The proliferation of Internet of Things (IoT) devices has introduced significant security challenges, making IoT networks prime targets for cyberattacks. Traditional security mechanisms often fall short in addressing the complexity and scale of these threats. This paper proposes an enhanced Intrusion Detection System (IDS) for IoT networks utilizing a hybrid deep learning approach combined with effective feature selection techniques. The proposed IDS integrates a Convolutional Neural Network (CNN) for feature extraction and a Recurrent Neural Network (RNN), specifically Long Short-Term Memory (LSTM), for temporal sequence analysis of network traffic. A feature selection method based on Information Gain and Variance Thresholding is employed to reduce dimensionality and improve the efficiency and accuracy of the deep learning model. The performance of the hybrid IDS is evaluated using benchmark IoT network datasets, demonstrating superior detection accuracy, lower false positive rates, and enhanced resilience against various attack vectors compared to existing state-of-the-art methods. The results highlight the potential of this approach to significantly improve the security posture of IoT environments.
The smart grid represents a significant advancement in energy infrastructure, promising enhanced efficiency, reliability, and sustainability. However, realizing its full potential requires sophisticated control strategies capable of adapting to the dynamic and unpredictable nature of energy demand and supply. This paper presents a novel adaptive hybrid approach that combines machine learning (ML) techniques with optimization algorithms to improve energy efficiency in smart grids. The ML component leverages historical data and real-time sensor information to predict energy demand and renewable energy generation, while the optimization component uses these predictions to optimally allocate resources, schedule energy storage, and manage demand response programs. The proposed approach is evaluated through simulations using realistic smart grid scenarios, demonstrating significant improvements in energy efficiency, reduced peak demand, and enhanced integration of renewable energy sources compared to traditional methods. The results highlight the potential of the hybrid approach to address the challenges of modern energy management and contribute to a more sustainable and resilient energy future.
Federated Learning, Intrusion Detection System (IDS), Explainable AI (XAI), Network Security, Hybrid IDS, Machine Learning, Anomaly Detection, Signature-Based Detection, Distributed Learning, Privacy Preservation
The escalating sophistication and volume of cyberattacks demand robust and adaptable intrusion detection systems (IDSs). Traditional centralized IDSs often struggle with scalability, data privacy concerns, and the ability to detect novel attacks. This paper proposes a novel hybrid IDS framework that leverages federated learning (FL) and explainable AI (XAI) to overcome these limitations. The framework combines the strengths of signature-based and anomaly-based detection methods within a federated learning environment, allowing for collaborative model training across multiple network edge devices without sharing sensitive raw data. Furthermore, XAI techniques are integrated to provide insights into the IDS's decision-making process, enhancing transparency and trust. The effectiveness of the proposed approach is evaluated using a benchmark network intrusion dataset, demonstrating significant improvements in detection accuracy, reduced false positive rates, and enhanced model explainability compared to traditional centralized and non-federated IDS deployments. The results highlight the potential of FL and XAI to revolutionize network security by enabling decentralized, privacy-preserving, and interpretable intrusion detection.
Cloud computing has emerged as a pivotal paradigm for delivering scalable and on-demand computing resources. Efficient resource allocation is paramount to maximizing the benefits of cloud infrastructure, including performance, cost-effectiveness, and energy efficiency. This paper presents a novel hybrid metaheuristic optimization approach that combines the strengths of the Genetic Algorithm (GA) and Particle Swarm Optimization (PSO) for enhanced resource allocation in cloud environments. The proposed algorithm, named GA-PSO, leverages the global exploration capabilities of GA and the local exploitation capabilities of PSO to achieve a superior balance between exploration and exploitation. The performance of the GA-PSO algorithm is evaluated through extensive simulations under various workload scenarios and compared against traditional GA and PSO algorithms. The results demonstrate that GA-PSO significantly improves resource utilization, reduces makespan, and minimizes energy consumption compared to its counterparts, highlighting its potential as a robust and efficient solution for resource allocation in cloud computing.