| The rapid progress of development of the Smartphone as well as positioningtechniques has promoted various kinds of Location Based Services, such as the wellknown app Baidu Map and the Dazhong Dianping application. The common point forthose services is that they all need to gather users’ location information first. However,if users’ locations are gathered by some malicious adversaries, they can be faced withthreats of location privacy disclosure. Therefore, more and more people are interestedin this and much research efforts have been spent on this research area.For the most of the privacy protection techniques can be classified into threecategories, named as Fake ID Technique, Fake Location Technique and theAnonymous Space Technique. The Anonymous Space Technique has received lots ofresearch effort since it can both combine good privacy protection ability as well asvery little service influence. The existing solutions for Anonymous Space Techniquebasically form into two architectures. They are the P2P based architecture and theTrusted Third Party architecture. This paper focus on the P2P architecture and try tosolve an existing trust related flaw and also try to propose a new solution calledK-Zone which does not require every mobile user to share its location with each otherbut use a well generated anonymous space instead. This paper also proposes threealgorithms for our K-Zone solution, which are EA Algorithm, WA Algorithm andOEA Algorithm. Through a set of comprehensive experiments, we carefully compareour three algorithms in terms of user density, region area and running time.The main effort of our paper includes:(1) Analysis of current attack models for location privacy in mobile internet.(2) Analysis of current privacy protection techniques’ advantages as well asdisadvantages.(3) Propose the K-Zone solution for the existing trust related flaw which enablesmobile users to share well generated space instead.(4) Simulate the three algorithms in a real world map with well generated moving object data in terms of user density, region area and running time. |
