- Spring 2019-present
This project is a part of the Sensory Network Infrastructure for Scientific Laboratory Environments (SENSELET) project, which is an NSF-funded project.
The past few years have witnessed the rise of smart shoes, the wearable devices that measure foot force or track foot motion. However, people are not aware of the possible privacy leakage from in-shoe force sensors. In this paper, we explore the possibility of locating an indoor victim based on the force signals leaked from smart shoes. We present an attack scheme that reconstructs the corridor map of the building that the victim walks in based on force data only. The corridor map enables the attacker to recognize the building, and thus locate the victim on a global map. To handle the lack of training data, we design the stair landing detection algorithm, based on which we extract training data when victims are walking in stairwells. We estimate the trajectory of each walk, and propose the path merging algorithm to merge the trajectories. Moreover, we design a metric to quantify the similarity between corridor maps, which makes building recognition possible. Our current experimental results show that, the building recognition accuracy reaches 77.5% in a 40-building dataset, and the victim can be located with an average error lower than 6 m, which reveals the danger of privacy leakage through smart shoes.
- Tuo Yu, Haiming Jin, Klara Nahrstedt, “ShoesHacker: Indoor Corridor Map and User Location Leakage through Force Sensors in Smart Shoes”, Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies, vol. 3(3), pp. 1-29, 2019.
This project is supported by a collaborative award from the National Science Foundation (NSF award numbers CNS-1330491 and OAC- 1827126).