The growing number of cyber threats and the implementation of stringent privacy regulations have revealed significant shortcomings in traditional centralized machine learning models, especially in distributed systems like the Internet of Things (IoT). This study presents a Federated Learning (FL) framework designed for intrusion detection and malware classification. This framework enables decentralized model training while preserving data locality and minimizing communication overhead. The proposed architecture incorporates lightweight, privacy-preserving techniques, including gradient clipping, differential privacy, and encrypted model aggregation, to ensure secure and efficient collaboration across heterogeneous clients. Experimental results on two widely adopted cybersecurity benchmarks demonstrate that the framework achieves detection accuracies above 90%, maintains privacy loss below 5%, and improves communication efficiency by over 25%. These results confirm the viability of FL as a scalable, privacy-compliant approach for next-generation cybersecurity systems in highly distributed infrastructures.

This work explores the new approach of integrating Artificial Intelligence (AI) into an adaptive mesh security system and discusses the challenges and opportunities of this innovative method. In a digital world where cyberattacks and technological innovations are rapidly increasing, making AI a core tool in security architectures is not just an upgrade to defensive efficacy, but also a strategic necessity. By examining technical and operational barriers, as well as the requirement of advanced computational resources and specific expertise, the paper emphasizes the need to tailor security system design to each IT infrastructure. Access to extensive, diverse data to facilitate the training of AI algorithms for efficient threat recognition is needed, but also raises challenges in data collection, processing, and confidentiality. This paper also stresses the urgent need for cross sector collaboration and standardization in adaptive security as key to efficiently meet challenges and accelerate innovation adoption. In conclusion, though facing significant challenges, incorporating AI into adaptive mesh security systems blazes a path towards developing innovative solutions that enable organizations to navigate effectively in a constantly evolving cyber landscape. It highlights the need for research, development, and collaboration across disciplines to create a robust, adaptable, and transparent security ecosystem.

In this paper we are proposing an authentication method based on the energy usage patterns of connected devices. This approach marks a step up from traditional security protocols, using artificial intelligence and predictive analytics to identify the unique energy behaviors of each device. We implemented this system on Raspberry Pi 5, a small and affordable microcomputer, to show its efficiency and practicality in the real context of IoT applications. The solution we propose provides a new level of security by detecting subtle changes in energy usage that could mean suspicious activity or deviations from normal operation. The adaptability and scalability of this system makes it suitable for a wide range of IoT applications, from smart home devices to complex industrial systems. By closely studying energy data, our work highlights the importance and efficiency of integrating emerging technologies into IoT infrastructures for a safer and more resilient IoT environment.

The increasing frequency of cyber threats and the enforcement of strict privacy regulations have exposed critical limitations of traditional centralized machine learning models, especially in distributed environments such as the Internet of Things (IoT). This study presents a federated learning (FL) framework tailored for intrusion detection and malware classification that enables decentralized model training while preserving data locality and minimizing communication overhead. The proposed architecture incorporates lightweight privacy-preserving techniques-including gradient clipping, differential privacy, and encrypted model aggregation-to ensure secure and efficient collaboration across heterogeneous clients. Experimental results on benchmark datasets, such as CICIDS2017 [1] and TON_IoT [2], show that the framework achieves detection accuracies above 90%, while maintaining privacy loss below 5% and improving communication efficiency by more than 25%. These results confirm the viability of federated learning as a scalable and privacy-compliant approach for next-generation cybersecurity systems in highly distributed infrastructures.

Securing the IT infrastructure in today’s digital age against cyber-attacks has become vital, one of the most used robust solutions for cyber security being a DD-WRT solution. This article focuses on DD-WRT as a security solution for a TP-Link AC1750 router, model Archer C7. In this article, we will explore deploying DD-WRT on a network as a basic solution, followed by an analysis of port scanning and vulnerabilities. We will also examine the ways in which DD-WRT help increase network security by installing additional functionality such as VPN, firewall, and other applications. We will study how an implemented DD-WRT solution will prevent cyber-attacks and vulnerability scans, also referring to research studies conducted to date. This paper provides a starting point for understanding the role of DD-WRT in enhancing IT security against cyber threats.

Internet of Things (IoT) devices face exponential growth daily across diverse sectors. The need for robust security measures to safeguard sensitive data and ensure privacy is imminen ...