AI Capacity Forecasting for CBRS Channel Reconfiguration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In spectrum-controlled wireless communication networks, particularly in CBRS networks, there is a challenge in accurately predicting radio frequency (RF) resource utilization over time, leading to potential service disruptions due to insufficient or excessive resource allocation, as existing methods lack predictive capabilities to preemptively reconfigure channels efficiently.
Innovation Solution
The implementation of a method and apparatus that uses AI techniques to monitor RF usage, identify patterns, and predict capacity utilization, allowing for preemptive reconfiguration of network channels to efficiently allocate resources, utilizing a capacity forecasting module connected to a self-organizing network element and a domain proxy to reallocate channels based on predicted needs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional static channel allocation methods are used in CBRS networks, then network configuration is simple, but service disruptions occur due to insufficient or excessive resource allocation when capacity utilization changes over time
Solution Approach 1:
The system performs preliminary actions by predicting future capacity utilization patterns using AI/ML models and proactively reconfiguring channel allocations before service disruptions occur. The capacity forecasting module analyzes historical RF usage data to identify patterns and makes advance predictions, allowing the network to preemptively adjust resource allocation rather than reacting to congestion or underutilization after they occur.
Solution Approach 2:
The system implements continuous feedback loops where RF usage data is constantly monitored, capacity utilization is predicted based on identified patterns, and channel allocations are adjusted accordingly. The domain proxy receives predictions and executes reconfiguration, then continues monitoring to verify effectiveness and refine future predictions, creating a closed-loop control system that adapts to changing network conditions.
2Reliability
If AI techniques are implemented to predict capacity utilization and preemptively reconfigure channels, then service reliability improves, but computational complexity and processing requirements increase
Solution Approach 1:
The patent introduces a domain proxy as an intermediary component that acts as a bridge between the capacity forecasting module and the network configuration system. The domain proxy receives capacity utilization predictions, translates them into appropriate reconfiguration actions, and manages the actual channel allocation changes. This intermediary layer simplifies the overall system architecture by centralizing the complex AI/ML processing and pattern recognition functions while presenting a simplified interface to the rest of the network infrastructure.
3Productivity
If frequent network reconfiguration is performed to optimize channel allocation, then resource utilization efficiency improves, but network stability and configuration management become more difficult
Solution Approach 1:
The system employs periodic action by reconfiguring channel allocations based on predicted capacity utilization patterns rather than continuously or reactively. The AI/ML models identify periodic trends in RF usage and schedule reconfiguration actions at optimal intervals to match these patterns. This approach ensures resources are reallocated efficiently during high-utilization periods while maintaining stability during low-utilization periods, avoiding unnecessary configuration changes.
Data Source
AI summary
A method and apparatus for predicting capacity utilization and preemptively reallocating channels in a spectrum-controlled network such as a Citizen's Band Radio Service (CBRS) network. The network includes a plurality of Base Stations/Access Points (BS/APs) which monitor RF resource usage of the channels at each BS/AP and provide time series data relating to capacity utilization over a period of time. The data is analyzed, and patterns are identified in the collected capacity utilization data using Artificial Intelligence (AI) techniques. Looking forward, capacity utilization is predicted using AI techniques. Responsive to the capacity utilization predictions, the channels pre-emptively reallocated among BS/APs. The enterprise network provides feedback regarding prediction accuracy, which is utilized in a machine learning process to modify the models and provide greater prediction accuracy. The prediction apparatus includes a capacity utilization module and a training/retraining module, which may be located remotely from the enterprise network.