ML-Based Mobile Network Load Relief With Temporary UE Downgrades
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Mobile networks experience overload conditions during peak periods, leading to the dropping of low-priority messages, which degrades service quality for UEs without prior notice, negatively impacting user experience.
Innovation Solution
Implementing a Network Data Analytics Function (NWDAF) that uses machine learning to predict load conditions and UE-specific session behavior, allowing for temporary network service downgrades with user authorization and incentives to alleviate network load.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If network overload is managed by dropping low-priority messages, then network load is reduced, but service quality and user experience deteriorate
Solution Approach 1:
The system performs preliminary actions by proactively notifying users of anticipated network congestion and obtaining their authorization before the overload occurs. This allows the network to preemptively downgrade service levels for authorized users, avoiding the need to reactively drop messages during congestion, thereby maintaining service quality while managing load.
Solution Approach 2:
The system dynamically adjusts service levels for individual users based on real-time network conditions and user authorization status. Instead of static message dropping, the network flexibly modulates service quality by temporarily downgrading authorized users' service levels, creating a dynamic balance between load management and service quality maintenance.
2Quantity of substance
If message dropping is used to manage overload, then network capacity is preserved, but user experience and satisfaction worsen
Solution Approach 1:
The system implements feedback by continuously monitoring network load conditions and user authorization status, then adjusting service levels accordingly. This closed-loop approach ensures that message throughput is managed while minimizing user experience degradation, as users are informed and can provide feedback through their authorization decisions.
Solution Approach 2:
Users actively participate in load management by authorizing service downgrades themselves in exchange for incentives. This self-service approach transforms users from passive victims of message dropping into active participants who control their own service levels, thereby preserving user experience while maintaining network throughput.
3Reliability
If proactive user authorization is implemented, then service quality is maintained, but system complexity increases
Solution Approach 1:
The network management system performs multiple functions: it monitors network load, predicts congestion, communicates with users, manages authorization, and adjusts service levels. By consolidating these diverse functions into a single multi-functional system, the patent reduces overall complexity compared to having separate specialized systems for each function.
Data Source
AI summary
A network device receives, from a machine learning (ML) engine, predicted future network load conditions associated with UE traffic at nodes, network elements (NEs), and/or network functions (NFs) in a mobile network. The network device applies policies to the predicted future network load conditions to select UEs as candidates for temporary downgrades in mobile network service, and initiates sending of authorization requests to the selected UEs to request authorization for the implementation of a temporary downgrade in mobile network service for each of the selected UEs. The network device causes mobile network service to be downgraded to one or more of the selected UEs, for a temporary time period, based on responses to the authorization requests.