Demonstrating the Threat of Hardware Trojans in Wireless Sensor Networks
To learn more about the Andrew Young School of Policy Studies and Evidence-Based Cybersecurity Reaserch visit https://aysps.gsu.edu/ and https://ebcs.gsu.edu/.
Abstract
As the demand for cheaper electronic devices has increased, the location of manufacturing foundries has changed, sometimes to untrusted places in foreign countries. Some of these locations have limited oversight of the manufacturing of complicated and sensitive electronic components including integrated circuits (ICs). The integrated circuits are key component in all current electronic devices and can be modified to be malicious or to monitor the functions of their applications. These malicious modifications on the ICs are called hardware trojans (HWTs). HWTs an be designed to quietly monitor, to actively send out sensitive information, or to destroy their host device completely. The idea of hardware trojans in Wireless Sensor Networks (WSNs) has not been investigated before; thus, our goal is to demonstrate the potential threat that hardware trojans pose for sensor networks. This is important to study, given that in WSNs hundreds of sensors are deployed and in most cases left unattended, which gives the opportunity to an attacker to trigger a HWT on the sensors. For our investigation, we used TelosB sensors that have been used for some WSN applications. An attacker in a network can, for example, take advantage of the SPI bus that is used by the radio to eavesdrop messages and even disrupt communications completely. Currently, security breaches through software is given great importance in the WSN academic and research community. Our research shows that the same level of importance must be given to attacks through hardware to ensure a trusted and secure network.