Tutorial 1: Feature Extraction leveraging Programmable Data Planes for Traffic Analysis based on Machine Learning
Sergio Gutierrez (Universidad Autónoma Latinoamericana, Colombia), Juan Felipe Botero (University of Antioquia, Colombia), Adrian Lara (University of Costa Rica, Costa Rica)
Room 2 (Parati 2), Wednesday, November 30, 2022 – 9:30-13:00 (BRT)
Programmable Data Planes have created an expanded landscape for the complete realization of the Software Defined Networking paradigm. Programmability enables further customization of the logic of packet processing within forwarding devices. Thanks to the capabilities of Programmable Forwarding Devices (PFD), it is possible to implement personalized functions that make it possible to introduce additional intelligence for packet processing at the data plane while preserving the benefits of centralized logical view of the network state at the control plane. Among the functionalities that can leverage the capabilities of programmable devices, literature reports the incorporation of Machine Learning (ML) based algorithms for traffic analysis. In this tutorial, we explore how to take advantage of the functionalities of PFDs for one of the crucial operations of machine learning algorithms which is Feature Extraction. Given the visibility that PFDs have of the traffic while considering their computational limitations, it is possible to use these devices, located at the data plane, to extract features which can be used either in very simple algorithms embedded into the PFD or passed up as input to complex algorithms executed as applications in the control plane.
Tutorial 2: Malware Analysis and Detection
Ashu Sharma (WatchGuard, India), Hemant Rathore (BITS Pilani, K K Birla Goa Campus, India)
Room 3 (Parati 3), Wednesday, November 30, 2022 – 9:30-13:00 (BRT)
Often computer/mobile users call everything that disturbs/corrupts their system a VIRUS without being aware of what it means or accomplishes. This tutorial systematically introduces the different malware varieties, their distinctive properties, different methods of analyzing the malware, and their detection techniques.
Tutorial 3: Evolution of NOMA Toward Next Generation Multiple Access
Zhiguo Ding (University of Manchester, United Kingdom), Yuanwei Liu (Queen Mary University of London, United Kingdom)
Room 2 (Parati 2), Wednesday, November 30, 2022 – 16:30-19:45 (BRT)
User data traffic, especially a large amount of video traffic and small-size internet-of-things (IoT) packets, has dramatically increased in recent years with the emergence of smart devices, smart sensors and various new applications such as virtual reality and autonomous driving. It is hence crucial to increase network capacity and user access to accommodate these bandwidth consuming applications and enhance the massive connectivity. As a prominent member of the next generation multiple access (NGMA) family, non-orthogonal multiple access (NOMA) has been recognized as a promising multiple access candidate for the sixth-generation (6G) networks. The main content of this tutorial is to discuss the so-called “One Basic Principle plus Four New” concept. Starting with the basic NOMA principle to explore the possible multiple access techniques in a non-orthogonal manner, the advantages and drawbacks of both the channel state information based successive interference cancellation (SIC) and quality-of-service based SIC are discussed. Then, the application of NOMA to meet the new 6G performance requirements, especially for massive connectivity, is explored. Furthermore, the integration of NOMA with new physical layer techniques is considered, followed by introducing new application scenarios for NOMA towards 6G. Finally, the application of machine learning in NOMA networks is investigated, ushering in the machine learning empowered NGMA era, for making multiple access in an intelligent manner for the next generation networks.
Tutorial 4: Machine Learning for CPS Security: Limitations and Novel Attack Discovery Techniques
Chuadhry Mujeeb Ahmed (University of Strathclyde, United Kingdom), Muhammad Azmi Umer (DHA Suffa University, Pakistan)
Room 3 (Parati 3), Wednesday, November 30, 2022 – 16:30-19:45 (BRT)
Machine learning has found applications in the security domain, especially for attack detection. At the same time, adversarial learning has received a lot of attention from the research community in the cybersecurity domain. Adversarial learning is used to test the robustness of machine learning algorithms under attack and create attacks that deceive the anomaly detection methods. However, little attention is paid to the exhaustive search of the attack space. This tutorial focuses on a particular class of Cyber Physical Systems (CPS), called the Industrial Control Systems (ICS). In this tutorial, first, we will summarize the challenges of deployment of machine learning based anomaly detection techniques in a real-world scenario from a corpus of studies, followed by the recommendations. To address the limitations of machine learning based techniques considered in the first part, next, we propose a technique to derive an exhaustive set of possible attack patterns to enhance the anomaly detection and security analysis of a CPS. Further, a live demo will follow a hands-on tutorial to generate attack patterns using a real-world water treatment plant data.