Keio Secure Mobile Sensing Platform

🖥️ Platform

Context and problem statement

A smart city IoT platform earns its value by being open, scalable and able to carry many kinds of city data. Those same properties make it hard to defend: devices come from different manufacturers, data flows outward to third parties, and there is no single perimeter. Two earlier EU–Japan projects had built the sensing capability in Fujisawa but left the security question unanswered.

Solution overview

Keio University’s mobile sensing platform had been running with Fujisawa City for more than three years of demonstration experiments before this work began, collecting environmental data from sensor-equipped refuse collection vehicles. Rather than build a new deployment, M-Sec applied protection across the existing one at three levels: hardening and an intrusion detection system on the IoT devices themselves, Transport Layer Security providing point-to-point encryption for device-to-cloud communication, and sophisticated authentication with a lightweight encryption mechanism in the cloud.

The result secures the data stream from sensor to storage while leaving the platform open and scalable — the condition that made it worth securing in the first place.

Functional Scope and Features

– Device hardening and on-device intrusion detection
– TLS point-to-point encryption for device-to-cloud communication
– Cloud-side authentication with a lightweight encryption mechanism
– Support for heterogeneous devices and manufacturers
– Controlled distribution of collected data to third parties
– Retro-fitted onto an operating deployment rather than a purpose-built demonstrator

Use Cases and Deployments

Deployed as M-Sec Use Case 3 in Fujisawa, Japan, on the sensing fleet Keio operates jointly with the city. The platform traces back to ClouT (2013–2016) and BigClouT (2016–2019) and was secured end to end under M-Sec through 2021. COVID-19 restricted access to city staff and cleaning companies, so a substantial part of the verification had to be carried out at laboratory level rather than as a full field trial.

Related Challenges

Interested in this solution?

Get in touch to explore deployment opportunities or propose a complementary approach.

Share the Post: