Meso-Scale GENI WiMAX Project

Faculty: Jie Wu (PI), Eugene Kwatny (Co-PI), Shan Lin (Co-PI), Chiu C. Tan (Co-PI)

The popularity of wireless networks has led to significant investments in next generation wireless technologies such as WiMAX and LTE. These new technologies promise, among other things, better coverage and higher bandwidth rates, than existing networks. The objective of this project is to build an open, large-scale, outdoor wireless testbed to advance the state of next generation wireless network research. The testbed will be a joint participation of Temple University and Drexel University.

Body Sensor Networks and Their Applications in Maternal Fetal Monitoring

Faculty: Jie Wu (PI), Dimitrios Mastrogiannis (Co-PI)(Medical school), Li Bai (Co-PI)(ECE), Chiu C. Tan (Co-PI)

Assessment of fetal health during pregnancy constitutes a very important task of modern obstetrics. It is applied in high risk patients in the third trimester and in almost all patients during labor and delivery. Currently, the monitoring devices needed for fetal heart rate (FHR) and uterine contractions are hardwired to a large monitor (about 15 lbs), and require the patient to remain relatively immobile in order for the monitor to function optimally and continuously. The project seeks to design a body sensor network (BSN), a network consisting of one or more on-body sensing units coupled with a smart local processing unit, to allow normal mobility during the monitored period.

Mobile Content Sharing Networks: Theory to Implementation

Faculty: Jie Wu (PI), Xiaojiang Du (Co-PI)

The last few years have witnessed an explosive growth in the popularity and capabilities of mobile handheld devices such as smartphones, tablets, and laptops. These portable devices’ rapidly expanding computational power has enabled their users to access, process, and share content anytime, anywhere. Mobile content sharing has several salient features such as mobility of numerous users, rich content, and individual users’ ever-growing need for communication and computation capacity. The goal of this project is to investigate how to exploit the key attributes of mobile content sharing environments to dramatically increase network performance.

Hybrid Wireless Network Infrastructure for Integrated Research and Education

Faculty: Xiaojiang Du (PI)

This project seeks to build a Hybrid Wireless Network (HWN) consisting of two WiMAX base stations, forty-eight WiMAX/Wi-Fi mobile stations, one computing server, one storage server, and a Gigabit Ethernet switch. The proposed HWN infrastructure will support research and education in broadband wireless networking and communications, and enable high-quality and high-accuracy performance evaluations of protocols and schemes designed for HWNs.

Towards Robust and Self-Healing Heterogeneous Wireless Sensor Networks

Faculty: Xiaojiang Du (PI)

Research has shown that Heterogeneous Sensor Networks (HSN) can significantly improve performance of sensor networks. The objective of this project is to investigate innovative network architectures of HSN, design energy-efficient and self-healing routing protocols, and develop effective security schemes for HSN. Specifically, we will investigate efficient and robust HSN architectures, design self-healing and energy-efficient routing protocols for HSN, and propose effective security schemes.

A New Algorithmic and Graph Model for Networking in Challenged Environments

Faculty: Jie Wu (PI)

Networking in Challenged Environments (NICE) is designed to meet the special networking needs of the 21st century. These networks operate under special environments which pose unique challenges to the network design. One particular challenge is modeling and analyzing mobility. This project presents a generalized graph model that can capture mobility in NICE. This model is called a weighted evolving graph which captures time-space dynamics while remaining simple enough to maintain most of the elegant structure of the traditional graph model.

Energy-Efficient Design in Wireless Networks Using Cooperative Communication

Faculty: Jie Wu (PI)

The rapidly increasing capabilities and declining costs of computing and communication devices have made it possible to use wireless networks in a wide range of applications that can improve quality of life, and even save lives. One of the key challenges in the deployment of wireless networks is how to prolong the lifetime of the networks. The project studies energy management techniques for wireless sensor networks. The key idea is that we take advantage of the physical layer design that facilitates the combining of partial information.

Dynamic Carrier-Assisted Routing in Mobile Networks

Faculty: Jie Wu (PI)

The traditional connection-based approach to mobile networks views node mobility as undesirable. In this project, we consider a mobility-assisted model considers mobility as a desirable feature, where routing is based on the store-carry-forward paradigm with random or controlled movement of mobile nodes. This project will be useful in various applications of mobile networks, including MANETs, WSNs, and DTNs and the proposed study will contribute to making these networks more practical.

