Network Load Traffic Management via Device Location Probability
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing network traffic management systems fail to effectively anticipate and manage network congestion, often resulting in reduced Quality of Service (QoS) and inadequate relief from congestion due to poorly targeted traffic management rules, which can lead to network congestion even in the absence of heavy users or low-priority traffic.
Innovation Solution
A system and method for detecting and managing network data traffic that includes a Radio Access Network (RAN), a network load traffic management server, and a database, which calculates probabilities of mobile device location and utilization to apply dynamic traffic management policies, identifying and mitigating congestion by pacing downlink traffic and offloading to Wi-Fi networks when necessary.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If simple traffic management rules are applied to predetermined traffic types, then congestion is avoided, but Quality of Service is reduced by affecting clients even when there is no actual congestion
Solution Approach 1:
The system dynamically changes traffic management parameters based on real-time network conditions. Instead of applying fixed rules to predetermined traffic types, the system adjusts management policies based on actual congestion levels, client behavior patterns, and network utilization metrics, thereby avoiding QoS degradation when congestion is absent while maintaining reliability when needed
Solution Approach 2:
The system transitions from static traffic management rules to dynamic policies that adapt to changing network conditions. By continuously monitoring network state and adjusting management actions in real-time, the system avoids affecting clients during non-congested periods while effectively managing congestion when it occurs
2Reliability
If bandwidth per user is limited during peak hours, then network congestion is managed, but clients are restricted from using available bandwidth even when no actual congestion exists
Solution Approach 1:
The system performs preliminary identification of clients likely to cause congestion before peak congestion occurs. By predicting which clients will generate heavy traffic based on historical patterns and current behavior, the system can prepare targeted management actions that only affect specific clients when necessary, rather than broadly limiting all clients during peak hours
Solution Approach 2:
The system applies differentiated bandwidth management to different clients based on their individual congestion risk profiles. Instead of uniformly limiting all clients during peak hours, the system selectively manages bandwidth for identified heavy users while allowing other clients to utilize available bandwidth, thereby maintaining service quality for non-problematic clients
3Reliability
If poorly targeted traffic management rules are applied, then network congestion events may still occur despite few or no heavy users, but the complexity of the system increases
Solution Approach 1:
The system enables clients to effectively manage their own traffic behavior through automated identification and targeted policies. By using machine learning to predict congestion-prone clients and applying selective management actions, the system achieves effective congestion relief without requiring complex manual configuration or oversight, as the system self-adjusts based on observed patterns
Solution Approach 2:
The system implements continuous feedback loops that monitor network congestion events and client behavior patterns. By analyzing this feedback data, the system automatically refines its traffic management policies to accurately identify and manage congestion sources, reducing false positives and minimizing the need for complex rule sets while maintaining reliable congestion relief
Data Source
AI summary
A system for detecting and managing network data traffic and a network load traffic management module is described. The system for detecting and managing network data traffic includes a Radio Access Network (RAN), a plurality of mobile devices, a network load traffic management server, a first dataset and a corresponding first baseline, a second data set and a corresponding second baseline, a data flow dataset, and a probability for one or more mobile devices to remain within each location associated with each RAN site, wherein the probability is calculated by the network traffic management module.