TEID-Based Analytics Traffic Filtering in Communication Networks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Generating comprehensive analytics in communication networks is challenging due to the large volume of user plane traffic, which requires significant processing, memory, and storage resources. Existing approaches that sample or filter user plane traffic to reduce resource demands still face high processing overheads and limitations in the number of subscribers that can be analyzed.
Innovation Solution
The proposed solution exploits tunnel endpoint identifiers (TEIDs) to implicitly determine whether user plane traffic should be processed for analytics generation. By dividing the TEID space into subsets, a network node can allocate TEIDs based on whether traffic is to be processed for analytics, thereby reducing the need for correlating user plane traffic with the control plane.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If comprehensive analysis of user plane traffic is performed, then analytics quality and insight are improved, but hardware resource requirements (processing, memory, storage) increase significantly
Solution Approach 1:
The patent segments user plane traffic into different categories based on tunnel endpoint identifiers (TEIDs). By dividing the TEID space into subsets corresponding to different traffic types (e.g., analytics-relevant vs. analytics-irrelevant), the system can selectively process only the relevant segments, thereby maintaining analytics quality while reducing overall hardware resource consumption.
Solution Approach 2:
The patent applies local quality by treating different traffic flows differently based on their TEIDs. Instead of uniformly processing all user plane traffic, the system identifies specific local regions (traffic flows with particular TEIDs) that require analytics processing and applies processing only to those local segments, optimizing the balance between analytics quality and resource usage.
2Quantity of substance
If user plane traffic is sampled or filtered to reduce resource demands, then hardware resource requirements are reduced, but processing overhead remains high due to control-plane correlation
Solution Approach 1:
The patent performs preliminary action by pre-categorizing traffic flows using TEID-based filtering before analytics processing. The network node determines in advance which traffic flows require analytics processing based on their TEIDs, allowing for efficient pre-screening that eliminates the need for complex runtime control-plane correlation and significantly reduces processing overhead.
Solution Approach 2:
The patent extracts the traffic identification function from the control plane by using TEIDs embedded in the user plane packets themselves. This extraction allows the system to identify and filter relevant traffic directly from user plane data without requiring control-plane involvement, thereby reducing processing overhead and device complexity.
3Loss of information
If user plane packets are correlated with control plane to identify subscribers, then subscriber-level analytics are achieved, but processing overhead and hardware resources increase
Solution Approach 1:
The patent introduces TEIDs as an intermediary that bridges the gap between user plane traffic and subscriber identification. Instead of directly correlating user plane packets with control plane data, the system uses TEIDs as a mediator that encapsulates subscriber-related information, enabling efficient subscriber-level analytics without complex control-plane correlation.
Solution Approach 2:
The patent creates a simplified copy of subscriber identification information within the TEID field of user plane packets. This copying mechanism allows the system to carry essential subscriber identification data in the user plane itself, eliminating the need for complex control-plane correlation while maintaining subscriber-level analytics capability.
4Productivity
If subscriber whitelist filtering is applied to reduce traffic volume, then analytics processing load is reduced, but the number of subscribers that can be whitelisted is limited in practice
Solution Approach 1:
The patent changes the filtering parameter from subscriber-specific criteria to TEID-based classification. By using TEIDs as the filtering parameter, the system can efficiently categorize and process traffic from a large number of subscribers without being constrained by the practical limitations of whitelist management, thereby increasing both processing efficiency and the number of supported subscribers.
Data Source
AI summary
A network node is configured for use in a communication network. The network node determines whether or not user plane traffic for a subscription to the communication network is to be processed for generating analytics. The network node then allocates a tunnel endpoint identifier, to identify an endpoint of a tunnel through which the user plane traffic for the subscription is to be communicated, based on whether or not the user plane traffic for the subscription is to be processed for generating analytics according to the determination.