Predictive Policy Control for Telecommunications Network Congestion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Telecommunications networks face challenges in managing network capacity due to increasing data usage, leading to RAN congestion, which existing solutions like trigger-based policies and Deep Packet Inspection are costly and not scalable for wide-scale deployment.
Innovation Solution
A computer-implemented method that utilizes predictive indicators to proactively manage network capacity by determining policy control decisions based on subscriber data, historical data, and real-time network resource utilization, allowing for precise and targeted policy enforcement to prevent congestion and ensure Quality of Experience (QoE).
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If trigger-based policies are deployed to manage RAN congestion, then network service quality can be improved, but deployment cost and system complexity increase significantly
Solution Approach 1:
The patent introduces a policy management system as an intermediary between the RAN and core network elements. This mediator collects predictive indicators from multiple sources, processes them centrally, and generates policy decisions, thereby avoiding the need to modify numerous RAN components individually and reducing overall deployment complexity while maintaining service quality
Solution Approach 2:
The system uses predictive indicators to anticipate congestion conditions before they occur and pre-configures policy rules in advance. This preliminary action allows the network to proactively manage capacity and prevent congestion rather than reactively responding to it, improving service quality while simplifying real-time decision-making complexity
2Measurement precision
If Deep Packet Inspection platforms are deployed to detect RAN congestion, then congestion detection accuracy improves, but hardware cost and deployment time increase
Solution Approach 1:
The patent replaces physical Deep Packet Inspection hardware with a software-based policy management system that utilizes predictive indicators from existing network elements. This substitution eliminates the need for additional inspection hardware while maintaining congestion detection accuracy through alternative measurement approaches
Solution Approach 2:
Instead of deploying physical DPI platforms, the system creates virtual copies of congestion detection capabilities by collecting and analyzing predictive indicators from existing network elements. This virtualization approach achieves the same detection accuracy without requiring additional hardware resources
3Reliability
If network capacity is increased to relieve RAN congestion, then service quality improves, but network cost increases
Solution Approach 1:
The patent implements dynamic capacity allocation through policy rules that adjust network resource distribution in real-time based on predictive indicators and current network conditions. This dynamic approach allows the network to optimize service quality by allocating capacity where and when it is needed rather than maintaining excess capacity throughout, reducing overall network cost while maintaining high service quality
Solution Approach 2:
The system changes operational parameters such as bandwidth allocation, priority levels, and service thresholds based on predictive congestion indicators. By dynamically adjusting these parameters rather than increasing physical capacity, the network maintains optimal service quality while avoiding the costs associated with permanent capacity expansion
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A computer-implemented method of determining policy control decisions in a telecommunications network. The method comprises the steps of receiving a predictive indicator from a forecasting system, wherein the predictive indicator includes predictive network resource utilization information, responsive to receiving a service request from a gateway, determining a policy decision based on the predictive indicator, and sending the policy decision to the gateway.