Abstract
The tremendous number of Internet of Things (IoT) devices and their widespread use have made our lives considerably more manageable. At the same time, however, , the vulnerability of these innovations means that our day-to-day existence is surrounded by insecure devices, thereby facilitating ways for cybercriminals to launch various attacks by large-scale robot networks (botnets) through IoT. This problem is further heightened by the constraints of the IoT on security techniques due to limited resources including central processing units (CPUs), memory, and power consumption. In consideration of these issues, we propose a lightweight neural network-based model to rapidly detect IoT botnet attacks. The model was developed using FastGRNN algorithm which is a lightweight and fast version of the recurrent neural network. In addition, it is independent and does not require any specific equipment or software to fetch the required features for learning and detection processes. Therefore, only packet headers are required to complete learning and detection. Furthermore, the model provides multi-classification, which is necessary for taking appropriate countermeasures to understand and stop the attacks. According to the conducted experiments, the proposed model is accurate and achieves 99.99%, 99.04% as F1 score for MedBIoT and Mirai-RGU datasets in addition, to fulfilling IoT constraints regarding complexity and speed. It is less complicated in terms of computations, and it provides real-time detection that outperformed the state-of-the-art, achieving a detection time ratio of 1:5 for MedBIoT dataset and a ratio of 1:8 for Mirai-RGU dataset.
Highlights
The dominant features of the modern era can be illustrated by the abundant data that are collected and monitored via Internet of Things (IoT) devices, as well as by the endless functionalities enabled by this innovation
The results proved that using FastGRNN [14] provided high speed for training the model and detecting attacks, with much less complexity compared to the state-of-the-art while preserving a high F1 score, where it attained a score of 99.04% with the RGU dataset [8] in comparison to the gated recurrent unit (GRU) model’s 97.82%, and the long short-term memory (LSTM) model’s 98.60%
The model was compared with the LSTM-based model proposed by [9], but because there is no published information regarding time or the MedBIoT dataset in the paper by [9], we implemented their model and trained it ourselves
Summary
The dominant features of the modern era can be illustrated by the abundant data that are collected and monitored via Internet of Things (IoT) devices, as well as by the endless functionalities enabled by this innovation. The hack value of IoT devices is not confined to the critical information stored, collected, or monitored by these technologies but extend to any other assets that can be breached via large-scale botnets. This problem is further exacerbated by the fact that the IoT ecosystem imposes constraints on security techniques because of limited resources with respect to central processing units (CPUs), memory, and power consumption. After breaking into an unprotected gadget, the bot embeds itself into the equipment and waits for instructions from a botmaster to perform malicious activities An example of these attacks is the collaborative flooding of a target (an IoT or non-IoT device) with numerous illegitimate requests, preventing the device from processing legitimate ones and causing a distributed denial-of-service (DDoS) attack. The other ill-intentioned activities of IoT botnets [2] include cryptocurrency mining, password cracking, and email spam sending, keylogging
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
More From: International Journal of Advanced Computer Science and Applications
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.