|
Research on QoE/QoS and Wireless Multimedia Communications: Multimedia is the
dominating traffic in internet and nowadays unreliable wireless access networks.
Mobile devices are typical energy-constrained and resource limited. Our research objective is
to design a new solution to accommodate quality multimedia service over
energy-constrained wireless networks, which are typical unreliable and
unsecure. New approaches including data-driven algorithm implementation, and
joint design of codec and protocols. |
Prototype of energy efficient mobile wireless multimedia systems. CEENBoTs carrying Gumstix computer boards taking camera pictures, invoking light-weight JPEG2000 codec and WiFi drivers in Linux OS, transmitting multimedia streams to client computers. Graduate and undergraduate students are working on theoretical model and experimental testing of wireless network performance. |
|
Research on Wireless Sensor Networks, Cyber Physical
Systems and Internet of Things: Low cost wireless sensor
devices provide ubiquitous data access bridging the internet and the physical
world. Our current research is focused on optimization and scheduling
algorithm design to bridge the gap between emerging applications and
energy-constrained internet of things. Applications include
real-time soil condition monitoring in precision agriculture, intelligent
transportation system design using heads-up display and smart-phone for
driver safety warning, as well as supply chain management for pharmacy
anti-counterfeiting. |
Prototype
of Heads-Up-Display for Driver Warning: A Smartphone APP Gets Real-time Road
Information, transmits information through Bluetooth to Arduino board, and
Arduino boards drive the LED to project information on car windshield
(letters and icons will be flipped) |
|
Research on Breast Cancer Imaging Classification: Early screening of breast
cancer is critical. Our research objective is to create a cost-effective
screening solution by leveraging modern microwave tomography imaging
techniques, as well as state of the art machine learning and stochastic
classification algorithms. New approaches include effective tumor
identification and classification study leveraging big data opportunities. |
New microwave tomography imaging provides tumor
information from different perspectives. Unlike traditional mammogram or MRI,
tomography image equipments are typically cheaper
while providing relatively good quality of patient data. New algorithms and
big data mining techniques are under investigation to provide useful
information for diagnosis. |
