The Open House at CEWIT
May 7, 2014 | 1:30-4:30pm
The Open House at CEWIT is a collaborative effort to display the Center’s technologies and commercialization possibilities to participants from industry and start-up entrepreneurs. CEWIT will be opening its doors to attendees from different constituencies for an opportunity to tour the facility, its labs, visit incubator companies, and learn about Center’s research initiatives. Faculty, incubators, and supporting companies will be presenting talks, demos, and providing discussion on their research and technologies.
We welcome you to join us at the Center of Excellence in Wireless and Information Technology (CEWIT) on May 7, 2014. Directions to the Center here. For any questions or concerns, please contact the CEWIT Team at firstname.lastname@example.org. This event will be held in between in the XLDB Healthcare and CDDA Spring Workshops being hosted here at CEWIT.
Open House Program & Floor Map
Featured CEWIT LabsVisualization Lab, CEWIT 131
Professor Arie Kaufman, CEWIT/Department of Computer Science
Student: Koosha Mirhosseini
Computer Graphics, Visualization, Medical Imaging, Virtual Reality
Current Project: CAVE
Immersive Cabin (IC) is a 5-wall CAVE(CAVE Automatic Virtual Environment) in Stony Brook University, which consist of 5 back projected walls and a front projected floor. This facility delivers a 360 horizontal field of view and saturates most of vertical view of user. The IC uses two projectors per wall to deliver stereoscopic images that are viewed using active shutter glasses. By combining motion parallax from head tracking with depth perception from stereoscopic display, this facility can help users to explore complicated virtual environments and easily distinguish between objects in a complex scene. The immersive stereoscopic experience of IC makes it suitable for various applications such as medical visualization, virtual architecture tours and urban planning. Visualizing the E-coli bacteria and virtual colonoscopy are good examples of how the IC can be used in the context of medical and scientific visualization. During the last years a numbers of buildings (Simon Center, RSS) have been constructed around the campus. Prior to each construction, building architects virtually explored the CAD drawings in 1-to-1 scale and iterated on their designs. Last but not least a futuristic visualization of Manhattan urban area is also available to demo on IC.
Professor Allen Tannenbaum, Department of Computer Science
Student: LiangJia Zhu
The Biomedical Imaging Lab performs various medical imaging analyses; currently researching the use of MRI imaging to predict outcomes in ablation of the left atrial fibrillation
Consortium for Digital Arts, Culture and Technology (CDACT)
Professor Margaret Schedel, Department of Computer Music
Students: Ken Fehling, Ally Mihailovich, Max Carmack, Tim Vallier, Matthew Blessing, Kate Schwarting, Marley Rosner, Jay Loomis, Derek Kwan, Vinny Metas, Michael Liuzzi, Jim Palmeri, Josh Blessing
Current project: Drum Circle
Cuddle time is a place where students from different majors come together in order to share ideas and work on projects collaboratively. Here, ideas are created and have the ability to culminate into a game such as Drum Circle. Inspired by a recent project, Chrome Racer, Drum Circle fosters a unified environment where players make music together with gamification elements.
Student: Shebuti Rayana
Our group is interested in studying, understanding, and modeling large-scale data. We focus on problems involving massive graphs (GATA); including link analysis, pattern discovery, and outlier/anomaly/fraud/event/outbreak detection. Our goal is to develop algorithms and scalable analytics for mining & learning from (graph) data.
Current Project: An Ensemble Approach for Event Detection and Characterization in Dynamic Networks.
