Privacy Preserving in Ubiquitous Computing: Architecture Information Technology Journal 8 (6): , 2009 Tinghuai Ma, Sen Yang, Wei Tian, Wenjie Liu 報告者 : 許睿中 日期 :10.18

Outline Introductions The Structure of PAS Spatiotemporally-based anonymous Service matching Random coordinates Fixed coordinates Discussion Conclusion

Introduces Ubiquitous computing represents the concept of seamless everywhere computing and aims at making computing and communication essentially transparent to the users. Personal informational con be divided into two part: ◦ identification ◦ Profile

The Structure of PAS Focus on three main principle in our Privacy-Aware System(PSA): ◦ Anonymity( 匿名性 ) ◦ Untraceability( 不留痕跡 ) ◦ Confidentiality( 機密性 )

The Structure of PAS Access base ◦ Is a software and responsible to deal with the query. ◦ The access base is separated from privacy system, because it can deal with all entities with no authentication. ◦ It will respond entity’s query quickly, as avoiding complexity authentication in privacy system. ◦ At the same time, the malicious entities will be filtrated outside the privacy system.

The Structure of PAS Entity Privacy system Authentication center Service provider Access base trustful domain distrustful domain 1.RSN and query service service list 4.the list 5.e-RSN and PPL 6.identification 7.ps-RSN 8.verify its validity sending the ps-RSN 9.notarization taking ps-RSN as identification 10.PPL 11.apply the service utilizing the ps-RSN 12.teansfers service

System divides time into pieces, carves out the entire domain into some certain areas. When a person’s random data stream is due or out of the confined area, the random data stream updates itself and distributes a new data stream to represent the person, then invalidates the old random data stream. Spatiotemporally-based anonymous

1:00 The prescriptive time The time interval is 30 min The RSN refreshed itself

Spatiotemporally-based anonymous 1:00 More than 30 min here The time interval is 30 min The RSN refreshed itself

Service matching All the services in ubiquitous computing have the limits of spatial and temporal. Only the service’s spatial and temporal preferences strictly match with the entity’s requirements, the service is valid for entity.

Service matching Example: Service: Entity: The service match can be processed as follow: If mSer is true, the service matching is successful.

Random coordinates The service is available in a relatively large domain, the privacy system mask the entity’s precise spatiotemporal information, so system randomly chooses a geographic location coordinates within the valid area and sent it to the relevant SP. Service: Entity: The privacy system can disturb the entity’s location information to random data as:

Fixed coordinates Privacy system provides a fixed coordinate to whoever enters the service area regardless of any different coordinate, if the enters are in the service valid area.

Discussion Anonymity: ◦ First, user are anonymously while in querying, applying and utilizing service, as only RSNs used to indicate the users. ◦ Secondly, the RSNs can be changed using a spatiotemporally-based anonymous strategy. Protection of user preference: ◦ We assume the AC is trustful. So, we can say users’ preference won’t be leaked.

Discussion Non-linkability: ◦ Each session, every users has different temporal RSN as identifier under spatiotemporally-based anonymous. Performance evaluation: ◦ We divide the steps into two processes, one is query service process, which step 1 to 4. The other is applying service process, which include step 5 to 12.

Conclusion This study is the first step to integrate privacy protection technologies into access control architecture for ubiquitous computing.