Colloquium - WARP: On-the-fly Program Synthesis for Agile, Real-time, and Reliable Wireless Networks

Date: 
September 4, 2020 - 4:00pm to 5:00pm
Location: 
Zoom - See emails for details
Speaker: 
Octav Chipara
The University of Iowa

Over the last decade, we have seen wireless communication used successfully in process control industries to provide real-time and reliable communication between sensors, controllers, and actuators over real-time flows. The emergence of Industrial Internet-of-Things (IIoT) is bringing forth a new class of applications that, in addition to real-time and reliable communication, also requires higher data rates, lower latency, and support for variable workloads. Examples of such applications include smart factories that continuously monitor the quality of produced silicon chips, automated warehousesthat use robots to load and unload merchandise, or smart buildings that rely on increasingly complex sensors for heating and air conditioning.

To address the needs of IIoT applications, we developed WARP -- a novel approach that uses software synthesis techniques to build wireless software-defined networks that provide real-time and reliable performance in the presence of network dynamics. We overcame three key challenges to meet this goal:

  1. create a language to specify a "smart" data plane can adapt to short-term variations in link quality more effectively than traditional schedules;
  2. software synthesis techniques that are scalable and guarantee that the reliability and deadline constraints of flows are met; and
  3. a control plane design that can effectively handle workload and topology changes by performing on-the-fly synthesis on constrained embedded devices.

Extensive experiments using simulations and testbed deployments demonstrate that WARP provides significantly higher performance and agility than traditional scheduling approaches. 

Bio

Octav Chipara

Octav Chipara is an Associate Professor of Computer Science at the University of Iowa. His research focuses on systems and networks that must provide predictable performance. His work combines the design of communication protocols, middleware, and software tools with large-scale real-world deployments of working systems. He received his B.S. from Indiana University Bloomington and his Ph.D. from Washington University in St. Louis. He is a recipient of an NSF Career Award in 2018. He received the best paper awards at IEEE/ACM Middleware 2015 and IEEE International Symposium on Computer-Based Medical Systems (CBMS 2013). His interdisciplinary research aimed at developing novel methods to evaluate and configure hearing aids was recently recognized with the Editor's Award for the Best Published Article in Ear and Hearing during 2019.


More on Prof. Chipara's research in this video produced by CLAS: