Malware and malicious code do not only incur considerable costs and losses but impact negatively the reputation of the targeted organisations. Malware developers, hackers, and information security specialists are continuously improving their strategies to defeat each other. Unfortunately, there is no one‐size‐fits‐all solution to detect and eradicate any malware. This situation is aggravated more by the undetected vulnerabilities that usually impair computer software and internet tools. Such vulnerabilities will remain undetected until fully exploited by malware developers, which will eventually cause considerable financial and reputation losses. In this paper, we propose a novel scheme to detect and classify malware using only image representations of the malware binaries. Highly discriminative features of the malware category and structure are extracted in a compact subspace using principal component analysis. Then, an optimised support vector machine model classifies the extracted features into malware categories. Unlike existing classification models, our solution requires simple algebraic dot products to classify malware based on representative digital images. To assess its performance, publicly‐available image datasets, Malimg , Ember and BIG 2015 , are considered. Our performance analysis indicates that their classifier outperforms state‐of‐the‐art models and attains classification accuracies of 0.998, 0.911, and 0.997 using Malimg , Ember and BIG 2015 malware datasets, respectively.

The increasing sophistication of cyber threats has led to the evolution of Security Operations Centers (SOCs) towards more intelligent and adaptive systems. This review explores the integration of Artificial Intelligence (AI) in SOCs, focusing on their current state, challenges, opportunities, and advantages over traditional methods. We address three key questions: (1) What are the current AI approaches in SOCs? (2) What challenges and opportunities exist with these approaches? (3) What benefits do AI models offer in SOC environments compared to traditional methods? We analyzed 38 studies using a structured methodology involving database searches, quality checks, and data extraction. Our findings show that Machine Learning (ML) techniques dominate SOC research, with a trend towards multi-approach AI methods. We classified these into ML, Natural Language Processing, multi-approach, and others, forming a detailed taxonomy of AI applications in SOCs. Challenges include data quality, model interpretability, legacy system integration, and the need for constant adaptation. Opportunities involve task automation, enhanced threat detection, real-time analysis, and adaptive learning. AI-driven SOCs show better accuracy, reduced false positives, greater scalability, and predictive capabilities than traditional approaches. This review defines Cognitive SOCs, emphasizing their ability to mimic human-like processes. We offer practical insights for SOC designers and managers on implementing AI to improve security operations. Finally, we suggest future research directions in explainable AI, human-AI collaboration, and privacy-preserving AI for SOCs.

The Internet of Things (IoT) links various devices to digital services and significantly improves the quality of our lives. However, as IoT connectivity is growing rapidly, so do the risks of network vulnerabilities and threats. Many interesting Intrusion Detection Systems (IDSs) are presented based on machine learning (ML) techniques to overcome this problem. Given the resource limitations of fog computing environments, a lightweight IDS is essential. This paper introduces a hybrid deep learning (DL) method that combines convolutional neural networks (CNN) and long short-term memory (LSTM) to build an energy-aware, anomaly-based IDS. We test this system on a recent dataset, focusing on reducing overhead while maintaining high accuracy and a low false alarm rate. We compare CICIoT2023, KDD-99 and NSL-KDD datasets to evaluate the performance of the proposed IDS model based on key metrics, including latency, energy consumption, false alarm rate and detection rate metrics. Our findings show an accuracy rate over 92% and a false alarm rate below 0.38%. These results demonstrate that our system provides strong security without excessive resource use. The practicality of deploying IDS with limited resources is demonstrated by the successful implementation of IDS functionality on a Raspberry Pi acting as a Fog node. The proposed lightweight model, with a maximum power consumption of 6.12 W, demonstrates its potential to operate effectively on energy-limited devices such as low-power fog nodes or edge devices. We prioritize energy efficiency while maintaining high accuracy, distinguishing our scheme from existing approaches. Extensive experiments demonstrate a significant reduction in false positives, ensuring accurate identification of genuine security threats while minimizing unnecessary alerts.

Unmanned Aerial Vehicles (UAVs) have gained significant popularity in both military and civilian applications due to their cost-effectiveness and flexibility. However, the increased utilization of UAVs raises concerns about the risk of illegal data gathering and potential criminal use. As a result, the accurate detection and identification of intruding UAVs has emerged as a critical research concern. Many algorithms have shown their effectiveness in detecting different objects through different approaches, including radio frequency (RF), computer vision (visual), and sound-based detection. This article proposes a novel approach for detecting and identifying intruding UAVs based on their RF signals by using a hierarchical reinforcement learning technique. We train a UAV agent hierarchically with multiple policies using the REINFORCE algorithm with entropy regularization term to improve the overall accuracy. The research focuses on utilizing extracted features from RF signals to detect intruding UAVs, which contributes to the field of reinforcement learning by investigating a less-explored UAV detection approach. Through extensive evaluation, our findings show the remarkable results of the proposed approach in achieving accurate RF-based detection and identification, with an outstanding detection accuracy of 99.7%. Additionally, our approach demonstrates improved cumulative return performance and reduced loss. The obtained results highlight the effectiveness of the proposed solution in enhancing UAV security and surveillance while advancing the field of UAV detection.