- AI-Enabled Networking for Internet of Things
- Big Data Analytics for Smart Connected Communities
- Virtual Reality by Autonomous Aerial Systems
- Predictive Signal Acquisition and Processing
- Nano-Scaled Holographic Image-based Security Solutions
- Practical Wireless (Passive) Sensor Network: Design and Performance Analysis
- Computational Framework for Molecular Interaction Modeling
- Multi-User Information Theory
In WiNeSH research laboratory we are working on different projects centering around the following topics. Our research projects are supported by supported by NSF, NIH U54, US Airforce Research Laboratory, and Arizona Board of Regents (ABOR).
In Wireless Networking and Smart Health (WiNeSH), we are working on harnessing remote sensing, wireless networking, signal processing and machine learning algorithms to solve health-related problems. Some of the current projects are
- Predictive Communications for UAV Networks: In this project, we plan to build the next generation of communication and control protocols for flying Adhoc Networks (FANET). The idea is to move beyond the current practice of adaptive communications by predicting network topology dynamics and incorporating it into the different levels of communication protocols including routing, connectivity, topology control, task management, optimal sampling, and etc. This project is in the intersection of wireless networking, machine learning, graph theory, and distributed optimizations. For more information please check out these papers (paper1, paper2, paper3, paper4, paper5, paper6) . This is an ongoing project and I welcome your comments and feedback..
- Remote Heart Monitoring project: In this project, we are working on developing an individualized kit for remote monitoring of heart-related vital signals in order to trigger local and remote alarms when a life-threatening abnormality is predicted through analyzing the vital signals. The vital signals may include the heart ECG signal, environmental conditions (temperature and humidity), light, chemical pollutants, and activity level of a patient. RFID technology is used to identify patients. In training phase, the informative features of the vital signals will be extracted in order to capture the normal pattern of the signals at different states (sleep, wakefulness, physical exercise, ...). The kit will constantly monitor the signals and send it through a wireless channel to a central processing unit for further analysis. The system will ignite an alarm on the kit and will send an alarm message to the physician if abnormality is detected. Some related preliminary works are published in this paper and this paper. For more information check out this website.
- Optimal Packetization Policy for Secondary Users in Cognitive Radio Networks: In this project, we consider the entry point of Wireless Sensor Networks, where the sensor measurements are bundled into packets to be sent to a data fusion center. In current practical WSN platforms, the packet lengths are either constant or sole defined by input traffic statistics. In this project, we design an optimized packetization policies that constantly tracks the wireless channel conditions and adjusts the packetization criterion based on the current channel conditions. For more information check out this website.