Interference Classification Platform for Local 6 GHz WiFi Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing techniques for detecting interferers in the 6 GHz spectrum for WiFi networks are challenging due to the wide variety and specificity of potential interferers, especially non-WiFi devices, which vary by geographical location, making centralized detection infeasible.
Innovation Solution
A machine learning (ML) model is trained using interactive user feedback to classify interferers, where system administrators capture and label spectrogram data to create a classifier model that can be deployed on wireless network devices to identify and avoid interferers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If centralized detection techniques are used to detect interferers, then detection can be performed with existing methods, but the approach becomes infeasible due to the wide variety and specificity of potential interferers in 6 GHz spectrum
Solution Approach 1:
The patent segments the interferer detection task by deploying distributed machine learning models across multiple access points rather than using a single centralized detection system. Each access point independently classifies interferers in its local environment, enabling the system to handle the wide variety and geographical specificity of 6 GHz interferers while maintaining detection reliability through localized adaptation
Solution Approach 2:
The patent applies local quality by training machine learning models at each access point using locally captured spectrogram data and geographical information. This allows each access point to develop specialized detection capabilities tailored to its specific environment, improving adaptability to location-specific interferers while maintaining overall system reliability through consistent classification frameworks
2Measurement precision
If machine learning models are trained with interactive user feedback to classify interferers, then classification accuracy for unknown interferers improves, but system complexity and training time increase
Solution Approach 1:
The patent implements self-service by enabling access points to automatically capture spectrogram data, train local machine learning models, and deploy classification algorithms without requiring extensive manual configuration or centralized coordination. The system serves itself by autonomously adapting to local interferer environments while maintaining high classification precision through interactive user feedback mechanisms
Solution Approach 2:
The patent applies preliminary action by pre-training machine learning models with spectrogram data captured during normal operation before deployment. The models are prepared in advance with geographical and environmental context, allowing them to quickly adapt to specific locations and achieve high classification precision without requiring complex real-time training procedures
3Adaptability or versatility
If spectrogram data is captured and labeled by system administrators to create classifier models, then the model can effectively classify location-specific interferers, but data collection and labeling effort increases
Solution Approach 1:
The patent applies periodic action by having access points continuously capture and update spectrogram data at regular intervals, allowing the machine learning models to progressively improve their classification capabilities over time. This periodic data collection approach enables the system to adapt to changing interferer environments without requiring intensive one-time data gathering efforts
Solution Approach 2:
The patent implements feedback mechanisms where classification results and interferer identification outcomes are used to refine and retrain the machine learning models. This feedback loop allows the system to improve its geographical adaptability progressively, using real-world performance data to enhance future classification accuracy without requiring manual re-labeling of all training data
Data Source
AI summary
Techniques for classification of wireless signals are disclosed. These techniques include receiving data describing radio frequency characteristics of a wireless network transmission environment and presenting the data for display on a graphical user interface (GUI) associated with a network device in the wireless network. The techniques further include receiving input from the GUI identifying one or more portions of the data, and training a machine learning (ML) model to classify an interferer in the wireless network transmission environment based on the identified one or more portions of the data.