Event detection in temporal graphs, such as cyber and communication network traffic data, has many key applications in systems monitoring and security. It also has useful applications in detecting real life events from news corpus, email communications, cell phone communications and so on. Dynamically monitoring the network over time to detect suspicious changes in its structure and behavior is a challenging task due to the evolving nature of the data. In this work, we develop an ensemble approach for real-time event detection and characterization for dynamic graphs. We propose to use five detection techniques that can rank the time points by their anomalousness, the results of which we combine to come up with a final ranking. What is more, we characterize the events; by identifying the main network entities, i.e. specific nodes and edges, that are responsible for the detected changes the most. Our ensemble also employs a rank merging strategy to rank these entities by the magnitude of their suspiciousness. There are two key features of our approach: (1) it is designed to operate in a real-time fashion, where the decisions are made as new data arrives, and (2) the ensemble approach yields a better ranking, thanks to its voting mechanism. We use both simulated data (network flow) and real data (New York Times news corpus and email communication of Enron Inc.) for our experiments. At first experiments are performed on cyber network flow data, carefully simulated at a large corporation with ground truth events. Our proposed method has successfully identified the time points in which the network went through suspicious state changes, as well as the key network agents that instantiated these changes which are verified quantitatively. Experiments on New York Times news corpus in the setting of temporal graphs with co-occurrence of named entities in the same article and also on email communication network of Enron Inc. successfully identified some important real life events and the entities associated with those events.
Professor Steven Skeina, Department of Computer Science
Students: Bryan Perozzi, Yanqing Chen, Rami al-Rfou
Our research covers a range of topics in natural language processing. A current focus is using Deep Learning techniques to build concise representations of the meanings of words in all significant languages, and use these powerful features to recognize entities and measure sentiment and other properties of texts. Another focus involves analyzing Wikipedia to identify the fame and significance of historical figures as reported in our book Who's Bigger?
Current Project: Polyglot: NLP for all the World's Languages
Big Data makes for big and interesting problems! Our lab focuses on analyzing large-scale text streams such as news, blogs, and social media to identify cultural trends around the world's people, places, and things.
Professor Samir Das, CEWIT/Department of Computer Science
Students: Zafar Qazi, Ayon Chakraborty, Fatima Zarinni, Jihoon Ryoo
Smart Clinical Environment Lab, CEWIT 260
Professor Rong Zhao, CEWIT/Department of Computer Science
Staff: Tianyun Ling, Matthew Cordaro
Develop data modeling, machine learning and visual analytics tools and clinical decision support applications
Current Project: Computer Augmented Rehabilitation
The goal of our project is to develop innovative computer games and virtual reality environments and integrate them into physical and cognitive rehabilitation processes to deliver high quality care to the aging population and patients recovering from injuries, such as soldiers returning from the two wars. We focus on areas with the greatest needs, including balance training, gait training, strength training, and portable/wireless telerehabilitation solutions.
Professor Jennifer Wong, Department of Computer Science
Matthew Cordaro, CEWIT Programmer/Analyst
Professor Petar M. Djuric, Department of Electrical Engineering
Students: Li Geng and Zhe Shen
The Wireless Sensing and Auto ID Laboratory supports research efforts in the areas of signal processing, communications, and networking and all related to wireless sensor networks and auto identification of objects. Current and recent research projects involve indoor localization and tracking using the Radio Frequency Identification technology.
Current Project: RFID Sense-a-Tags for the Internet of Things
We aim at addressing the question if it is feasible to design backscattering devices that can communicate independently. We investigate the design of a novel component for the Internet of Things that we refer to as sense-a-tag and that is passive or semipassive. It has the following functionalities: (a) initiation and generation of query signals for the tags in its proximity, (b) passive detection and decoding of backscatter signals from RFID tags in its vicinity, (c) processing of information that it acquires during its operation, and (d) communication of the collected information by backscattering. The research also involves development of protocols for the sense-a-tags.
Professor Erez Zadok, Department of Computer Science
Student: Ming Chen
Researchers and students in the MSL group perform research in operating systems with focus on file systems, storage, security, and networking. Emphasis is given to new methods, interfaces, and APIs that increase system security, usability, and performance significantly, improve the portability of operating system code, speed the productivity of development of new code, and more.
