Deep Learning to Predict the Feasibility of Priority-Based Ethernet Network Configurations

Machine learning has been recently applied in real-time systems to predict whether Ethernet network configurations are feasible in terms of meeting deadline constraints without executing conventional schedulability analysis. However, the existing prediction techniques require domain expertise to choose the relevant input features and do not perform consistently when topologies or traffic patterns differ significantly from the ones in the training data. To overcome these problems, we propose a Graph Neural Network (GNN) prediction model that synthesizes relevant features directly from the raw data. This deep learning model possesses the ability to exploit relations among flows, links, and queues in switched Ethernet networks and generalizes to unseen topologies and traffic patterns. We also explore the use of ensembles of GNNs and show that it enhances the robustness of the predictions. An evaluation on heterogeneous testing sets comprising realistic automotive networks shows that ensembles of 32 GNN models feature a prediction accuracy ranging from 79.3% to 90% for Ethernet networks using priorities as the Quality-of-Service mechanism. The use of ensemble models provides a speedup factor ranging from 77 to 1,715 compared to schedulability analysis, which allows a far more extensive design space exploration.

PENGEMBANGAN INFRASTRUKTUR JARINGAN BERBASIS NIRKABEL KE JARINGAN BERTEKNOLOGI GPEN (GIGABIT PASSIVE ETHERNET NETWORK) DI KAMPUS IAIN BENGKULU

The purpose of this research is to build an interconnection network using GPEN (Gigabit Passive Ethernet Network) technology so as to provide convenience for network admin users in managing, controlling, and maintaining each client, making it easy for network technicians when a breakdown occurs, and Providing easy client user authentication in the area (PUSKOM) IAIN Bengkulu. The research method used is research and development research, the research method used to aim at developing, producing products and waiting for the effectiveness of the research method. Developing a wireless-based network, namely a point-to-point radio to a network with GPEN technology (Gigabit Passive Ethernet Network) as a switch and router that supports gigabit ports and to control and monitor bandwidth on each client. Better network quality and efficiency so that activities such as browsing, uploading and downloading are faster.

A study of securing in-vehicle communication using IPSEC protocol

Current vehicles are increasingly dependent on Electronic Control Units (ECUs) that control virtually every system of the vehicle. To enable advanced features automotive embedded systems are opening to external world, which raises security concerns. At the same time these innovative systems require more complex software and higher bandwidth for information exchange. Thanks to its bandwidth, payload size, and openness, Ethernet is a candidate technology for future in-vehicle architectures. This paper deals with design of a novel approach to secure In-vehicle Systems by taking advantage of Ethernet/IP technology and proven security mechanisms from TCP/IP model. Main goal is to design an efficient solution that meets requirements for latency without requiring high amounts of processing power and provides secure exchange of control messages. The work is mainly focused on the widespread Controller Area Network (CAN). The presented solution is based on encapsulation of CAN frames into UDP datagrams with added authenticity, integrity, and (if required) confidentiality of communication using IPsec protocol in transport mode. This creates a “secure tunnel across backbone Ethernet network in a vehicle. Next part of the paper presents extensive tests in simulation that are based on our previous experiments on hardware, in order to evaluate the characteristics of the designed security extension. The results indicate that using IPsec is a viable solution for securing in-vehicle communications.

Post-quantum MACsec in Ethernet Networks

The demand on MACsec in Ethernet is increasing substantially since MACsec fits well for industrial applications which require strong security as well as efficiency. To provide a long-term security, the MACsec protocol should be resistant to future attacks including quantum attacks. In this paper, MACsec is analysed under a quantum attack scenario. To achieve 128-bit quantum security, AES (Advanced Encryption Standard) algorithms defined in MACsec should mandate to use 256-bit keys. On the other hand, classical public-key cryptosystems in MKA are not secure at all against quantum attacks so that they need to be replaced by post-quantum crypto schemes in a quantum world. We propose an authenticated post-quantum key establishment protocol which is suitable for long-term secure MACsec. The proposed protocol is used in the hybrid mode, an ephemeral key exchange, and an end-to-end encryption. We verified by experiments that the proposed protocol can be deployed in existing a MACsec-enabled Ethernet network.

Development of Interactive PLC Learning Media Based on Virtual Learning

This research aims to develop interactive PLC learning media based on virtual learning. The interactive PLC learning media uses a 3D simulator design to simulate the application of PLC in the industry, so it looks like real and makes students more interested in learning PLC This study uses experimental methods. The 3D simulation software is Factory I/O developed by Real Games. It enables us to build a conveyor virtual plant and to bring it into the classroom. The automatic controller used to control the virtual plant is the PLC Siemens S7-1200 with Tia Portal software. The PLC is connected to the computer server through an ethernet network via TCP/IP protocol. With the rapid development of current internet technology, PLC learning can be done in virtual learning. This PLC learning media can be run remotely through the internet with team viewer program so that PLC learning is not only done on campus but can be done anywhere. From the results of research that has been done, the user can create PLC programs remotely to control conveyor virtual plant on a computer server through an internet network without having to install the program on their computers. Keywords: PLC, interactive learning media, virtual learning, remote desktop