5G Network Capacity Adjustment via ML Load Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The network capacity of a wireless network, particularly a 5G network, can be overwhelmed by unpredictable events within its coverage area, leading to performance issues and potential failures of network components.
Innovation Solution
A method that uses a computing device to access event data and user data, employing a machine learning model to generate an expected network load. Based on this analysis, the system determines if the network will exceed its dynamic threshold limits, and if so, generates new network components to meet the anticipated load.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the network capacity is increased to handle unpredictable traffic spikes, then the network reliability is improved, but the device complexity and resource utilization deteriorate
Solution Approach 1:
The patent implements dynamic network capacity adjustment by using machine learning models to predict traffic loads and automatically provisioning network components. The system transitions from static capacity planning to dynamic scaling, where network resources are adjusted in real-time based on predicted demand, resolving the contradiction between maintaining high reliability and avoiding excessive complexity.
Solution Approach 2:
The system performs preliminary actions by predicting future network traffic loads using machine learning models before traffic spikes occur. This advance prediction allows the network to proactively provision capacity rather than reacting to failures, improving reliability while maintaining efficient resource utilization through targeted, pre-planned capacity additions.
2Quantity of substance
If additional network components are added to meet expected network load, then the network capacity is improved, but the loss of energy and resource utilization worsen
Solution Approach 1:
The patent changes the parameter of network capacity dynamically based on predicted traffic load. Instead of maintaining fixed high capacity, the system adjusts capacity parameters in response to ML-predicted demand, ensuring sufficient capacity during spikes while minimizing energy consumption during low-traffic periods, thus resolving the contradiction between capacity and energy efficiency.
Solution Approach 2:
The network system performs self-service by using machine learning to autonomously predict traffic patterns and trigger capacity provisioning without manual intervention. This automated self-adjustment ensures capacity is added only when predicted to be needed, avoiding energy waste from unnecessary components while maintaining adequate capacity during high-demand events.
3Productivity
If manual provisioning of network components is used, then the device complexity is reduced, but the productivity and response time to traffic spikes deteriorate
Solution Approach 1:
The patent replaces manual mechanical provisioning processes with an automated machine learning-based system. The ML model predicts traffic loads and automatically triggers component provisioning, substituting human-operated mechanical processes with an intelligent automated system that increases productivity while accepting the necessary complexity for autonomous decision-making.
Solution Approach 2:
The system implements feedback loops where machine learning models continuously monitor traffic patterns, predict future loads, and trigger provisioning actions based on these predictions. This closed-loop feedback mechanism enables rapid automated response to traffic spikes, significantly improving provisioning speed compared to manual processes while managing complexity through systematic automation.
Data Source
AI summary
A method may include accessing event data corresponding to an event affecting a region covered by a 5G network including a plurality of network components. The method may include accessing user data corresponding to a user equipment within the region covered by the 5G network. The method may include generating, using a machine learning model, an expected network load. The method may include accessing, a dynamic threshold associated with the 5G network. The dynamic threshold may include one or more limits associated with the plurality of network components. The method may include determining that the expected network load will cause the 5G network to exceed at least one limit of the dynamic threshold. In response to determining that the expected network load will exceed the limit, the method may include generating a new network component in the 5G network based at least in part on the expected network load.