Node Rx Load Control for Policy Controller Overload

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

In network policy control systems, such as 3GPP PCC architecture, nodes interacting with the policy controller can generate a high load, leading to overloading of the policy controller, especially when handling a large number of services, which can result in network congestion and inefficient resource management.

Innovation Solution

Implementing an Rx load control mechanism in nodes that determines a load value of messages sent to the policy controller within a given time interval and limits further messages by selectively buffering or discarding them based on message types, such as initial vs. update authentication/authorization requests and session termination requests, to prevent overload.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If nodes send all generated messages to the policy controller, then complete service authorization and control is achieved, but the policy controller becomes overloaded

Engineering Contradiction:

Improveservice authorization completenessVSAvoidpolicy controller processing capacity

Core Design Contradiction:

ReliabilityVSProductivity

Solution Approach 1:

The node performs preliminary assessment of incoming messages against current session state and load conditions before forwarding to the policy controller. This preliminary filtering action prevents unnecessary messages from reaching the controller, reducing its load while ensuring only relevant authorization requests are processed

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The node acts as an intermediary between the service flow and the policy controller, intelligently selecting which messages to forward based on session state and load conditions. This intermediary function filters and conditions messages before they reach the controller, protecting it from overload while maintaining authorization completeness

Inventive Principle:

Principle #24Intermediary (Mediator)

2Productivity

If the node buffers all messages during high load, then the policy controller is protected from overload, but message delivery is delayed

Engineering Contradiction:

Improvepolicy controller load managementVSAvoidmessage delivery time

Core Design Contradiction:

ProductivityVSLoss of time

Solution Approach 1:

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The message buffering behavior dynamically adapts based on current load conditions and session state. During high load, selective buffering occurs; during normal conditions, messages are forwarded promptly. This dynamic adjustment optimizes both load management and delivery timing

Inventive Principle:

Principle #15Dynamics

3Productivity

If the node selectively discards messages to reduce load, then policy controller overload is prevented, but service quality may deteriorate

Engineering Contradiction:

Improvepolicy controller load reductionVSAvoidservice authorization accuracy

Core Design Contradiction:

ProductivityVSReliability

Solution Approach 1:

The node applies different handling strategies to different message types based on their specific characteristics and importance. Critical authorization requests are prioritized for forwarding, while less critical messages may be buffered or discarded. This localized quality differentiation ensures service authorization accuracy is maintained for important messages while managing overall load

Inventive Principle:

Principle #3Local quality

Data Source

PatentEP2837145B1Limitation of signaling messages to a policy controller of a network

Publication Date: 2020.01.22 TELEFONAKTIEBOLAGET LM ERICSSON (PUBL)

Figure 1
Figure 2
Figure 3

AI summary

A node (36'; 50) generates messages to be sent to a policy controller (30). Further, the node (36'; 50) determines a load value. The load value represents a number of the messages which are sent from the node (36'; 50) to the policy controller in a given time interval. In response to the load value reaching a threshold, the node (36'; 50) limits sending of at least one of the messages to the policy controller.