Paper Info
Title
SALS: semantics-aware location sharing based on cloaking zone in mobile social networks
Abstract
There is a potential privacy breach when users access various location-based social applications on a mobile social network (MSN), e.g., sharing locations with friends. To preserve location privacy, one of the most common methods is to use a coarse or fake location instead of a user's exact location. However, most of these previous approaches only provide geometric strategies without considering the semantic context of the geographical locations. For example, if a cloaked region contains a part of a lake, where no boats are allowed, an adversary can easily prune the cloaked region to a smaller range covering a user's actual location. In this paper, we propose SALS, a semantics-aware location sharing framework based on cloaking zone for an MSN environment. By considering a user's social relations and activities which are available in an MSN environment, SALS does not assume any trustworthy entities, including strangers, friends or any third parties. As a solution, SALS enables users to cooperate with each other, in a Peer-to-Peer (P2P) way, to generate the cloaking zones, which will be used instead of the actual locations. Different from the previous cloaking techniques, SALS considers the semantic location which can influence the distribution probability of a user's locations. We also propose metrics for measuring the quality of the cloaking zone. The evaluation shows that our method can well defend the semantic-location attack.
Year
DOI
Venue
2012
10.1145/2442810.2442820
MobiGIS
Keywords
Field
DocType
location privacy,semantics-aware location sharing framework,semantic location,cloaked region,mobile social network,geographical location,fake location,msn environment,actual location,exact location,cloaking zone
Cloaking,Social network,Mobile social network,Computer security,Trustworthiness,Computer science,Probability distribution,Adversary,Location sharing,Semantics
Conference
Citations 
PageRank 
References 
5
0.41
14
Authors
4
Name
Order
Citations
PageRank
Yanzhe Che1353.11
Kevin Chiew211611.06
Xiaoyan Hong31665163.41
Qinming He437141.53