Uniqueness Assessment of Human Mobility on Multi-Sensor Datasets - Archive ouverte HAL Access content directly
Reports (Research Report) Year : 2016

Uniqueness Assessment of Human Mobility on Multi-Sensor Datasets


The widespread adoption of handheld devices (e.g.,smartphones, tablets) makes mobility traces of users broadlyavailable to third party services. These traces are collected bymeans of various sensors embedded in the users’ devices, includ-ing GPS, WiFi and GSM. We study in this paper the mobility of300 users over a period up to 31 months from the perspective ofthe above three types of data and with a focus on two cities, i.e.,Lausanne (Switzerland) and Lyon (France). We found that users’mobility traces, no matter if they are collected using GPS, WiFi orGSM antennas, are highly unique. We show that on average onlyfour spatio-temporal points from the WiFi, GSM and GPS tracesare enough to uniquely identify 94% of the individuals, on bothdatasets. In addition, we show that using the temporal dimension(i.e., whether users move or are in a meaningful location such asthe home or the working place) drastically improves the capacityto uniquely identify users compared to when only exploiting thespatial dimension (by 14% on average). In some cases, usingthe temporal dimension alone can represent a better mobilityfootprint than the spatial dimension to discriminate users. Wefurther conduct a de-anonymisation attack to assess how mobilitytraces can be re-identified, and show that almost all userscan be de-anonymised with a high success rate. Finally, weapply different location privacy protection mechanisms (LPPMs),applying spatial filtering, temporal cloaking, adding spatial noiseto mobility data, or using generalisation, and analyse the impactof these mechanisms on both the uniqueness of users’ mobilitytraces and the outcome of the de-anonymisation attack. We showthat spatially obfuscating mobility data is not enough to protectusers, and that classical LPPMs are not able to protect usersagainst a de-anonymisation attack. We finally conclude this paperby drawing some insights towards future spatio-temporal LPPMs.
Fichier principal
Vignette du fichier
paper.pdf (2.58 Mo) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-01381986 , version 1 (08-12-2016)


  • HAL Id : hal-01381986 , version 1


Antoine Boutet, Sonia Ben Mokhtar, Vincent Primault. Uniqueness Assessment of Human Mobility on Multi-Sensor Datasets. [Research Report] LIRIS UMR CNRS 5205. 2016. ⟨hal-01381986⟩
422 View
323 Download


Gmail Facebook Twitter LinkedIn More