| Title | On-line Detection and Localization of DoS Attacks in NoC |
| Publication Type | Conference Paper |
| Year of Publication | 2020 |
| Authors | Yao, Jiaqi, Zhang, Ying, Mao, Zhiming, Li, Sen, Ge, Minghui, Chen, Xin |
| Conference Name | 2020 IEEE 9th Joint International Information Technology and Artificial Intelligence Conference (ITAIC) |
| Date Published | dec |
| Keywords | denial of service, Hardware, Hardware Trojans, Internet of Things, Metrics, network on chip, network on chip security, pubcrawl, random forest algorithm, random forests, Real-time Systems, Resiliency, Scalability, security, system-on-chip, Trojan horses |
| Abstract | Nowadays, the Network on Chip (NoC) is widely adopted by multi-core System on Chip (SoC) to meet its communication needs. With the gradual popularization of the Internet of Things (IoT), the application of NoC is increasing. Due to its distribution characteristics on the chip, NoC has gradually become the focus of potential security attacks. Denial of service (DoS) is a typical attack and it is caused by malicious intellectual property (IP) core with unnecessary data packets causing communication congestion and performance degradation. In this article, we propose a novel approach to detect DoS attacks on-line based on random forest algorithm, and detect the router where the attack enters the sensitive communication path. This method targets malicious third-party vendors to implant a DoS Hardware Trojan into the NoC. The data set is generated based on the behavior of multi-core routers triggered by normal and Hardware Trojans. The detection accuracy of the proposed scheme is in the range of 93% to 94%. |
| DOI | 10.1109/ITAIC49862.2020.9338861 |
| Citation Key | yao_-line_2020 |
On-line Detection and Localization of DoS Attacks in NoC