Home / Work / Research · UTHM · 2021–2023
Water quality monitoring with machine learning
Connected sensing for a clearer picture of water conditions. Designed and deployed an IoT sensor network across 10 remote sites, combining water measurements with machine-learning analysis.
The challenge
Remote readings are difficult to use when measurements remain scattered. The project explored bringing sensor data and analysis into one monitoring workflow.
The engineering
The work covered sensor-network design, data collection, machine-learning models and power optimisation through prototyping and field testing.
Outcome & context
The research involved 3,000+ data points and reported 98% model accuracy and a 40% reduction in system power consumption. These are founder-reported results from this academic project, specific to its testing conditions; they are not performance guarantees for a Mesab AI product.
Why it matters
A foundation for exploring remote environmental monitoring, data visibility and earlier investigation of changing conditions.
This is the founder’s academic work, not a Mesab AI customer case study. See his engineering portfolio ↗ for further project context.