Correlating miniscule variations in mobile signal strength with data generated by physical rain gauges, radars and satellites will allow the Public Utilities Board to develop more accurate rainfall predictions and flood management solutions
Singapore, 21 January 2022 – Hydroinformatics Institute (H2i) will tap StarHub’s ubiquitous network of mobile base stations as “opportunistic” rainfall sensors, creating a cost-effective rainfall monitoring system in a PUB pilot to be launched in the second quarter of the year.
The two companies will mine data on the impact rain has on mobile signal strength to provide PUB with an additional source of rainfall intensity, allowing the national water agency to better anticipate and prepare for heavy rain across the island.
The idea was among four PUB Global Innovation Challenge winners, picked for its potential to drive operational excellence and address Singapore’s future water needs through the application of digital solutions and smart technologies. The idea, which won the “Cost-effective Rainfall Monitoring” challenge, was chosen from some 57 applicants across four challenges. The team will receive mentorship, test-bedding opportunities, and support from PUB to further commercialise and scale the idea.
The H2i-StarHub team will be commissioned for a Proof-of-Concept (PoC) in Singapore’s southwestern district. If the pilot is successful, the project could be extended to the Proof-of-Value stage and cover more of the island, and eventually lead to a full national roll-out. This is the first time that such a system is being trialled in Singapore.
Accurate rainfall readings are critical for water resource management, early flood warnings and weather predictions. Traditional rain gauges are often spaced too far apart to collect high-resolution data. In the tropics, where rainfall varies greatly over space and time, it can often be challenging to quantify and forecast. Having a greater variety and density of data sources can make modelling more accurate, and predictions more precise.
Microwave backhaul links – or the wireless connections between mobile base stations – provide an efficient and cost-effective way to plug some of the information gaps. “The large network of base stations outnumber existing rain gauges, cover a much wider area than weather radar networks, and collect data around the clock,” explains the project’s principal investigator Keem Munsung, a radar specialist with the H2i.
As the World Meteorological Organisation recognises extreme weather events as the “new norm” in its newly-released State of the Climate report, there is even greater urgency to address their impact – and the novel pairing between H2i, a water technology scale-up and StarHub can help.
“From implementing energy saving measures through the use of technology, to adopting more renewable energy sources for our network infrastructure, StarHub has been proactively addressing climate change and mitigating its impact to us and our community,” said Chong Siew Loong, Chief Technology Officer, StarHub. “Ranked as the world’s most sustainable Wireless Telecommunication Service Provider by Corporate Knights, we are delighted to collaborate with H2i to expand the use of our existing signal attenuation data for an important and meaningful purpose, and to help Singapore become more green and sustainable. Beyond staying committed to our climate goals, we are glad that our technology and infrastructure can make a palpable difference to build a safer environment for all."
StarHub analyses signal strength data as part of its day-to-day network operations. Leveraging these insights, the company makes continual enhancements to the thousands of mobile base stations it owns and operates, for customers to enjoy the best network quality.
These insights on micro changes in signal strength, when processed by H2i’s data scientists and rainfall models, can be used to map rainfall intensity across the island. When correlated with other data sources, they support more accurate rainfall intensity measurement and flood management solutions. The data provided relates only to signal strength, and does not contain any personal or customer data.
“The beauty of this solution is that we get ready data that enhances information from existing rain measurement systems without the need to create additional infrastructure or make additional investments. The data will simply be further mined using Machine Learning, and be translated into rainfall intensity insights that can be visualised on a dashboard,” says Dr Keem.
Similar projects have been successfully undertaken in the Netherlands and Germany.