Date of Award
Doctor of Philosophy (PhD)
With the emergence of industrial standards such as WirelessHART, process industries are adopting Wireless Sensor-Actuator Networks (WSANs) that enable sensors and actuators to communicate through low-power wireless mesh networks. Industrial monitoring and control applications require real-time communication among sensors, controllers and actuators within end-to-end deadlines. Deadline misses may lead to production inefficiency, equipment destruction to irreparable financial and environmental impacts. Moreover, due to the large geographic area and harsh conditions of many industrial plants, it is labor-intensive or dan- gerous to change batteries of field devices. It is therefore important to achieve long network lifetime with battery-powered devices.
This dissertation tackles these challenges and make a series of contributions. (1) We present a new end-to-end delay analysis for feedback control loops whose transmissions are scheduled based on the Earliest Deadline First policy. (2) We propose a new real-time routing algorithm that increases the real-time capacity of WSANs by exploiting the insights of the delay analysis. (3) We develop an energy-efficient routing algorithm to improve the network lifetime while maintaining path diversity for reliable communication. (4) Finally, we design a distributed game-theoretic algorithm to allocate sensing applications with near-optimal quality of sensing.
Kunal Agrawal, Roger D Chamberlain, Christopher D Gill, Humberto Gonzalez