Implanting Life-Cycle Privacy Policies in a Context Database

Ambient intelligence (AmI) environments continuously monitor surrounding individuals’ context (e.g., location, activity, etc.) to make existing applications smarter, i.e., make decision without requiring user interaction. Such AmI smartness ability is tightly coupled to quantity and quality of the available (past and present) context. However, context is often linked to an individual (e.g., location of a given person) and as such falls under privacy directives. The goal of this paper is to enable the difficult wedding of privacy (automatically fulfilling users’ privacy whishes) and smartness in the AmI. Interestingly, privacy requirements in the AmI are different from traditional environments, where systems usually manage durable data (e.g., medical or banking information), collected and updated trustfully either by the donor herself, her doctor, or an employee of her bank. Therefore, proper information disclosure to third parties constitutes a major privacy concern in the traditional studies. On the contrary, AmI is based on autonomous and invisible data collection with weak durability requirements (from donors’ point of view), which puts regulation of context data life-cycle on the hot seat. More precisely, we propose to bind Life-Cycle Policies (LCP) to context data regulating its progressive degradation. This paper makes the following contributions. (i) It introduces the Life Cycle Policy (LCP) model to regulate the content of context databases; (ii) it investigates the problem of correctness of the LCP model when used to implant one-way degradation (i.e., ensure that degraded information can no more be recovered from the current database content); (iii) it implants LCP on top of a traditional DBMS, to provide a practical understanding of the model and show the feasibility of the proposed techniques. Finally, it presents new challenges linked to our approach and concludes the paper. We are convinced that providing LCP on autonomous systems paves the way to new privacy solutions.