RAN Intelligent Controller for Predictive 5G Resource Allocation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing 5G networks face challenges in efficiently managing network resources and energy consumption due to dynamic demand scenarios, leading to inefficient resource allocation and increased energy usage.
Innovation Solution
Implementing a radio access network intelligent controller (RIC) with real-time data and artificial intelligence (AI) to predict network demands and dynamically allocate resources, utilizing a microservices framework for granular control and energy savings.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional static resource allocation is used in 5G networks, then network infrastructure is maintained and basic service is provided, but resource allocation efficiency deteriorates and energy consumption increases under dynamic demand
Solution Approach 1:
The patent implements dynamic resource allocation by transitioning from static configuration to real-time adaptive resource management. The RIC controller continuously monitors network state and dynamically adjusts resource allocation based on current demand conditions, allowing the system to optimize resource utilization and reduce energy consumption as traffic patterns change.
Solution Approach 2:
The system performs preliminary actions by predicting future network demand using machine learning algorithms and proactively allocating resources before peak demand occurs. This anticipatory resource allocation prevents resource starvation during high-demand periods while avoiding over-provisioning during low-demand periods, thereby improving overall efficiency and reducing energy waste.
2Productivity
If AI-based predictive resource allocation is implemented, then resource allocation efficiency is improved and energy consumption is reduced, but system complexity increases
Solution Approach 1:
The patent introduces a RIC controller as an intermediary layer between the radio access network and core network functions. This intermediary consolidates the complexity of AI/ML algorithms, data collection, and resource allocation decisions in a centralized entity, allowing the underlying network infrastructure to remain relatively simple while still achieving intelligent adaptive resource management.
Solution Approach 2:
The system segments resource allocation into distinct functional modules: data collection module, machine learning prediction module, decision-making module, and execution module. This segmentation allows each component to be optimized independently and facilitates easier maintenance, debugging, and updates of the complex AI-based system without requiring complete system redesign.
Data Source
AI summary
A framework for dynamic network resource allocation and energy saving based on the real-time environment, radio network information, and machine learning (ML) can be utilized via a radio access network (RAN) intelligent controller (RIC). Real-time and predicted network utilization can facilitate resource and energy savings by leveraging the RIC platform. For example, a network information base (NIB) in the RIC platform can collects RAN and user equipment (UE) resource related information in real time and provides the abstraction of the access network in the real time. ML can predict real-time information about the UEs at time t based on data analytics and real time radio resource needs. The RIC can then instruct the network to reduce or increase resources.