TUTORIAL 1 – Sunday May 26, 09:40 – 10:25
Yasir Naveed Malik, Assistant Professor, New York Institute of Technology, Vancouver Campus Canada
Yasir Malik received his Ph.D. from Department of Computer Science at Sherbrooke University, Canada. His research interest includes the discipline of software engineering and applied machine learning with applications in security and vulnerability assessment of distributed and mobile computing system. He is interested in developing systems, that are not only intelligent in their applications but also are capable to handle intelligent and adaptive adversaries. In his current research, he is investigating how machine learning techniques can help to optimize traditional software analysis approaches to detect and analyze security vulnerabilities and mine errors. The outcome of this research can help to develop predictive security controls at runtime to improve the overall design and security of the system.
Title: Applications of Artificial Intelligence in Network Security Management
Co-authored by Adetokunbo Makanju
With the advancement in computing and communication technologies, the IT infrastructure of today enjoys exponential growth, with more users and devices becoming increasingly connected and supporting mobility. The market leaders estimate around 25 billion more devices will be added to the global network by 2020, which will add new services and traffic to the network. The complexity of network architecture and the growing threat environment makes it challenging for service providers and network administrators to efficiently manage the network resources, whilst maintaining its security. Recently, industry and academic researchers have shown keen interest in studying applications of machine learning and artificial intelligence being applied to network management. Network security management tools provide administrators the global view of numerous endpoints, firewalls and security controls. This helps administrators to analyze network security requirements, maintain, configure, automate and deploy security solutions across the network. The data produced at each node provides Intel to various network transactions. This provides an opportunity to translate this information into intelligence that can be used to mine the security related information. In addition, this information can be used to build an intelligent system with the ability to detect vulnerabilities, run security analysis, automate firewall configurations, learning new symptoms and develop predictive security controls. In this tutorial, we will highlight the potential applications of AI in network security management, discuss current trends and future research prospects along with demonstration of tools and techniques.
Dr. Makanju received his Ph.D. from Dalhousie University, Halifax Nova Scotia in 2012. A recipient of the 2013 IEEE IM/NOMS Doctoral Dissertation Award he is currently an Assistant Professor with College of Electrical Engineering and Computer Science at the Vancouver Campus of the New York Institute of Technology (NYIT). Prior to working with NYIT held several positions including Post-Doctoral researcher with IBM/CIVDDD in Toronto and Research Engineer with KDDI Inc. in Japan. KDDI Inc. is the 2nd largest telecommunications network provider in Japan. His research interests include security for autonomic vehicles, application of evolutionary computation to security in Software Defined Networks and privacy in web browsing.
TUTORIAL 2 – Sunday May 26, 11:00 – 12:30
Sun Junshuai, China Mobile Research Institute
Sun Junshuai received the M.S. degree in CST from Xidian University, Xi’an, China, in 2005. From 2005 to 2013, he worked in CATT as a TD-SCDMA/TD-LTE L2 engineer, SE, team leader and the director of high layer technology department. Since 2013, he has worked as a researcher in CMRI. He has great R&D and industry experience in both telecommunication and radio resource management, based on which he puts forward MCD (Multiple centralized and distributed) design logic of the protocol stack. His current research interests focus on the architecture and functionalities of wireless protocol stack.
Title: 5G driven by AI
5G is deployed widely in the word. In this tutorial, 5G is introduced from the requirements, the architecture, the primary procedures and the trial progress. Based on the current status of 5G, AI-driven idea is introduced. The target of 5G is ‘Information a Finger Away, Everything in Touch ‘. 5G system should be a system with low cost, high spectrum efficiency, low power assumption. The architecture of RAN is composed of integrated and distributed architecture which is defined by 3GPP. The brief of 5G should be introduced in this tutorial. Based on the basic architecture of 5G, O-RAN, AI-driven RAN architecture facing the diverse traffic models and huge amount of UEs, is introduced with Open and Smart. O-RAN alliance is an international organization composing of EC, TSC and eight work groups. This tutorial will provide a clear picture of O-RAN from the view of the structure, the development history, and the technical groups. AS the evolution of RAN architecture, O-RAN gives a good way to take native AI in RAN.
