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.

The growing number of cyber-crimes is affecting all industries worldwide, as there is no business or industry that has maximum protection in this domain. This problem can produce minimal damage if an organization has information security audits periodically. The process of an audit includes several steps, such as penetration testing, vulnerability scans, and network assessments. After the audit is conducted, a report that contains the vulnerabilities is generated to help the organization to understand the current situation from this perspective. Risk exposure should be as low as possible because in cases of an attack, the entire business is damaged. In this article, we present the process of an in-depth security audit on a distributed firewall, with different approaches for the best results. The research of our distributed firewall involves the detection and remediation of system vulnerabilities by various means. In our research, we aim to solve the weaknesses that have not been solved to date. The feedback of our study is revealed with the help of a risk report in the scope of providing a top-level view of the security of a distributed firewall. To provide a high security level for the distributed firewall, we will address the security flaws uncovered in firewalls as part of our research.

From the perspective of services to the population, with vast social implications, in which the security, confidentiality, and access to personal data represents a critical region, the medical services and information systems that are on the base of the strategic management in health care systems, are a theme of maximum interest and rather less approached. In particular, the prospect of at-home health care systems for screening and rehabilitation has raised enormous interest and is seen as a new method to approach the disease more efficaciously. With the development of Information Technology and its major breakthroughs in health care information systems, bringing the rehabilitation and monitoring of the elderly patients from in-hospital to more advanced at-home health care systems, information security measures has been raised. A survey of security measures and data communication security involved in health care systems in order to ensure information protection is presented in this paper. The specific security issues involved in the development of a health care system that manages data to support monitoring and rehabilitation of patients with Parkinson's disease is subject of this paper.

In computer networks the traffic control is an essential management issue and a permanent challenge for network engineers. The paper is evaluating the main QoS (quality of service) technologies used in networks management. The article is mainly focusing on classful queuing disciplines and HTB (hierarchy token bucket) Linux implementations. The shaping and prioritization mechanisms are explained, and are proposed three different practical solutions for implementing HTB, under a common Linux environment, for a defined QoS scenario.

Operating in a dynamic and competitive global market, railway companies have realized many years ago that better management of their logistical operations will enhance their strategic positions on the market. The financial component of daily operations is of utmost importance these days and many companies concluded that maximizing the profit relies on the integration of logistical activities with better income management. This paper presents a system consisting of three components: Ferodata BOX, Ferodata MOBILE, and Ferodata SYS, used to transmit to a web-server the status and operating information of an electric or diesel train. Train information includes data from locomotives, wagons, train driver, route, direction, fuel or electric consumption, speed, etc. All this information is processed in real-time and can be viewed in the web-server application. Additionally, the web-server application could manage and report details that are coming from the wagons, such as valuable information regarding the bogie wear, the identification of the wagons attached to a gasket, and identification the situations in which a wagon or group of wagons comes off the gasket configuration. All information about the status of trains is available on-line and at any moment the person responsible for management can use these data in their work.

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.