Keynote Speakers:


黃有榕.jpg


Pro.Yu-Jung Huang

I-Shou University,Taiwan


Title:Detection of the mixed-type 3D defect with machine learning techniques


Abstract:

As technology node moves to 10 nanometers and beyond, defects become more detrimental to device performance and harder to detect. The TSV-based interconnect of 3D-SICs constitute a new multilayer structure, The new 3D processing steps lead to new intra-die defects and faults that are not covered by traditional 2D fault models. The defect inspection for the semiconductor industry's future technology has become a significant challenge. The International Roadmap for Devices and Systems (IRDS) shows it is important to explore new alternative technologies that can meet inspection requirements to discriminate defects of interest. Several mixed-type defects occur due to 3D integration process; therefore, it is important to identify these defect patterns in order to know the root causes of failure and to take actions for quality management and yield enhancement. In order to electrically detect these failures, it is necessary to study and analyze the electrical characteristics of defects in advance. In this talk, thenondestructive flaw detection in 3D die-stacked structure by implementing machinelearning methods in physical data will be presented.



姚修慎.jpg

A.Pro.Yao, Hsiu-Hsen

Yuan-Ze University,Taiwan


Title:A Case Study of Data Science --- Driving Data Analysis For Advanced Transportation Service 


Abstract:

Collecting a set of driving data via vehicles' sensors including GPS, OBDII, and telematics,

one may generate one driving feature database for a lot of transportation applications, consisting of driving behavior recognition, driving safety management, driving fuel consumption prediction, driver tutorial materials generation, advanced route navigation development, and route characteristics analysis. 

In this talk, the first topics we discuss is, how to collect driving data for generation of driving behavior indexes, driving safety indexes, and derived fuel consumption data. After that, using machine learning, analyzing the relationship between driving behavior and fuel consumption. 

Historical driving data could provide a better route navigation suggestion.  One compound route navigation method is developed based on driving time factor or on fuel consumption saving factor.  One driver may find a shortest path, some fast enough routes, a fuel consumption saving route, or a "comfortable" route.  Collected probe driving data were analyzed to generate route characteristics data warehouse for transportation management.

Finally, Some taxi driving data and a set of bus driving and passenger boarding data are collected into one travel data warehouse. Based on some data transformation approach, several innovative transportation services can now be provided.


Giampaolo-Buticchi.jpg

A.Pro.Giampaolo Buticchi

The University of Nottingham Ningbo, China


Title:Power Electronics and Electric Drives for the pursuit of an improved and sustainable transportation


Abstract:

The progressive pursuit for lower pollutant emission and for a more convenient transport system has been pushing the electrification to all the transportation sectors. This lecture overviews the advancement in the power electronics systems with particular attention to the aircraft transportation. Topologies, controls and electric machine drives will be discussed, highlighting the actual challenges and the future development.



黃依賢.jpg

Pro.I-Shyan Hwang

Yuan-Ze University,Taiwan


Title:Enabling SDN/NFV for 5G RAN-Optical Networks


Abstract:

5G is expected to radically transform the communications industry and reshape peoples’ daily lives and business operations. 5G will create new opportunities to simulate social and economic development of new innovative services with the SDN/NFV. The radio access network (RAN) has been in use since the beginning of cellular technology and has evolved through the generations of mobile communications (1G through 5G). Cloud-RAN (C-RAN), BBU-Base band units from Backhaul and RRH-Remote radio head are Fronthaul networks and connected for the Common public radio interface (CPRI). Therefore, the communication through optical networks between the Fronthaul and Backhaul is merged as an X-haul or Cross-haul. In this talk, we propose the hybrid Mobile-Xhaul-PON with software-defined networking (SDN) to provide intelligent networking, allowing a highly scalable, distributed, stateless forwarding plane to support functions like mobility management, policy, subscription control and maintain end-to-end path information for each service; and network functions virtualization (NFV) to virtualize entire classes of network node functions into building blocks that may connect, or chain together, to create communication services. Furthermore, new trends and challenging issues on future C-PON-RAN for different 5G RAN-Optical networks will be elaborated. 





Supported by

Media Support

Indexing Information

Official Wechat