Collusion Resistant Inference Control for Cadastral Databases

Abstract

Abstract. In this paper we present a novel inference control technique, based on graphs, to control the number of accessible parcels in a cadastral database. Different levels of collusion resistance are introduced as part of the approach. The dynamic aspect of the cadastral application, caused by mutation operations, is handled. We propose a scheme for gradually resetting the inference graph allowing continuous access to the data.

Quantity based aggregation for cadastral databases

“…Section 3 defines the security policy and presents the new model. Section 4 compares the new model to our previous work [2,3], and provides a detailed analysis on why dominant zones achieve higher availability. Section 5 describes the application workflow, the role of QBA control in that workflow, and authentication.…”

“…In [2,3] we proposed a first model and its implementation. The main contribution of this paper is a new version of the model to achieve more availability on cadastral data where we introduce "Dominant Zones" (Definition 5) that requires less storage space, provides more availability, and gives more flexibility to the security administrator controlling the trade-off between confidentiality and availability of data.…”

“…The only work we are aware of, treating QBA directly, is that of Motro, Marks and Jajodia [21,23] twenty years ago. In a previous paper [2] we addressed two QBA problems in the context of cadastral databases. In fact, we proposed a solution to enforce the following security policy: In the cadastral database, any user has the right to know the owner's name of any given parcel.…”

“…Solutions presented for P r1 can be easily applied to P r2 with minimal modifications due to the similarity of both problems. For more details about the legislative context, the reader is invited to read [2]. This paper is organized as follows (we would like to stress out the fact that all sections, except Section 3.1, are original contributions): Section 2 gives some preliminary definitions that help identifying aggregation problems.…”

“…In this paper we work on QBA problems in the context of cadastral databases, and we focus on one particular problem: how to prevent a user from accessing all parcels in a region. This work is a new version of the model presented in [2,3] where we introduce the concept of "Dominant Zones". This concept increases the availability of the data while preserving their confidentiality.…”