Paper #:[4.3.1 Privacy] # 8
Title: "Location Privacy in Pervasive Computing" by Alastair R. Beresford and Frank Stajano 

PROBLEM
As location-based services become more prevalent, the privacy and security of these technologies also come to the forefront.  A solution is needed that allows users to benefit from location-based applications without giving up their location privacy.  While using pseudonyms to hide user identities is a step in  the right direction, a persistent "hostile observer" may still extract privacy information from existing data.

NEW IDEAS AND STRENGTHS
*The solution utilizes an anonymizing proxy in order to ensure privacy during communication between users and applications.
*The paper makes a good point that long term pseudonyms don't work in office settings because one can determine based upon which desk the user spends the most time, the identification of the user.
*The concept of mix zones is used to improve privacy by having a user be connected to a group of users in a spatio-temporal region.  The zone can't be to much larger than the area a user can cover over one location update period.  Otherwise, the user will be at risk.
*It is a good idea to allow the user to refuse to give his location if the minimum amount of anonymity does not exist.

WEAKNESSES AND EXTENSIONS
*With respect to the anonymity set, anonymity may be underestimated because new users who haven't visited the zone aren't included.  Therefore an accurate portrayal of anonymity is not given.
*An extension needs to be made to ensure that the anonymity set includes members that are equally likely to be of interest to "hostile observers".