Current Project: How Fast is NFSv4.1 -- A Benchmark Study of the Linux NFSv4.1
We present a benchmarking study of Linux’s implementation of NFSv4.1, the latest version of the NFS protocol. The NFSv4.1 servers in the 2.6.32 and 3.12.0 kernels performed well in most of our experiments. However, we also observed that Linux’s memory management can waste up to 80% of NFS I/O throughput. NFS performance also suffers from unfair networking behavior that causes the throughputs of identical NFS clients to differ by a factor up to 19×. We show that NFS delegations can boost performance by saving up to 90% of network traffic, but a delegation conflict can cause a delay of at least 100ms—more than 500× the RTT of our 1GbE network. We also found that NFS can exhibit counterintuitive performance behavior due to its intricate interactions with networking and with journaling in local file systems.
Web-scale Knowledge Representation Lab, CEWIT 331
Professors Michael Kifer and Paul Fodor, Department of Computer Science
Students:Reza Baseda, Mohammad Amin, Tiantian Gao, Vikas Ashok
Current Projects: Flora-2, Planning for Security Policies, Ontology and Reasoning System for Access Policies to Software Services, Natural Language Understanding with Logic Programming
The Reality Deck, CEWIT 335
Chief Scientist Dr. Ari Kaufman, CEWIT/Distinguished Professor & Chair, Department of Computer Science
Break the barrier between illusion and reality with CEWIT's Reality Deck, which will be the world's largest visualization facility with more than 1.5 billion pixels worth of resolution.
The Reality Deck is a unique visualization facility located at the Center of Excellence in Wireless and Information Technology (CEWIT). Supported by the National Science Foundation (NSF) and Stony Brook University, the Reality Deck is the world’s first immersive gigapixel resolution display, offering more than 1.5 billion pixels. The Deck inhabits a 40' x 30' x 11' room and contains 308 LCD display screens driven by an 85-node graphics computing cluster that rivals the performance of modern supercomputers, approaching the visual acuity of the human eye. The product of years of research and engineering, the Reality Deck facility’s goal is to break barriers in data visualization and help scientists deal with the challenges presented by the massive datasets of today and tomorrow.
Professor Ellen Li, Stony Brook Medicine; and Wei Zhu, Department of Applied Mathematics and Statistics
Student: Yuanhao Zhang
Current Project: Comparing the RNA-seq and Microarray Platforms with Generalized Linear EIV Model
The discovery of gene biomarkers and related pathways is critical in the studies of disease and therapeutic treatment. The rapid evolution of modern biotechnology has generated several waves of measurement platforms – with some phasing out, while others co-sharing the market. Two front runners are the more traditional gene microarray technology and the newly arrived RNA-Sequencing (RNA-Seq) platform. An intensive literature review revealed that the prevalent statistical method for the comparison of the Microarray and the RNA-Seq platforms is the Pearson product moment correlation coefficient. However, the Pearson correlation is unable to provide a calibration formula to convert the expression levels between the platforms. It also fails to account for the fixed and proportional biases between the two platforms. To fill this void, we have developed a method based on the generalized linear errors-in-variable (EIV) model that is able to provide both a calibration equation as well as a statistical measure of the two biases.
Professor Song Wu, Department of Applied Mathematics and Statistics
Students: Fei He, Jianjin Xu
The focus of the lab is on developing and applying new statistical/computational methodologies for the analysis of genetic and genomic data.
Current Project: Scalable Parallel Processing Algorithms for Sequence Alignment and Assembly
High-throughput next generation sequencing (NGS) technology has quickly emerged as a powerful tool in many aspects of biomedical research. However, along with its rapid development, the data magnitude and analysis complexity for NGS far exceed the capacity and capability of traditional small-scale computing facilities, such as multithreading algorithms on standalone workstations. To address these issues, we try to develop solutions using the ever-increasing supply of processing power by massive parallel processing (MPP) systems. We have designed a scalable hierarchical multitasking algorithm for importing classical sequencing algorithms to modern parallel computers. More specifically, we have developed a novel parallel infrastructure, which includes a portable NGS-oriented messaging package that adapts well to heterogeneous communication systems, and a scheduling package that provides a dynamic balancing strategy for efficient task scheduling. Based on these, a unified software suite, entitled “PPSeq”, has been constructed to import serial bioinformatics algorithms.
