Application of Virtual and Augmented Reality Technologies in Remote Technical Maintenance Systems for ICT Infrastructure
Ochiq kirish: yoʻq
Annotatsiya
The article explores the use of virtual and augmented reality technologies for remote maintenance of information and communication infrastructure. A mathematical model for optimising the diagnostic process has been developed, taking into account data transmission delays, object recognition accuracy, and operator cognitive load. An algorithm for adaptive video stream compression based on predicting the importance of image areas and an anomaly detection system using an ensemble of three machine learning models are proposed. The experimental part includes testing a prototype based on Microsoft HoloLens 2 during the maintenance of Tier III data centre equipment with the participation of 23 technical specialists over 8 weeks. The results show a 34% reduction in troubleshooting time (from 47.3 to 31.1 minutes, $\mathbf{p}\lt 0.001$) and a $51 \%$ reduction in errors (from 3.7 to 1.8 per procedure, $p=$ 0.003) compared to traditional methods. The anomaly detection system achieved 94.7% accuracy with a false alarm rate of 4.1% and a response time of less than 200 ms. The adaptive compression algorithm reduced channel bandwidth requirements by $37 \%$ while maintaining subjective quality.
Hali tarjima qilinmagan