Dr Lanlan Su
Dr. Andrea Lecchini-Visintini
Wireless sensor network is a group of a spatially distributed and dedicated sensors communicating via a wireless medium and cooperatively monitoring the physical conditions of the environment such as temperature, sound, speed, pressure, humidity, etc. Beyond the sensing task, a modern sensor network is equipped with actuators, so that each individual sensor can activate the actuator devices to accomplish a specific task when a certain event or fault is detected. Sensor networks play significant roles in practical applications such as health care monitoring, environmental tracking, monitoring of traffic, threat detection, industrial monitoring and many more.
A conventional sensor network has a centralized architecture, and it typically requires a leader node to gather data from all sensors and make a decision on the actuation task based on the combined information. There exist significant challenges in using a centralized sensor network, as it can only work in a short communication range, yields long response delays, and consumes a lot of power. In contrast, a distributed sensor network requires each node (sensor) to communicate with its immediate neighbours and an individual node can decide on its own if its output has exceeded a specified threshold in an actuation task. Such decision-making capability brings advantages in terms of ease of re-configuration, high scalability, reduced communication costs, and ability to execute local actuation tasks in response to local phenomena.
Nevertheless, measurement noise and the involvement of a wireless communication network can complicate the filtering process embedded in the distributed sensor, and therefore degrade the performance and even destroy the reliability of the sensor network. This could lead to detrimental consequences in some scenarios, like the failing of fault detection for aircraft systems. Hence, it is of great importance to obtain a theory for analysing robustness of the sensor network against the uncertainty and disturbance caused by the network-induced factors like data propagation delay, quantization effects, limited channel capacity and environmental noise. This project aims at developing a theoretical framework that can be used to establish the reliability of the distributed sensor network in the presence of networked imperfections and uncertainties.
Y. Yan, L. Su, V. Gupta and P. Antsaklis, Analysis of Two-Dimensional Feedback Systems Over Networks Using Dissipativity, in IEEE Transactions on Automatic Control, vol. 65, no. 8, pp. 3241-3255, Aug. 2020, doi: 10.1109/TAC.2019.2945038.
Su, L., Li, M., Gupta, V., & Chesi, G. (2019). Distributed resource allocation over time-varying balanced digraphs with discrete-time communication. arXiv preprint arXiv:1907.13003.
Khong, S. Z., & Su, L. (2021). On the necessity and sufficiency of the Zames–Falb multipliers for bounded operators. Automatica, 131, 109787.
Rogers, E., Galkowski, K., Paszke, W., Moore, K. L., Bauer, P. H., Hladowski, L., & Dabkowski, P. (2015). Multidimensional control systems: case studies in design and evaluation. Multidimensional Systems and Signal Processing, 26(4), 895-939.