Professor Wei Zhu, Department of Applied Mathematics and Statistics
Student: Jinmaio Fu
Current Projects: Platform Comparison via Errors in Variables Models with or without Replicates
Platform or instrument comparison is a critical task in many lines of research and industry. In biotechnology, for example, with the rapid development gene sequencing platforms – we now have the first and the next generation sequencing platforms, each with several brands manufactured by different companies, co-sharing the market – at the same time, the third generation sequencing method built upon the sequencing of a single molecule of DNA, has already emerged. An accompanying critical task for the statisticians is how to best compare and combine results from different platforms. Previously, we have demonstrated the advantages of the errors-in-variable (EIV) model as an optimal instrument comparison and calibration device. However, one limitation to the traditional EIV modeling approach is its reliance on the availability of repeated measures of the same sample. Such replicates can be expensive and at times unattainable. Two methods by Wald (1940) and Kukush (2005) respectively are applicable for estimating the EIV model in the absence of replicates -- both relying heavily on the pre-cluster of data points. In this work, we aim to combine and improve these two methods through a better clustering strategy.
Gauging Biomarker Consistency between Different Measurement Platforms
With the rapid development of biotechnology, an increasing number of platforms for measuring gene expression levels are co-sharing the market – with the older technology such as the gene microarray being phased out, while the newer ones based on the RNA sequencing (RNAseq) being brought in, generation after generation. A crucial question to the entire biomedical community is whether biomarkers detected through these different platforms are consistent or not? In this work, we present a theoretical framework on biomarker consistency study based on the errors in variable (EIV) model. First, we calibrate the measurements between two platforms through an EIV model featuring indices for the constant bias and the proportional bias. Subsequently we demonstrate how different biomarker detection algorithms including the fundamental fold change and Z –test, T-test, will be influenced by such biases. Finally, we discuss strategies to combine measurements from different platforms for better biomarker detection.
Mobile Computing Lab, CEWIT 388
Professor Yuanyuan Yang, CEWIT/Department of Electrical Engineering
Students: Zhiyang Guo, Dawei Gong, Zhenhua Li, Jun Duan
Conducts cutting-edge research related to cloud computing, data center networking, wireless/mobile networks, and optical networks.
Current Projects: High speed multicast scheduling for all-optical packet switches
In this research, We propose a novel optical buffer structure, a Low Latency Multicast Scheduling (LLMS) Algorithm that guarantees delay upper bound, and a pipeline and parallel architecture that enables line-rate scheduling, for all-optical packet switches.
Channel assignment in 802.11n WLAN
We study how to assign channels to APs in 802.11n WLANs, focusing on the new challenges from the 802.11n standard. We first introduce a throughput estimation model. Based on the model, we propose a distributed channel assignment algorithm, in which APs update their channels iteratively to maximize local network throughput. Simulation results show that the proposed algorithm can greatly improve network throughout.
Buncee.com is a fun and easy way for individuals to design and share engaging and interactive multi-media presentations, greetings, lesson plans, advertisements and more. Without ever having to leave the buncee platform, users can easily add any of our custom-made stickers, animations, and backgrounds, include personal photos, text, drawings, hyperlinks, and recorded audio, as well as online content such as YouTube videos, SoundCloud audio, Instagram, Google, or flickr images into a digital canvas called a ‘buncee’. Your personalized buncee creations can then be shared with your social and private networks with just a few clicks.
Technology Spotlight: Buncee Web, Buncee EDU, Buncee Mobile
Buncee leverages modern web and mobile technologies to provide an innovative digital canvas that enables users to create and present fresh, interactive multi-media content while on the go, in school, at the office or while relaxing at home on their favorite device. In addition to our extensive library of hand crafted DRM free media and artwork, users can also source content from their personal libraries or popular Internet sources such as Google, Instagram and SoundCloud. Complementing our web portal is a suite of HTML5 and iOS mobile products.
