Towards Secure and Practical Targeted Mobile Advertising

Abstract

Mobile advertising has become a key economic force with the growing mobile app markets. Targeted advertising has been attracting the attention of business and research communities because of its high return on investment. However, the fact that it requires continuous collection of user profiles raises severe privacy concerns. In this paper, we propose a secure and practical mobile advertising architecture, which aims to enable targeted advertising over encrypted mobile user profiles. The proposed idea is based upon a cryptographic primitive, i.e., private stream searching (PSS). We first propose a basic construction of secure targeted mobile advertising. Thorough analysis shows that our basic design achieves semantic security. After that, we introduce a number of techniques to make our design more efficient for mobile devices. Experimental results demonstrate the practicality of our proposed architecture.