Video Call Diagnostics Using Machine Learning Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Video calls are challenging to diagnose and troubleshoot due to the complexity of network and service issues, leading to compromised user experiences and difficulties in identifying and correcting problems in real-time.
Innovation Solution
An electronic device that uses a pretrained machine-learning model to diagnose video-call problems by collecting communication-performance metrics and video-service performance metrics during a video call, providing corrective information, and optionally performing remedial actions to improve user experience.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If video call performance monitoring is implemented, then user experience quality is improved, but system complexity increases
Solution Approach 1:
The patent introduces a machine learning model as an intermediary component that automatically analyzes video call performance metrics and diagnoses problems. This intermediary handles the complex analysis work, allowing the monitoring system to maintain high reliability while managing complexity through automated intelligent processing rather than manual analysis
Solution Approach 2:
The system performs self-diagnosis by automatically collecting performance metrics, analyzing them through the machine learning model, and identifying issues without requiring external intervention. This self-service capability improves reliability through continuous monitoring while managing complexity through automated self-analysis rather than requiring complex external diagnostic tools
2Reliability
If real-time problem diagnosis is performed, then user experience is improved, but processing time increases
Solution Approach 1:
The machine learning model is pre-trained on video call performance data before deployment. This preliminary training action enables the model to rapidly diagnose issues during actual video calls without requiring time-consuming analysis during the diagnostic process, thus improving reliability while minimizing time loss
Solution Approach 2:
The patent replaces manual or mechanical diagnostic processes with machine learning-based automated analysis. This substitution enables rapid real-time diagnosis by using intelligent algorithms to quickly process performance metrics and identify issues, improving reliability while reducing the time required compared to traditional diagnostic methods
Data Source
AI summary
During operation, an electronic device receives, from a second electronic device in a network, a request for testing. In response, the electronic device set ups a video call with a video-call service. Then, the electronic device provides, to the second electronic device, an invitation for the video call. When the electronic device receives a notification (e.g., from the video-call service) that the video call has started, the electronic device provides content via the video-call service for the second electronic device. Next, the electronic device obtains communication-performance metrics associated with communication via the network during the video call and video-service performance metrics associated with the video call. Furthermore, the electronic device diagnoses a type of problem experienced at the second electronic device during the video call based at least in part on the communication-performance metrics, the video-service performance metrics and a pretrained machine-learning model.