A Hybrid High-Performance GPU/CPU System

Faculty: Jie Wu (PI), Saroj K. Biswas (Co-PI), Michael L. Klein (Co-PI), Igor Rivin (Co-PI), Yuan Shi (Co-PI)

This project is to design and operate a hybrid high-performance GPU (graphics processing unit)/CPU system that will complement existing and future federal and state investments at Temple University and will help drive related research and educational activities. As GPUs are about to become an integral part of mainstream computing systems, the hybrid GPU/CPU system enables support for three groups of applications: traditional CPU-based, GPU-based, and hybrid GPU/CPU-based. The proposed hybrid system enables broader heterogeneous computing by deploying multiple types of computing nodes and allowing each to perform the tasks to which it is best suited.

Mobile Multicore Computing

Faculty: Jie Wu (PI)

Multicore technology is a breakthrough technology developed in recent years, which extends the lifetime of Moore’s Law by changing its applicability from uniprocessors to multicore processors. Multicore technology is entering the mobile phone domain, and the key challenge in mobile multicore phones is making a good tradeoff between performance and power. This project proposes software-oriented approaches including power-aware parallelization of mobile applications and power-aware task scheduling to meet this challenge.

An Architecture for Joint Integration of Inter and Intrasession Network Coding in Lossy Multihop Wireless Networks

Faculty: Abdallah Khreishah (PI), Jie Wu (Co-PI)

Maximizing the throughput while achieving fairness among the flows is one of the fundamental research problems in multihop wireless networks. Network coding has emerged as a promising approach to enhance the performance of wireless networks. The goal of this project is to provide a framework to study and deploy intrasession network coding (IANC), where only packets of the same flow or session are coded together, and intersession network coding (IRNC), which exploits the broadcast advantage of wireless links by mixing different flows at intermediate nodes to resolve bottlenecks, jointly under different wireless network settings.

Auction-based Cloud Computing

Faculty: Justin Y. Shi, Abdallah Khreishah, Slobodan Vucetic

Auction-based cloud computing can reliably reveal the true cost of computation. It also promises the ultimate resource efficiency. This project investigates new computational models and methods that can deliver performance and reliability at the same time when acquiring computing and communication components. These sustainable applications should also survive harsh processing environments.

Google Docs for Finance Curriculum (uFin Model)

Faculty: : Justin Y. Shi, Michael Bolton (Finance)

Google Docs provide seamless authentication that traditional spreadsheets do not have. The availability of Google Finance is also a valuable teaching tool. This project investigates the effective use of Google Docs for teaching financial trading courses.

Integrated Data Warehouse with Entity Matching

Faculty: : Justin Y. Shi, Zoran Obradovic

Integrating heterogeneous data sources concerning the same population requires entity matching using reliable unique identifiers. In practice, this requirement is often broken. This project investigates entity matching methods that can tolerate typos, transposes, aliases and frequently changed mutable properties. We also investigate database clustering technologies for delivering extremely high performance and high availability at the same time when adding components.

Multiresolution Video Streaming with Network Coding

Faculty: Abdallah Khreishah, Jie Wu

Video streaming over the communication networks requires unique treatments due to its the special properties. First, the video can be divided into multiple layers that represent different resolutions. Second, the recovery of one layer at the destination is meaningless without recovering all of the lower layers by that destination. Network coding has been shown to simplify the operations and enhance the throughput of multicasting. However, network coding has been studied under homogeneous environments where all of the destinations have the same bandwidth from the source. The objective of this project is to study the integration of network coding with a heterogeneous environment where the destinations have different bandwidth requirements. This creates different challenges on how to create the different video layers and how to code them at the intermediate nodes such that every destination can recover the appropriate amount of video layers.

Project 910: Smart Phone and Social Media for Public Safety

Faculty: Justin Y. Shi

Staffing the 911 call portals is a non-trivial financial and technical challenge for law enforcement offices. Although we have many security cameras mounted, we simply do not have enough eyes to watch them 24/7. This project seeks to dramatically enhance per person surveillance ability by leveraging smart phones, web-based security webcams and social media.

Temple CAVE: Virtual Environments with Motion Analysis

Faculty: Emily Keshner (Public Health), Justin Y. Shi, Haibin Ling

Virtual Reality (VR) technologies are typically used for gaming, simulations and training. We investigate human reactions to 3D visual stimuli. There are many applications. These include physical therapy, medical education, athlete training, scientific visualization and engineering designs. With the advent of low cost 3D TVs and motion sensors, low cost personal CAVEs are also possible. When powered by health social network, these devices may deliver services that are not possible before.

Temple Elastic HPC Cloud

Faculty: Justin Y. Shi, Jie Wu, Abdallah Khreishah

Not all computing intensive applications can use Linux and batch processing for results. Many require human interactions and different operating systems and different “application stacks”. The TCloud project investigates sustainable infrastructure for scientific research offering heterogeneous high performance processors and networks and multiple operating systems.