Epidemic tracking

The scenario

An epidemic may be controlled or predicted, if we can monitor the history of physical human contacts. As we leave our digital footprints wherever we go, they could in principle be used to track such contacts, to study the disease spreading in real-time, and to provide personalised warnings of infection risk. However, this information is largely unexplored in the public health research community thus far.

My approach

As most people possess a smartphone, a contact between two persons can be regarded as a handshake between the two phones. More importantly, modern smartphone is equipped with multiple sensors that are capable of passively scanning the surroundings. Hence, our task becomes how to detect when the two devices are close by (i.e. co-location of devices).

Over the years, we have proposed several novel approaches to detect such co-location. Our assumption is that, when two mobile devices observe similar time-stamped sensors’ readings, they should be nearby, which in turn, guarantees that their respective owners should also in close proximity.

In 2014, we investigated the possibility of using the outdoor WiFi signals, provided by public Access Points, for off-line mobile phones collision detection. With our idea, we need not maintain a map for the devices, nor require knowing the exact location of the phone at any moment. The challenge with this approach is that even in the same position, two phones may observe different WiFi signal strength and availability, because of different antenna's sensitivity. Below is a video demonstrating our matching rate metric.

magnetic tracking

In 2017, we proposed a novel usage of the geomagnetic field sensor (magnetometer). Our inspiration was that the natural magnetic field generated by the Earth’s core is heavily distorted by the metal bars, ferrous tubes and reinforced concrete, which are commonly found within the building structure. Additionally, an electric current that moves in metal wires (e.g. power lines) will alter the nearby magnetic field. However, this challenge provides a unique opportunity for our purpose. That is, the magnetic field is not uniformly perturbed, so that, different locations experience different magnetism anomalies (see the left figure).

Critically, since every passenger must share the same journey between at least two consecutive stations, which may last up to 10 minutes on the trains or buses, we have a window of opportunity to assess co-location of people.

Below are some videos assessing our system in large scale real-world settings covering 150 kilometres of travelling distance in different parts of London on all types of public transports (i.e. the overground trains, underground tubes, and buses).


Interested researchers are invited to follow our papers and journals.

1) Nguyen, Khuong An, Zhiyuan Luo, and Chris Watkins. "On the feasibility of using two mobile phones and WLAN signal to detect co-location of two users for epidemic prediction." In Progress in Location-Based Services 2014, pp. 63-78. Springer, 2015.

2) Nguyen, Khuong An, Chris Watkins, and Zhiyuan Luo. "Co-location epidemic tracking on London public transports using low power mobile magnetometer." In 2017 International Conference on Indoor Positioning and Indoor Navigation (IPIN), pp. 1-8. IEEE, 2017.

3) Nguyen, Khuong An, Raja Naeem Akram, Konstantinos Markantonakis, Zhiyuan Luo, and Chris Watkins. "Location Tracking Using Smartphone Accelerometer and Magnetometer Traces." In Proceedings of the 14th International Conference on Availability, Reliability and Security (ARES), pp. 1-9. 2019.

4) Nguyen, Khuong An, and Zhiyuan Luo. "Cover your cough: detection of respiratory events with confidence using a smartwatch." In Conformal and Probabilistic Prediction and Applications, pp. 114-131. 2018.