Location-dependent disclosure risk based decision support framework for persistent authentication in pervasive computing applications

In pervasive computing applications (e.g. electronic health records), the amount of information permissible to be shared or accessed by mobile users results in high disclosure risks. Obfuscation techniques are desirable in reducing the impact of disclosing confidential information but with a significant loss of utility of information content. Thus, accesses to confidential data by mobile users need to be controlled so as to minimize the disclosure risks. To achieve these requirements, we propose a novel location-dependent disclosure risk based decision support framework for persistent authentication and data access management. We have derived the location dependent identity based disclosure risks at record level and file level by using the search theory and entropy. We have experimentally evaluated our proposed model using multi-level security model and fuzzy sets. We have further proved that our proposed technique can significantly reduce the impact of common privacy attacks by performing a comprehensive security analysis. In conclusion, this research presents a novel location-dependent disclosure risk-based decision support framework persistent authentication and a pragmatic data access management approach for highly privacy-sensitive pervasive computing applications.