ML Interference Classification Using OAM Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The increasing strain on limited radio-frequency (RF) spectrum in wireless communication networks due to interference from various sources, such as industrial machinery, electronics test equipment, and passive intermodulation (PIM) interference, leads to degraded service quality and reduced network capacity and coverage.
Innovation Solution
A system and method utilizing machine learning models to detect and classify interference sources in wireless communication networks by analyzing Operations, Administration, and Management (OAM) data, including Configuration Management (CM), Performance Management (PM), and topology data, to produce per-Physical Resource Block (per-PRB) interference data and identify interference source types with corresponding confidence levels.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional interference detection methods are used, then additional measurement devices are required, but system complexity and cost increase
Solution Approach 1:
The network elements themselves (base stations, user equipment) perform interference detection and classification using their existing measurement capabilities and operational data. The machine learning model processes OAM data that is already being collected for network management, eliminating the need for separate dedicated measurement devices while maintaining high detection accuracy
Solution Approach 2:
The machine learning model serves multiple functions: it detects interference presence, classifies interference types, and identifies affected network elements all using a single integrated system that processes existing OAM data. This multi-functional approach replaces what would traditionally require multiple specialized measurement devices
2Measurement precision
If manual interference classification is performed, then expert knowledge is required, but time consumption and operational complexity increase
Solution Approach 1:
The patent replaces manual expert analysis with an automated machine learning system that processes OAM data to classify interference types. The model has been trained on labeled interference data to recognize patterns and characteristics of different interference sources, enabling automated classification that is both accurate and rapid without requiring human expert intervention
Solution Approach 2:
The system uses labeled interference data from historical network operations to train the machine learning model. The model continuously improves its classification accuracy by learning from past interference events and their confirmed types, creating a feedback loop that enhances performance over time while maintaining rapid automated classification
3Measurement precision
If comprehensive OAM data analysis is performed, then interference classification accuracy improves, but data processing complexity increases
Solution Approach 1:
The machine learning model processes different segments of OAM data separately and systematically. It analyzes configuration management data, performance management data, and operational data in structured stages, processing each data type to extract relevant features before integrating them for final interference classification. This segmented approach manages complexity while comprehensively utilizing all available data
Data Source
AI summary
A method for classifying sources of interference provides Operations, Administration, and Management (OAM) data available in a wireless communication network to a trained machine learning model that outputs indications of the types of interference sources exhibited in the OAM data. The OAM data provided to the machine learning model may include per-Physical Resource Block (per-PRB) interference data for a cell, and may further include metadata corresponding to the configuration of the cell.