Project Overview
This project develops algorithms for automatic calibration of heterogeneous IoT sensor networks using AI and self-learning approaches. It focuses on environmental and industrial sensor systems with varying sensitivities and offsets.
Our Goal
Enable self-calibration and drift compensation in distributed sensor systems. Use machine learning to harmonise readings across heterogeneous sensors. Improve long-term accuracy of environmental and industrial IoT deployments.
Highlights
AI-based calibration without manual intervention. Applicable to large-scale deployments (smart cities, agriculture, environment). Integrates federated learning and statistical compensation methods.
Impact
Reduces cost and labour associated with manual recalibration. Extends lifespan and reliability of IoT infrastructures. Enhances data quality for AI-based analytics and decision systems.
