Cyber Security for Detecting Distributed Denial of Service Attacks in Agriculture 4.0: Deep Learning Model

Abstract

Attackers are increasingly targeting Internet of Things (IoT) networks, which connect industrial devices to the Internet. To construct network intrusion detection systems (NIDSs), which can secure Agriculture 4.0 networks, powerful deep learning (DL) models have recently been deployed. An effective and adaptable intrusion detection system may be implemented by using the architectures of long short-term memory (LSTM) and convolutional neural network combined with long short-term memory (CNN–LSTM) for detecting DDoS attacks. The CIC-DDoS2019 dataset was used to design a proposal for detecting different types of DDoS attacks. The dataset was developed using the CICFlowMeter-V3 network. The standard network traffic dataset, including NetBIOS, Portmap, Syn, UDPLag, UDP, and normal benign packets, was used to test the development of deep learning approaches. Precision, recall, F1-score, and accuracy were among the measures used to assess the model’s performance. The suggested technology was able to reach a high degree of precision (100%). The CNN–LSTM has a score of 100% with respect to all the evaluation metrics. We used a deep learning method to build our model and compare it to existing systems to determine how well it performs. In addition, we believe that this proposed model has highest possible levels of protection against any cyber threat to Agriculture 4.0.

A Critical Cybersecurity Analysis and Future Research Directions for the Internet of Things: A Comprehensive Review

“…Unauthorized Access Attack [52][53][54][55] The term "unauthorized access" describes when hackers enter a system without authorization. Passwords that are too easy to crack, a failure to safeguard against thought control, hacked credentials, and even insider threats all contribute to the prevalence of these types of attacks.…”

A concept for application of integrated digital technologies to enhance future smart agricultural systems

“…[2,8,29]. Deep learning is being introduced in smart agriculture and used to process images and data and handle complex problems [30]. IoT links different objects through internet using smart technologies such as Bluetooth, wireless sensor network etc.…”

Internet of Things and Wireless Sensor Networks for Smart Agriculture Applications: A Survey

“…Additionally, the limited resources of sensor nodes in WSNs add complexity to implementing robust security measures while maintaining energy efficiency. Intrusion detection systems (IDS) are crucial in identifying and mitigating potential security breaches [69], [538]- [540]. However, adapting IDS to suit the specific characteristics and limitations of IoT-WSNs remains a significant challenge.…”