Security Issues on Outlier Detection and Countermeasure for Distributed Hierarchical Wireless Sensor Networks

Security Issues on Outlier Detection and Countermeasure for Distributed Hierarchical Wireless Sensor Networks

Yiying Zhang (Shenyang Institute of Engineering, China), Lin He (Korea University, South Korea), Lei Shu (Osaka University, Japan), Takahiro Hara (Osaka University, Japan) and Shojiro Nishio (Osaka University, Japan)
DOI: 10.4018/978-1-61350-110-8.ch009
OnDemand PDF Download:
List Price: $37.50


Outliers in wireless sensor networks (WSNs) are sensor nodes that launch attacks by abnormal behaviors and fake message dissemination. However, existing cryptographic techniques have difficulty in detecting these outliers, which makes outlier recognition a critical and challenging issue for reliable and secure data dissemination when outliers exist in WSNs. This chapter is concerned about detection and elimination problems of outlier. To efficiently identify and isolate outliers, we present a novel “Outlier Detection and Countermeasure Scheme” (ODCS), which consists of three mechanisms: (1) An abnormal event observation mechanism (AEOM) for network surveillance; (2) An exceptional message supervision mechanism (EMSM) for distinguishing fake messages by exploiting spatiotemporal correlation and consistency; (3) An abnormal frequency supervision mechanism (AFSM) for the evaluation of node behavior. The chapter also provides a heuristic methodology which does not need the knowledge of normal or malicious sensors in advance. This property makes the ODCS not only to distinguish and deal with various dynamic attacks automatically without advance learning but also reduces the requirement of capability for constrained nodes. In our solution, the communication is limited to a local range, such as one-hop or a cluster, which can reduce the communication frequency and circumscribe the session range further. Moreover, the chapter also provides countermeasures for different types of attacks, such as the rerouting scheme and the rekey security scheme, which can separate outliers from normal sensors and enhance the robustness of network, even when some nodes are compromised by adversary. Simulation results indicate that our approach can effectively detect and defend the outlier attack.
Chapter Preview

Problem Statement

Outliers (also called Inside attacker) in WSNs are some sensor nodes controlled by adversary, they do not perform tasks as normal nodes but exhibit different types of abnormal behaviors, e.g., dropping received messages from their neighbors, forwarding messages to enemy, broadcasting redundant messages, and disseminating fake messages (Akyildiz et al., 2002), (Liu, Cheng, & Chen, 2007). The outlier has the same network resource as a normal sensor node, but its behaviors are different. Typically, outliers are compromised and remote-controlled by adversary. Outliers also attack WSNs by tampering with messages transferred in WSNs or generating bogus messages and forwarding them to critical nodes (e.g., aggregation nodes or sink node), which typically reduces network performance in terms of reliability and security because of the following consequences (Bandyopadhyay & Coyle, 2003):

  • 1.

    Wasting network bandwidth;

  • 2.

    Increasing energy consumption;

  • 3.

    Interfusing illegal messages into sensory data streaming;

  • 4.

    Causing communication obstruction or dynamic holes.

However, although the outliers seriously threat the network, they are difficult to be detected by traditional cryptographic techniques for the attacks come from network inside (Ash & Moses, 2005). Outliers usually could obtain all or part of the security materials, such as keys, which makes it easy for the outlier to tamper with, inject or eavesdrop messages in network. Thus, it is critical to establish an efficient secure and reliable scheme to detect and prevent outliers. The conventional methods such as encryption, authentication, etc. have the ability to protect the correctness and integrity in WSN. However, they cannot withstand outliers’ attacks.

Complete Chapter List

Search this Book: