On privacy and security in distributed visual sensor networks

There is a critical need to provide privacy assurances for distributed vision-based sensor networking in applications such as building surveillance and healthcare monitoring. To effectively address protection and reliability issues, secure networking and processing must be considered from system inception. This paper presents attacks that affect the data privacy in visual sensor networks and proposes privacy-promoting security solutions based on opponent detection via game-theoretic analysis and keyless encryption.

1. INTRODUCTION

In this work, we consider privacy and security mechanisms for a heterogeneous wireless visual sensor network (VSN). The network is comprised of wirelessly communicating camera nodes and scalar sensors where the sensors trigger the cameras and provide specific privacy guarantees based on event detection. The network may be deployed in one or more zones such as throughout a building and its perimeters. The scalar sensor network may contain a number of malicious nodes that have been remotely or locally re-programmed to disinform regarding the presence of an event for the purpose of enabling and disabling camera privacy settings. Upon acquisition of the privacy enabled or disabled frames, camera nodes must encrypt and wirelessly transmit them to a local or off-site sink even if secret keys have been compromised. The surveillance network must thus remain dependable and provide privacy as well as confidentiality of surveillance despite disruption and disinformation activities.