CA Technologies helps customers succeed in a future where every business— from apparel to energy— is being rewritten by software. With CA software at the center of their IT strategy, organizations can leverage the technology that changes the way we live— from the data center to the mobile device.
Technology Spotlight: CA Mobile Application Security Demo
Provides granular and centrally managed mobile security, protecting applications, data, networks and device capabilities in a single product.
Frank Chau & Associates, LLC, CEWIT 261
Intellectual property law firm; ranked in 2013 the No. 1 NEW YORK FIRM which secured the highest quality U.S. patents in the category of consumer electronics (by Intellectual Asset Management).
Intelibs develops and provides unique 3G, 4G and WiFi wireless coverage and capacity solutions with products and services for Distributed Antenna Systems (DAS) to meet U.S. carrier and enterprise wireless needs. Intelibs specialized to offer the Hybrid DAS solutions for large corporate, higher educational institution campuses where they need indoor and outdoor seamless wireless mobility. Hybrid DAS solution provides unified RF coverage with multi-technology, multi-carrier platform with scalable and flexible network architecture for variety of venues from single high rise to multi complex such as University campus, corporate building, hospitality and healthcare. Intelibs provides the turnkey solution for the wireless carriers and venue owners including the wireless equipment procurement, site survey, design, commissioning, optimization and remote monitoring and maintenance. With the unique carrier grade WiFi and 3G/4G cellular coverage solution, Intelibs provides the innovative business model enabling both the venue ownership and wireless service providers to work together to build the true mobility with the simple finance model.
Hoffmann & Baron, LLP, CEWIT 273
Anthony E. Bennett
Hoffmann & Baron, LLP is Long Island's premier Intellectual Property law firm, possessing expertise in all areas of technology. Since 1984, Hoffmann & Baron has provided the umbrella of Intellectual Property protection that stimulates innovation and economic growth. We provide personalized attention and customize our services to fit the requirements of each client. Together with inventors and entrepreneurs, Hoffmann & Baron transforms ideas into assets.
Lexmark/Perceptive Software, CEWIT 361
Gail Zwerman and Thomas Wilton
Perceptive Software, a division of Lexmark, is a global leader in enterprise image and content management with an advanced innovative technology strategy and platform shaped by proactive intelligence and architected with visionary precision.
Technology Spotlight: Perceptive Acuo Vendor Neutral Archive
The Perceptive Acuo Vendor Neutral Archive is today’s solution for simplifying end-to-end lifecycle management of medical images and non-DICOM clinical content. A true vendor neutral archive (VNA) solution, the platform was designed for healthcare organizations that want to regain control of their data and achieve enterprise-wide interoperability across multiple disparate PACS with only one archive to manage. The Perceptive and Acuo unified content platform combines VNA and ECM technologies to create a powerful, integrated framework to manage, store and access all forms of content including clinical images, digital photos, audio and video content and all other unstructured data.
Softheon, CEWIT 343
Softheon is the emerging leader in next generation healthcare business process optimization, enabling payer, provider, and government organizations to measurably reduce administrative cost, improve member and provider satisfaction, generate new revenue opportunities, and comply with regulatory compliance by adopting best business practices. Softheon's healthcare best business practice solutions drive bottom-line benefits for health care payers, empowering them to improve competitiveness and comply with regulations. Softheon's process analysis mastery is enhanced by industry expertise, which helps Softheon work with clients to quickly identify the opportunities for process improvement that will have the greatest impact on their success.
SVAM International, Inc., CEWIT 280
SVAM International Inc. is a Global Information Technology (IT) services provider that delivers value and competitive advantage to our customers with high quality, cost effective software and related services that improve their access to critical information, automate their business processes, and help their personnel collaborate.
