time consolidated information of one or more protocols with attributes from that protocol, and the computed KPIs from other protocols that relate to or occur in an overlapping time-period in the vicinity of the anomaly, thereby minimizing an amount of information exported, but reflecting relationship between multiple control and user plane protocols. 
     
     
         14 . A method of correlating a plurality of protocols in user plane and control plane to identify per user, per network aggregation points or domain/website flows and estimate consolidated per user, per aggregation point or domain/website performance metrics, the method comprising:
 generating real-time consolidated information of one or more protocols with attributes from that protocol, and computed key performance indicators (KPIs) from other protocols that relate to or occur in an overlapping time-period for the same user, same aggregation point or same website; and   exporting the consolidated information.   
     
     
         15 . The method of  claim 14 , wherein the generating is initiated after identifying a protocol or KPI anomaly, so as to minimize the amount of exported or stored information. 
     
     
         16 . The method of  claim 15 , wherein the generating is limited to temporal or spatial vicinity of the anomaly.