An integrated multimedia data reduction and content confidentiality approach for limited networked devices

In this paper, we specifically address the problem of securing distributed multimedia systems, based on networked tiny devices (i.e., Internet of Things). Such systems have usually to deal with three antagonistic constraints: (a) the hardware limitations of the devices in terms of computational power, memory storage and energy provision as they are driven by batteries, (b) the intrinsic voluminous and time sensitive nature of the captured multimedia data, and (c) their relative vulnerability to attacks since they are open platforms communicating through unreliable wireless radio communications. We propose and study a novel integrated approach, specifically tailored to significantly reduce the size of the transmitted multimedia data whilst ensuring its content confidentiality. In opposition to traditional approaches, based on cryptographic systems which inquire a huge overhead when applied to multimedia data, our approach makes use of Voronoi tessellation to transform the input data. This transformation, performed on a random fashion basis, allows both significant data reduction and content confidentiality as well, while it exhibits a very low complexity. We performed an in-depth security analysis of our proposal that reveals its robustness against all known types of attacks. Moreover, we conducted an extensive series of simulations, based on realistic settings, to show the effectiveness of our approach over state-of-the-art approaches, in the context of Wireless Multimedia Sensor Networks (WMSNs) applications.