TUTORIAL 3 – Sunday May 26, 14:00 – 15:30
Howy Shu, Huawei
Howy Shu is Senior Director, Head of Wireless Connectivity and Intelligent Communication Department at Huawei Device Company. He joined Huawei at 2013 and leaded a Connectivity Innovation Group, which is focusing on research of innovative wireless system, including intelligent Antenna technology, machine learning, Cellular-Connectivity coordination, Wireless position technology., Moreover, He currently leads a task force of researching future Wireless demand , upon Full Scenario Connectivity, 5G NR dual connection network, low latency gaming and Smart home.
Title: How to integrate all wireless technologies to stop SMART HOME going stupid
Nowadays, smart phone/watch/mobile devices have changed people’s life. More smart devices are coming into your home, living room, dining room, bed room and even bath room . However, do we feel our life getting “Smarter” ? Or we still suffer in complicated pairing, multi-steps connection and non-friendly settings…Smartphone is equipped with more and more wireless technologies, such as WiFi, BT, GPS, NFC; Cellular like LTE even 5GNR, mm Wave..?? More Antennas, more RF components are squeezed into a small form factor; These Wireless never talk, but interfere to each other (ex:Side lobe, Intermodulation, De-sense, …). From standard perspective, Cellular discussed in 3GPP, Bluetooth in SIG, NFC in NFC forum, WiFi in IEEE. These wireless technologies never communicate under the same language and never share information to each other. However, they are stay in one device, one body… Mobile game as an example, can we take advantage that Cellular and WiFi cooperate to achieve shorter latency performance ? Or by aggregate dual bandwidths to accomplish higher throughput supporting cloud game streaming ? Huawei is committed to re-define a high-quality, easy-to-use, secure and digital life for consumers, also name as “Full-Scene-Solution”. At present, Huawei actively invests and participates in innovation research of fusion wireless technology and its corresponding use cases, including mobile phones, wearable devices, audio accessories, smart home and other Home Electronic. We are devoted to work with all partners to ensure an future easy-smart home.
TUTORIAL 4 – Sunday May 26, 16:00 – 17:30
Bo Sun, ZTE
Bo Sun, Director of wireless technology standardization, ZTE Corporation. Mr. Bo Sun, Director of wireless technology standardization in ZTE Corporation. He has extensive experience in the industry’s efforts on wireless technology standardization and is well versed in standard organization and industry forums collaborations. He is also an expert of wireless technologies, focusing on short range wireless communications, edge computing and AI. Now he is the chair of IEEE 802.11 TGbd (NGV, Next Generation V2X), the Co-chair of IEEE 802.11 TGax PHY adhoc, the chair of national Wireless PAN WG in NITS.
Title: AI/ML in Future Wireless Communications
The AI/ML is a rapidly growing phenomenon penetrating almost every industry area and people’s daily life. The result of the penetration shows profound changes in productivity gain. AI/ML has demonstrated a significant impact on the communication industry in recent years. Preliminary AI/ML research and implementation such as a cross-layer wireless network Root Cause Analysis (RCA) use case showing the extract/apply effective rules increasing up to 80%, time deploying new rules reduced up to 95%, and the number of alarms reduced up to 70%. However, the current AI/ML implementation for wireless networking is only the beginning. Many areas are still in its infancy stage. The AI/ML could benefit network slicing orchestration based on big data in 4/5G virtualization. Using the control plane data with deep leaning/training, the trained model can be used to inferring a new data model for smart SDN controller and smart network O&M which meet the dynamic network changing conditions. The introduction of AI/ML in wireless communication PHY and MAC layers with deep leaning and inferring method show many promises. Further studies including solid theoretical backing and practical implementation architecture will bring in the production level maturity.