ABSTRACT Intrusion Detection Systems play a key role in detecting cyberattacks within contemporary networks. However, more advanced cyber threats, especially botnet attacks, have revealed certain limitations of conventional Intrusion Detection Systems such as a high rate of false positives and a lack of real‐time detection capability. To address these issues, a DeepBotnet‐Boosted Stacked Bidirectional Autoencoder model is proposed for intrusion detection. The proposed model combines Stacked Bi‐directional Long Short‐Term Memory networks to learn the temporal network traffic pattern using a Deep Autoencoder to produce latent features and improve anomaly detection. A Crossover Boosted Bobcat Optimization algorithm is a bio‐inspired algorithm that balances exploration and exploitation to choose the most relevant network features. Also, the Synthetic Minority Oversampling Technique is employed in preprocessing to overcome the class imbalance issue. The effectiveness of the proposed framework is validated with various publicly available datasets, including the Wireless Sensor Network‐Detection System, the Network Security Laboratory‐Knowledge Discovery in Databases , the University of New South Wales‐Network‐Based 2015, and the Network Intrusion dataset, Canadian Institute for Cybersecurity‐Intrusion Detection System 2017. Experimental results demonstrate that the proposed model outperforms baseline techniques, including the Autoencoder‐Dense‐Transformer Neural Network, the Convolutional Neural Network‐Long Short‐Term Memory, the Graph Neural Networks‐based Network Anomaly Detection, the Gated Attention Dual Long Short‐Term Memory, the Genetic algorithm‐based random forest, and the Convolutional Neural Network‐Gated Recurrent Unit, achieving an accuracy of 98.8%, with a performance improvement of 2%–5% across the evaluated datasets. These results indicate the scalability and real‐time applicability of the hybrid model and optimization‐based feature selection plan to detect botnet intrusion.

Summary Vast usages of Internet of Things (IoT) devices in various smart applications have laid a foundation for the evolution of modern smart cities. The increasing dependency of smart city applications on communication and information technologies enhances operational efficiency, sustainability, and automation of city services. However, due to the heterogeneous nature of IoT devices, the network faces critical security issues while executing continued network operations and services, particularly by cyber‐attacks. One of the predominant and rampant cyber‐attacks in smart city applications is botnet attacks. Therefore, a novel deep learning model for the detection and isolation of cyber‐attacks is proposed in the cloud IoT‐based smart city applications to protect against such cyber‐attacks. The proposed framework utilizes two different modules to automatically detect and isolate the malicious traffic emanating from compromised IoT devices with more efficiency. Here, two different datasets namely the IoT network intrusion and the ISCX 2012 IDs datasets are utilized for the evaluation of the proposed framework. In the first phase, the compromised device which communicates malicious network traffics through the network is identified using a cascaded adaptive neuro‐fuzzy inference system (CANFIS). After detection, IP address of abnormal traffic is recorded and informed to the system administrator. In the second phase, communication pathways of compromised devices with other normal devices are blocked and the compromised devices are isolated from the network using the modified deep reinforcement learning (MDRL) approach. The analytic result shows that the proposed framework achieves a greater accuracy rate of about 98.7% as compared to other state‐of‐art methods.

Recently, Wireless Sensor Networks (WSNs) have emerged to provide a variety of important applications with low cost sensors. The task of the sensors is to collect data and send it to the sink node which delivers the data to a task manager. However, these sensors have limited power and thus limited lifetime. Another important consideration in WSNs is the level of security. Transmitting data from node to another can risk the security of the data. In this paper, we propose a novel trusted and energy efficient routing protocol (TERP) based on the Destination Sequenced Distance Vector Protocol (DSDV). TERP helps to increase the security level in the network and thus avoid any malicious nodes or untrusted nodes. It also reduces the power consumption by using the trust factor. The higher the degree of trust, the less encryption is used which results in less energy. Other factors such as drop ratio, delivery ratio, average delay, and delay jitter are analyzed along a comparison of DSDV protocol with the proposed TERP routing protocol.