Wireless Sensor Network Anomalies: Diagnosis and Detection Strategies

Abstract

Abstract. Wireless Sensor Networks (WSNs) can experience problems (anomalies) during deployment, due to dynamic environmental factors or node hardware and software failures. These anomalies demand reliable detection strategies for supporting long term and/or large scale WSN deployments. Several strategies have been proposed for detecting specific subsets of WSN anomalies, yet there is still a need for more comprehensive anomaly detection strategies that jointly address network, node, and data level anomalies. This chapter examines WSN anomalies from an intelligent-based system perspective, covering anomalies that arise at the network, node and data levels. It generalizes a simple process for diagnosing anomalies in WSNs for detection, localization, and root cause determination. A survey of existing anomaly detection strategies also reveals their major design choices, including architecture and user support, and yields guidelines for tailoring new anomaly detection strategies to specific WSN application requirements.

Nano-scale reservoir computing

“…ESNs with a one-dimensional topography have been investigated in [9], and a topography where several (randomly connected) reservoirs are regularly connected in a grid-like fashion using a wireless sensor network is used in our previous work [10]. Despite these constraints on connectivity between hidden units, e.g., prediction of future states is possible with a reservoir computing method -in the case of sensor networks with one of the applications to detect anomalies [11]. Use of piecewise linear approximations of the hyperbolic tangent transfer function in individual units, similar to our nonlinearities in Figs.…”

Detecting Malicious Data Injections in Wireless Sensor Networks

“…The closest survey to the one presented here is [Jurdak et al 2011]. It describes anomaly detection strategies for detecting faults due to environmental factors (e.g.…”

“…Their description of anomaly detection is similar to ours but the two surveys differ notably in the nature of the anomalies considered: attacks in our case, faults in theirs. Jurdak et al [2011] also claim that anomalies can be detected by spatial or temporal comparisons between sen-sors, since it is unlikely that many sensors will exhibit a calibration skew or failure at the same time (assuming there are no group failures). This assumption considers anomalies (faults) as independent but does not hold in the presence of malicious data injections, in particular when there is collusion between the compromised sensors.…”

Congestion control in wireless sensor networks based on bird flocking behavior

“…Firstly, such approach leads to excessive communication load in the network which rapidly depletes the batteries [1]. Secondly, centralized decision making for CC results in slow reaction to changes in network and traffic conditions [3]. Thirdly, a centralized approach does not take advantage of the in-network processing capability of WSNs which permits simple processing and decisionmaking by individual nodes [4].…”