Plenary Session

Prof. Marco Anisetti, Ph.D

Università degli Studi di Milano, Italy

Title: Trustworthiness in service-based systems

Abstract: The trend towards service miniaturization is revolutionizing the way in which modern IT systems are built. Business processes are increasingly built by composing existing services at high rates. Application context such as IoT is becoming preferred target for microservices supporting edge computation and scalable distributed deployment. In this context, the major drawback of current service-based application is the lack of trustworthiness. Service composition is mainly driven by the target functionalities, while less relevance is given to non-functional properties. The increasing pervasiveness and ubiquity of current systems point to the need of strengthen the role of assurance techniques in evaluating the properties of a composite service and, in turn, the level of trustworthiness guaranteed to its users. In addition ML based services are increasingly substituting the traditional algorithmic ones in the framework of composition, making the need of trustworthiness even more important. ML-based services are in fact less transparent and difficult to monitor than a traditional services, thus impeding the adoption of traditional assurance approaches.

The keynote will describe an approach for improving the service composition trustworthiness based on advanced assurance evaluation. The proposed assurance solution will support dynamic service composition based on blockchain and the assurance evaluation of ML services using a methodology based on Multi Armed Bandit.

Prof. Zonhgyu Lu, Ph.D

University of Huddersfield, UK

Tilte: Smart technology vs Intelligent learning

Abstract: The keynotes argue about the impact of smart technology on education environment through incorporating challenges in (1) pedagogy, e.g. curriculum improvements, engagements, confidence, and employability, (2) psychology, e.g. anti-guessing mechanisms, concentration functions, and (3) economic impact, e.g. about 300,000 users may save about £4 million in comparison with commercial systems in the markets. The intelligent system should be made affordable, easier to be adopted and applicable in a range of disciplines and geographic regions, such as Asian, Middle Easter and Europe, around world far more diverse than the traditional classroom scenario.

Prof. Gwanggil Jeon, Ph.D

Incheon National University, South Korea

Title: Color Filter Array Demosaicking Methods in Frequency and Spatial Domain

Abstract: This presentation addresses the problem of interpolating missing color components at the output of a Bayer color filter array (CFA), a process known as demosaicking. A luma–chroma demultiplexing algorithm is presented in detail, using a least-squares design methodology for the required bandpass filters. A systematic study of objective demosaicking performance and system complexity is carried out, and several system configurations are recommended. The method is compared with other benchmark algorithms in terms of CPSNR and S-CIELAB objective quality measures and demosaicking speed. It was found to provide excellent performance and the best quality–speed tradeoff among the methods studied.

Prof. Anh Ngoc Le, Ph.D

Electric Power University, Hanoi , Viet Nam

Title:  Hybrid Wireless Mesh Network (HWMN) – Routing and Channel Assignment Strategies

Abstract:

A Hybrid Wireless Mesh Network (HWMN) is the most generic type of Wireless Mesh Networks. In HWMN,  mesh routers can include gateway functionality (IGW) and provide connectivity to other networks, such as the Internet and other networks. While the mobile clients can act as a dynamic extension of the static infrastructure part of the network, by implementing routing and packet forwarding functionalities. The hybrid mesh architecture is the most applica­ble because mesh clients can not only directly communicate with other mesh clients, but also access the Internet service through mesh routers.

This keynote will introduce the Hybrid Wireless Mesh Network (HWMN), characteristics, and applications. Routing strategies and channel assignment problems will also be introduced. In particular, the emphasis is focused on the unique traffic characteristic of the HWMN network in the design of routing protocols. The keynote will also describe an approach for improving network performance using the gateway location and hello packet to limit the broadcast request routing. The results show the efficiency and low implementation complexity of the proposed solution.