Privacy and Trust Management Schemes of Wireless Sensor Networks: A Survey

Introduction

Security solutions based on cryptography are mainly used to provide protection against security threats, such as fabrication and modification of messages, unauthorized access, etc. For this purpose, assorted security mechanisms such as authentication, confidentiality, and message integrity are used. Additionally, these security mechanisms highly rely on a secure key exchange mechanism [Shaikh et al., 2006a]. However, these cryptography based security mechanisms alone are not capable of providing robustness, reliability and completeness in a security solution. They can only be achieved by incorporating privacy and trust features as described below.

Privacy features such as route anonymity of the data packets, identity anonymity of nodes and their locations are mainly used to provide protection against security threats such as traffic analysis and eavesdropping. Additionally, these privacy features could also be used to provide protection against security threats such as camouflage [Walters et al., 2006]. Therefore, the incorporation of these privacy features with cryptographic-based security mechanisms add to the degree of completeness of a security solution.

Trust management features, such as reputation is used to provide corresponding access control based on the judgment of the quality of sensor nodes and their services [Walters et al., 2006]. Also, it is used to provide complete reliable routing paths which are free from any malicious, selfish and faulty nodes [Liu et al., 2004]. Therefore, incorporation of trust management features with cryptographic-based security mechanisms help in increasing robustness and reliability of the overall security solution.

The soft relationship between privacy, trust, and cryptographic-based security is shown in Figure 1. This figure illustrates the related aspects of these terms with other commonly found terms used in the security domain. For example, secrecy is a mutual feature of cryptographic-based security and privacy aspects. In order to provide secrecy (also referred to as confidentiality), cipher algorithms (such as AES, DES) are used to prevent disclosure of information from any unauthorized entity. Similarly, an intrusion detection system may need a trust management feature such as reputation as well as a cryptographic-based security feature such as integrity checking to detect any malicious nodes. In like manner, solitude, which is used to isolate a node from the network either willingly or forcefully, is a mutual feature of trust and privacy aspects.

Figure 1.

Relationship between privacy, cryptographic-based security and trust

Current research so far seems to intensively focus on the cryptographic-based security aspects of wireless sensor networks. Many security solutions have been proposed such as SPINS [Perrig et al., 2002], TinySec [Karlof et al., 2004], LEAP [Zhu et al., 2003] and LSec [Shaikh et al., 2006b] etc., but surprisingly, less importance is given to privacy and trust issues of wireless sensor networks. Privacy and trust are as important as other security issues and they also contribute in increasing the degree of completeness and reliability of a security solution as discussed above.

In this chapter, we focus on the importance of privacy and trust establishment in wireless sensor networks. In Sections 2 and 3, we present generic and flexible taxonomies of privacy and trust respectively. These taxonomies are based on our specific experience with wireless sensor networks. Apart from these taxonomies, these sections also contain a detailed description of the privacy and trust issues of wireless sensor networks. This description is currently not available in the literature [Chan & Perrig, 2003; Djenouri et al., 2005; Perrig et al., 2004; Shaikh et al., 2006c; Walters et al., 2006]. Also, this chapter contains critical analyses of the state-of-the-art research work. Additionally, this chapter also highlights the challenging problems and issues in the field of privacy and trust in wireless sensor networks. Finally, last section concludes the chapter.