Dynamic Functional Splitting for Radio Access Network Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Static functional splits in radio access networks (RANs) fail to optimize network performance due to changing network conditions, such as varying traffic loads and infrastructure changes, leading to suboptimal performance and increased demands on fronthaul networks.
Innovation Solution
A method and system that dynamically determine an optimal functional split for a radio access network using a machine learning model, specifically a reinforcement learning model, to analyze network data and adjust the split based on current performance indicators like traffic load, latency, and bitrate requirements, allowing for continuous optimization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If higher functional splits (e.g., Split 8) are used to maximize centralized baseband processing benefits, then deployment effort and RU cost are reduced, but demands on the fronthaul network increase with high bandwidth and strict latency requirements
Solution Approach 1:
The patent implements dynamic functional splitting that allows the RAN to adaptively change the split option based on current network conditions, traffic load, and fronthaul capacity. This dynamic adjustment resolves the contradiction by enabling the system to operate at higher splits when fronthaul capacity is abundant, and switch to lower splits when fronthaul resources are constrained, thus reducing deployment effort while managing fronthaul bandwidth demand.
Solution Approach 2:
The system changes the functional split parameter dynamically based on network conditions. By monitoring fronthaul bandwidth availability, latency requirements, and traffic patterns, the system adjusts the split option to optimize the trade-off between centralized processing benefits and fronthaul resource consumption.
2Quantity of substance
If lower functional splits (e.g., Split 1) are used to place more baseband processing within the RU, then demands on the fronthaul network are reduced, but the RU becomes larger and more complex consuming more power
Solution Approach 1:
The dynamic functional splitting mechanism allows the system to shift processing functions between RU and CU/DU based on operational needs. When fronthaul bandwidth is limited, the system can dynamically move more processing to the RU. When fronthaul capacity is sufficient, processing is moved back to the CU/DU, reducing RU complexity and power consumption while maintaining fronthaul efficiency.
Solution Approach 2:
The RU is designed with multi-functionality to handle different processing loads dynamically. By implementing a universal RU that can operate effectively across different split options, the system avoids the need for specialized high-complexity RUs, thereby reducing power consumption while maintaining fronthaul efficiency.
3Ease of operation
If a static functional split is selected at deployment, then network configuration is simplified, but network performance cannot be optimized due to varying traffic loads and infrastructure changes
Solution Approach 1:
The system transitions from static to dynamic functional splitting, where the split option is automatically adjusted based on real-time network conditions including traffic load, latency requirements, and infrastructure changes. This maintains configuration simplicity while continuously optimizing network performance through automated adaptation.
Solution Approach 2:
The dynamic functional splitting system incorporates feedback mechanisms that continuously monitor network performance metrics and fronthaul conditions. Based on this feedback, the system automatically adjusts the functional split to optimize performance, resolving the contradiction between configuration simplicity and performance optimization.
Data Source
AI summary
A method for determining an optimal functional split for a radio access network (RAN), includes: obtaining network data relating to performance of RAN elements configured with a current functional split; analyzing, by a machine learning model, the obtained network data to determine an optimum functional split, from among a predetermined plurality of functional splits, for optimizing network performance under current network conditions; and outputting the determined optimum functional split for configuring the RAN elements.