首页机构设置政策法规科研项目基地及团队科技成果知识产权学术交流国际科技合作下载基地
学术交流
 内容列表 
 
内容列表
您的位置: 首页>学术交流>内容列表>正文

DCN下一代移动通信网国际研讨会—IoT专场

时间:[2017-10-10]  来源:

DCN下一代移动通信网国际研讨会—IoT专场

学 术 报 告

时间:2016年10月14日下午(星期五)14:00-17:30

地点:逸夫楼205会议室

主持人:通信与信息工程学院 曹傧博士

序号

报告题目

报告人

1

Smart and Connected Health Challenges, Opportunity and  Beyond

Honggang Wang,  Ph.D.

2

Towards Unified Decision-Making Framework for Mobile App Offloading  to Cloud Computing: From Theory to Products

Yonggang Wen, Ph.D.

3

3D  Surveillance and Reconnaissance by Wireless Sensor Networks: Challenges and  Findings

Hongyi Wu, Ph.D.

4

Multi-sensory Smart Device  User Behavior Classification and Learning

Shaoen Wu,  Ph.D.

报告内容简介

报告一:Connected health uses  Internet, sensing, communications and intelligent techniques in support of  healthcare applications. Wireless body area networks (WBANs) with various types  of biomedical sensors serve as one major infrastructure for connected health and  provide an opportunity to address issues in rapidly increasing connected health  applications. However, there are significant challenges in the area, such as  improving the performance of WBANs, analyzing large physiological data collected  from biomedical sensors, securing data transmission and protecting data privacy  especially in mobile and wireless environments. I will talk about the research  challenges, and opportunities in the field.

报告二:An  eminent tussle has emerged between the growing demand for rich applications on  mobile devices and the limited supply of  onboard resources (e.g., computing capacity, battery, etc). The latter is  fundamentally hindered by the physical size of mobile devices as dictated by its  mobility nature. In this research, we propose to leverage the seemingly  unlimited resources in cloud computing to extend the capability of mobile  devices via application offloading. In particular, resource-hungry mobile  applications are offloaded to a cloud platform in proximity.  The cloud platform provides a virtual  machine, as a digital clone of the physical device, to execute  dynamically-offloaded tasks.The mobile app offloading approach pivots on two  complementary technical contributions. First, previous research efforts have  been focusing on exploring alternative mechanisms to implement application  offloading, including Cloudlet, Cloud Clone and Weblet, etc. Second, it demands  a unified decision-making framework to understand a fundamental trade-off  between communication overhead and computation saving, in the presence of  inherent system uncertainty (e.g., varying channel conditions, dynamic computing  complexity, etc.).

In  this talk, we present a unified mathematical framework to determine whether it  is beneficial to offload a mobile app to the cloud in proximity. We model the  mobile workflow as a directed graph, whose vertices represent computing load of  a specific module and edges represent data exchange between modules. A policy  engine determines whether a vertex should be offloaded to the cloud for  execution, by trading the computation cost on the mobile device over the  communication overhead between the mobile device and the cloud. We formulate  this decision-making challenge as a constrained optimization problem, with an  objective to minimize a chosen cost metric (e.g., energy consumption) for either  the mobile user or the mobile service provider, under the constraint of quality  of service (QoS) requirements (e.g., delay deadline). Our analytical framework  builds on solving a series of progressively-challenging sub-problems with an  increasing complexity of the workflow graph, namely, graph of a single node,  linear topology and directed acyclic graph. For each of these sub-problems, we  have developed either closed-form solutions or efficient algorithms to provide  operational guideline for optimal mobile app offloading.

Our  theoretic framework is further verified by two real system developments. In our  group, we have developed a multi-screen cloud Social TV system (i.e., Yubigo),  allowing video-streaming session to seamlessly migrate across different screens  (e.g., TV, laptop and smartphone). It builds upon the clone-based model to  offload the session management onto a dedicated service container in the cloud.  Our research has also inspired an open-source cloud robotic framework (i.e.,  Rapyuta), which was developed by EPFL researcher by adopting our proposed  clone-based model. Both projects have been spun off as start-up with public  and/or private investment.