Tellabs, CEWIT 380
Tellabs delivers technology that transforms the new Enterprise Network. Tellabs experts design, develop, deploy and support wireless and wired network solutions. Tellabs comprehensive passive optical networks portfolio empowers the building of high performance LAN infrastructure that are simple, secure, stable, smart, and scalable.
Technology Spotlight: Gigabit Passive Optical Network (GPON)
CIOs and IT professionals are adapting to evolving network challenges associated with the introduction of big data, big data analytics, virtual desktop, hosted-managed services, software defined networks, cloud-based computing, wireless (3G/4G, DAS, Wi-Fi, BYOD), internet-of-things, and smart building technologies. Yet, oddly the local area network (LAN) infrastructure consisting of copper cabling and racks of Ethernet switches has remained unchanged. The legacy copper-based LANs architecture was implemented decades ago to support peer-to-peer desktop computer traffic flows because 80% of the traffic stayed local. Today it is expected that 90% of the LAN traffic flows directly to the wide area network (WAN) because of the new technologies identified above. Tellabs® Optical LAN solution builds on the cloud architectures value proposition lays the foundation for software defined LANs, and creates synergies with wireless technologies equally. Passive Optical LANs are a simple, secure, stable, scalable, sustainable, smart alternatives to legacy copper-based LANs. Federal government, commercial enterprise, healthcare, hospitality and education markets can obtain immediate benefit from Optical LANs that save energy, space and money.
Unify—formerly known as Siemens Enterprise Communications—is a premier, global communications software and services firm. Our solutions unify multiple networks, devices and applications into one easy-to-use platform that allows teams to engage in rich and meaningful conversations. The result is a transformation of how the enterprise communicates and collaborates that amplifies collective effort, energizes the business, and enhances business performance. Born out of the engineering DNA of Siemens AG, Unify builds on this heritage of product reliability, innovation, open standards and security to provide integrated communications solutions for approximately 75% of the Global 500. Unify is a joint venture of The Gores Group and Siemens AG.
Technology Spotlight: OpenScape UC Suite, OpenScape Moblity Client
OpenScape, Unify’s UC platform, is a complete stack of UC applications that includes built-in voice and conferencing services, voicemail, messaging, mobility, contact center and presence, which can be deployed either on premise or as-a-service in enterprises of all sizes.
Louis Tortora & Kin-Fung Chan
Verizon Communications Inc., headquartered in New York, is a Dow 30 company employing a diverse workforce of more than 180,000 dedicated employees around the globe. Verizon is a global leader delivering innovative communications and technology solutions that improve the way our customers live, work and play. Every day, we connect people, companies and communities with our powerful network technology. Not many companies get the chance to change the industry and change the world through innovation. We do. Verizon operates America’s largest 4G LTE network and most reliable 3G network. We also provide converged communications, information and entertainment services over America’s most advanced fiber-optic network, and deliver integrated business solutions to customers in more than 150 countries.
Technology Spotlight: Machine to Machine
Verizon M2M Technology automates processes and streamlines workflow by enabling machines to communicate with each other. Now you can stay focused on managing growth and gain greater visibility into your business—so you know what's happening and where.
Technology Spotlight: Mobile Workforce Manager
Mobility is a force changing and evolving the way our customers do business. The number of mobile workers continues to grow, so too does the number of devices which they use. Reports show that over half of mobile workers use at least two devices for work every day, corresponding to increasing mobile worker productivity. In addition, with the consumerization of IT now a given, users are driving the evolution of an organization’s IT strategy. The challenge facing IT departments is how to strike the right balance between mobile freedom demanded by users and the need for IT security and compliance.
Newest Release of the short summaries of Current, Funded Research Projects Now Underway.
The key to the CEWIT’s research achievements is developing successful relationships with affiliated institutions and centers. Sharing resources, research interests, and ideas contributes to exciting technological advances. Some of these affiliations result from successful grants and some are a natural extension of our Department.