Battery Polling and Trace Determination for Bluetooth Attack Detection in Mobile Devices

Abstract

This paper introduces a supporting model for a (Bluetooth) and IEEE 802.11 (Wi-Fi) attack activity that unique Battery-Sensing Intrusion Protection System (B-SIPS) for standard IDSs are incapable of detecting [1]. mobile computers, which alerts when power changes are detectedThis research further examines various means to refine the on small wireless devices. An analytical model is employed to B-SIPS detection capabilities. Smart battery polling rates, examine the smart battery characteristics to support the theoretical intrusion detection limits and capabilities of B-SIPS. system management bus speeds, and attack execution times This research explores the modification of the smart battery can be used to improve the theoretical accuracy of batterypolling rates in conjunction with the variance of malicious based anomaly detection. A methodology is then presented to network activity. Using the results from a previous study of identify an attack, using data sampled by an oscilloscope that optimized static polling rates to create minimum and maximum is filtered to characterize a unique trace signature. thresholds, a dynamic polling rate algorithm was devised. ThisThe rest of this paper is structured as follows. Section II algorithm allowed the smart battery to gauge the network's illicit attack density and adjust its polling rate to efficiently detect tpresents related work. Section III describes the ongoIng attacks, while conserving battery charge life. Lastly, a trace technology convergence trend in mobile devices. Secton IV signature methodology is presented that characterizes unique presents B-SIPS' implemented capabilities. Section V activity for IEEE 802.15.1 (Bluetooth) attack identification. discusses the smart battery polling model's design and outlines previous work on testing and optimizing static polling rate Index Terms-Battery, Intrusion Detection, Wireless Security solutions. Section VI utilizes the static polling results in formulating a dynamic solution and evaluating it in respect to

Bluetooth Network-Based Misuse Detection

“…[36] Buennemeyer et al provided a means of detecting Bluetooth battery depletion attacks using an anomaly detection system that measured the power levels of a particular Bluetooth device. [23] …”

“…Researchers at Virginia Tech have shown how to combine classic Internet protocol attacks such as the SYN flood with a Bluetooth distribution scheme. [23] As the number of attacks have grown, so have the severity of attacks and the ease of implementation. Repositories of attacks exist with source code.…”

Simulating the Deployment of Battery-Sensing Intrusion Protection Systems

“…Until recently, many Bluetooth device users considered their systems to be safe from attack. Table 1 [14] indicates that Bluetooth exploits and emerging attacks are increasing. The ever changing state of attack vectors has opened another avenue for attack signature development, which encompasses the characterization of Bluetooth wireless personal area network (WPAN) attacks.…”

“…This will offer greater opportunities to develop battery-based trace signatures, since Bluetooth capable devices tend to be of low powered design. Attacking exposed hosts through unsecured WPANs would allow the attacker direct access to the mobile device and its OS environment, completely bypassing any upstream defensive measures [14]. This observation suggests that mobile devices have an increasing need for hybrid IDS protection such as B-SIPS.…”

“…While considering plausible attacks that could disrupt PDA and smart phone operations, three original DoS attacks were crafted in the lab: BlueSYN, BlueSYN Calling DoS and PingBlender [14]. These blended attacks were designed to saturate the test devices multiple communication channels capabilities to hasten the drain time of its battery resources.…”