Classification of botnet attacks in IoT smart factory using honeypot combined with machine learning>
Peer J Computing – Seungjin Leeâ, Azween Abdullah, Nz Jhanjhi, Sh Kok
he Industrial Revolution 4.0 has brought a great innovation to the conventional manufacturing into the new era of smart factories (Oztemel & Gursev, 2020). A demand for the IoT equipment in smart factories has been increasing since 2012 as shown in Fig. 1. Especially in the last 5 years (2015â2020), the use of IoT devices has increased tremendously from 18.2 billion to 50 billion for application in the smart factories (Smith, 2015). Additionally, as smart factories are combined with Information and Communications Technology (ICT), all the facilities and devices are connected at the central wireless communication. Although IoT smart factories have been built and operated in the industry, standards of implementation for smart factories have yet to be established (Guo et al., 2020). With four layers arranged hierarchically, it starts at the physical resource layer, followed by the networking layer and the application layer, and ends at the terminal layer. >From the security perspective, research should focus more on the physical resource/sensing layer, as it is directly related to the vast usage of the IoT devices in order to reinforce the security network for smart factories. To mitigate the impact of data leakage and data abuse, real-time detection of cyber attacks to smart factory obviously becomes an extremely important factor to take into consideration of developing and improving security network of the smart factory (Brett et al., 2009).

Network security in the smart factory is highly at risk of being under cyber attacks due to the interconnection of a huge number of IoT equipment. According to a recent report, instability is recognized as one of the biggest limitations out of 250 vulnerable features found in the IoT devices (Casalinuovo, 2019). One of the most serious cyber attacks to smart factory is botnet. Among many detection methods, honeypot has been investigated to apply for detecting botnet attack in various studies in the recent years (Jaâfari et al., 2020). However, a huge volume of attacking data collected by honeypot is highly complex and non-classified. This ith an increasing interest in the potential application of machine learning, it offers a new solution for detecting abnormalities in the malicious Internet traffic. Applying machine learning in botnet detection for smart factories can become useful to enhance performance of the honeypot model in term of speeding up the processing time or detection time (Lim et al., 2019). Interestingly, there have been very few studies making attempts to mount both honeypot and machine learning on IoT device networks to target attacks on the IoT traffic.

Another approach to detect botnet is using machine learning which gave a high accuracy in detection at 91.66% (Wang et al., 2020). However, one disadvantage of using machine learning approach is that fast detection is hard to achieve in the randomized number of packets. Consequently, the feasibility of applying this approach for smart manufacturing needs more research looking into real-time factor and accuracy.

Simulation of hardware design with raspberry Pi transfer log files
Link: https://peerj.com/articles/cs-350/