报告三:There have been increasing interests in deploying wireless sensors in  three-dimensional (3D) space for such applications as underwater reconnaissance  and atmospheric monitoring. An individual sensor is highly resource-constrained,  with extremely limited computing, storage, and communication capacities. To  network a large number of such sensor nodes is nontrivial. Particularly,  compared with its 2D counterpart, the scalability problem is greatly exacerbated  in a 3D sensor network due to dramatically increased sensor quantity in order to  cover a 3D space. This talk will introduce a collection of unique and  challenging problems in 3D wireless sensor networks and present the recent  findings in this emerging area.

报告四:In  this project, we propose a stochastic non-password authentication solution that  models users’ finger gestures and handholding patterns as users’ profiles using  machine learning algorithms. The system leverages the data collected by three  sensors on mobile devices: accelerometer, orientation and touch screen input  measurements, and then trains a Hidden Markov Model (HMM) user profile model.  This solution is a software-only approach that does not require any  authentication hardware such as fingerprint sensor. The solution employs machine  learning algorithms to passively re-use the data provided by existing sensors  equipped on devices. This project is implemented and evaluated on Android  devices for performance assessment. In experiments, we measure the accuracy of  classification and authentication. The result shows a high accuracy of 89%.

报告人简介

报告人一:Honggang Wang, Ph.D.Associate Professor, UMass Dartmouth,  USA

Dr. Honggang Wang received his Ph.D. in Computer Engineering at  University of Nebraska-Lincoln in 2009. He is an associate professor at  University of Massachusetts (UMass) Dartmouth. His research interests include  Wireless Health, Body Area Networks (BAN) and biomedical sensor design, BIG DATA  in mHealth, Cyber and Multimedia Security, Mobile Multimedia and Cloud, Wireless  Networks and Cyber-physical System.

报告人二:Yonggang Wen, Ph.D.  Associate Professor&Assistant Chair  (Innovation), NTU, Singapore

Dr. Yonggang Wen is an associate professor with School of Computer  Science and Engineering (SCSE) at Nanyang Technological University (NTU),  Singapore. He is also the Assistant Chair for Innovation at SCSE. He received  his PhD degree in Electrical Engineering and Computer Science (minor in Western  Literature) from Massachusetts Institute of Technology (MIT), Cambridge, USA, in  2008. Previously he has worked in Cisco to lead product development in content  delivery network, which had a revenue impact of 3 Billion US dollars  globally.

报告人三:Hongyi Wu, Ph.D.Professor, Batten Chair of Cybersecurity, Old  Dominion University, USA

Hongyi Wu is the Batten Chair in Cybersecurity and the Director of  the Center for Cybersecurity Education and Research at Old Dominion University  (ODU). He is also a Professor in Department of Electrical and Computer  Engineering. Before joining ODU, he was an Alfred and Helen Lamson Endowed  Professor at the Center for Advanced Computer Studies (CACS), University of  Louisiana at Lafayette (UL Lafayette). He received the  Ph.D. degree in computer science from the  State University of New York (SUNY) at Buffalo in 2000 and 2002, respectively.  His research focuses on networked cyber-physical systems for security, safety,  and emergency management applications, where the devices are often light-weight,  with extremely limited computing power, storage space, communication bandwidth,  and battery supply.

报告人四:Shaoen Wu, Ph.D.Assistant Professor, Ball State University,  USA

Dr. Shaoen Wu received the Ph.D. degree in computer science from  Auburn University, Auburn, AL, USA, in 2008. He is currently a tenure-track  Assistant Professor of computer science with Ball State University, Muncie, IN,  USA. He was an Assistant Professor with the School of Computing, University of  Southern Mississippi, Hattiesburg, MS, USA, a Research Scientist with ADTRAN  Inc., Huntsville, AL, USA, and a Senior Software Engineer with Bell  Laboratories, Qingdao, China. His current research interests include Intelligent  Internet of Things, wireless and mobile networking, cyber security,  cyber-physical systems, and cloud computing.


欢迎全校师生参加!

期刊社 科技处 国际处

20161010


上一条:建校70周年学术会议--“电磁理论与射频技术”云学术会议

下一条:有关抗癌蔬菜以及植物生长激素的分子研究

关闭窗口

重庆邮电大学科技处  版权所有