Abstract

Nowadays, people can access a wide range of applications and services for mobile device users. Among them, location-based services (LBS), where the application needs the user's position to provide the service. Some examples of these applications are Uber and Waze. Nevertheless, the repetitive use of an LBS can reveal confidential user information; thus, behavior patterns-such as daily routes-could be deduced by some dishonest LBS. Furthermore, a query's keywords could provide information about a user's health status or future position when it inquires about hospitals or hotels. Therefore, an adversary can use this information for unethical purposes, and users need mechanisms that protect their privacy. At present, several approaches separately tackle location privacy, location security, and query privacy. To the best of our knowledge, no previous work deals with all these mentioned aspects simultaneously especially when users demand continuous protection when moving and accessing an LBS. This paper proposes two batch techniques to provide location privacy, location safety, and query privacy in an environment that considers a continuous LBS. These techniques apply l-diversity (query privacy) in a context that contemplates query semantics, as well as a diverse set of users' paths. Extensive experimentation shows that both techniques are cost-effective and scalable solutions that offer unified location privacy, query privacy, and location safety protection for many mobile users.