RAN RF Event Detection Using ML Metadata and Open RAN Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing communication systems lack efficient methods for analyzing low-level radio frequency signals to detect anomalies, unauthorized emitters, and control network operations in a standardized and interoperable manner.
Innovation Solution
Integrate machine learning networks into radio access networks to process low-level RF signal data, generating metadata for controlling network operations, using predefined protocols and interfaces within Open RAN architectures.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Difficulty of detecting and measuring
If machine learning networks are integrated into radio access networks to process low-level RF signal data, then detection capability for signal events and unauthorized emitters is improved, but device complexity increases
Solution Approach 1:
The patent introduces a RAN Intelligent Controller (RIC) as an intermediary component that hosts machine learning applications (xAPPs, rAPPs, zAPPs) separate from the core radio access network infrastructure. This mediator enables advanced detection capabilities while isolating the complexity of ML networks from the primary RAN operations, allowing standardized integration without overhauling existing systems.
Solution Approach 2:
The patent segments the RAN architecture into distinct functional units (RU, DU, CU, RIC) with the ML processing capability further divided into separate applications (xAPP for real-time, rAPP for near-real-time, zAPP for non-real-time). This segmentation allows selective deployment of detection capabilities and distributes computational complexity across multiple independent components rather than concentrating it in a single complex system.
2Productivity
If low-level signal data is transferred from radio unit to distributed unit or central unit, then processing capability is improved, but bandwidth requirements increase
Solution Approach 1:
The patent extracts only the essential low-level signal data features needed for detection purposes and transfers them to the RIC for ML processing, rather than transferring complete raw signal data. This extraction approach maintains processing capability by providing sufficient input data for ML algorithms while significantly reducing the bandwidth requirements through selective data transmission.
3Extent of automation
If machine learning networks are executed on radio access network components, then network control capability is improved, but ease of operation deteriorates
Solution Approach 1:
The patent creates a universal RIC platform that can host multiple different ML applications (xAPPs, rAPPs, zAPPs) with different functionalities using the same underlying infrastructure. This multi-functionality allows the system to perform various network control tasks through a single standardized interface, improving ease of operation compared to deploying separate specialized systems for each control function.
Data Source
AI summary
A method includes obtaining, using a specified protocol of a radio access network, low-level signal data corresponding to a radio frequency (RF) signal processed in the radio access network; providing the low-level signal data as input to at least one machine learning network; in response to providing the low-level signal data as input to the at least one machine learning network, obtaining, as an output of the at least one machine learning network, metadata providing information on one or more characteristics of the RF signal; and controlling an operation of the radio access network based on the metadata.