ETAREE: An Effective Trend-Aware Reputation Evaluation Engine for Wireless Medical Sensor Networks

Abstract

Wireless Medical Sensor Networks (WMSN) will play a significant role in the advancements of modern healthcare applications. Security concerns are still the main obstacle to the widespread adoption of this technology. Conventional security approaches, such as authentication and encryption, are able to defend against external attacks effectively. However, internally launched threats, either by compromised or selfish nodes, require further security measures to be detected. In this paper, an Effective Trend-Aware Reputation Engine (ETAREE) is proposed for WMSN. ETAREE uses a novel updating mechanism to evaluate the reputation value, which makes it effective in detecting malicious nodes. Moreover, the proposed updating mechanism of ETAREE can efficiently detect on-off attacks. ETAREE security evaluations have been presented and compared with different reputation evaluation models, demonstrating faster detection of malicious behaviours.

LTMS: A Lightweight Trust Management System for Wireless Medical Sensor Networks

“…The traditional updating mechanism uses a single weight exponential smoothing technique, which fails to reflect any sudden malicious behavior fast because the evaluated reputation value represents the long-term expected value of the probability distribution. Therefore, it needs more time to detect any malicious behavior [9]. This drawback may be exploited by a smart adversary to launch complicated attacks such as on-off attacks.…”

“…In [9], we have introduced a novel updating mechanism to allow fast detection of any behavior change. Although the proposed method shows prompt reaction to any sudden behavior change, smart adversaries can take advantage of the model dynamicity to launch complicated on-off attacks.…”

“…On the other hand, although TMSs show promising solution to detect packet drop attacks and other misbehaviors, they can be gamed by intelligent adversaries. TMSs are prone to on-off attacks, where adversary changes his behavior between good and bad alternately in order to redeem himself from the burden of bad behavior [9]. Moreover, adversaries can launch more sophisticated on-off attacks by changing the packet drop rates or launching on-off attacks with non-identical periods.…”

A survey on wireless body area networks: architecture, security challenges and research opportunities

“…• Sinkhole Attack: The malicious or the compromised node attempts to attract all the traffic inside the WBAN and then drop it [41]. The adversary can run this kind of attack by sending fake routing updates showing itself as the shortest path to the medical server.…”

TrustMod: A Trust Management Module For NS-3 Simulator

“…The first method LTMS(1), is a lightweight method used for inbody sensor nodes (SNs) as those SNs suffer from a stringent resource limitation, while LTMS(2) provides a further level of protection from on-off attacks and is designed for on-and off-body sensor nodes. Both LTMS(1) and LTMS(2) method are combined in algorithm 1, where α and β are the beta probability distribution levels, b t and d t are the slopes at the time unit t, Rep ij (t) is the reputation value maintained by the trustor i for the trustee j, thr1 is the defined threshold to differentiate between trustworthy and untrustworthy agents, which is usually set to 0.5 in the literature [17]- [20], thr2 represents the minimum trustworthiness for agents in normal operation and is set to 0.85 [18], ShRep ij (t) represents the short-term reputation value at the time unit t, and cycle and malicious are two parameters used to detect on-off attacks.…”