School of Electrical Engineering and Computer
Science
Sensor and Ad-hoc Network Systems
Sensor networks are resource-constrained in terms of size, battery, cost, and infrastructure, but still they are attractive for a wide variety of critical and emergency applications. Dr. Medidi’s research is focused on developing efficient protocols in sensor networks to address node localization, reliable communications, and energy-efficiency.
Cost-constrained sensor nodes are not typically equipped with positioning hardware such as GPS, so their position information or localization needs to be inferred. Existing solutions for localization, such as multi-dimensional scaling and mesh-relaxation techniques, are computationally expensive. Dr. Medidi has developed a sampling technique, using clustering, to identify a representative subset of nodes upon which the localization can be performed. The remaining nodes infer their positions based on the representative sample. This approach is more resilient in irregular-shaped sensor deployments and obstacles to wireless communications typical in sensor networking; furthermore, this energy-efficient approach scales to large and high-density sensor networks.
Traditional TCP (transport control protocol) designed for wired networking degrades significantly over wireless links and provides abysmal throughput. In over-provisioned sensor networks, an acceptable number of data packets must be assured to reach the base station; however all query and control packets from the base station to the sensor nodes must arrive reliably to ensure the application objectives. Dr. Medidi is developing a transport protocol for sensor networks given required bounds on reliable data delivery and complete dissemination of control messages.
Ad-hoc networks are preferred means of communication for information gathering and time-critical decision-making where infrastructure is not available, such as in emergency response. It is critical that these networks support secure communication while maintaining a high level of network performance. Dr. Medidi and her research group are currently synthesizing software-hardware monitoring techniques to enhance the security and performance of mobile ad-hoc networks.
Contact Information
Sirisha Medidi, Ph.D.
Assistant Professor
School of Electrical Engineering and Computer
Science
Washington State University
P.O. Box 642752
Pullman, WA 99164-2752
Telephone: 509-335-2343
E-mail: smedidi@eecs.wsu.edu
Safety and Security
|
|
.jpg)