| Article ID | Journal | Published Year | Pages | File Type |
|---|---|---|---|---|
| 722110 | IFAC Proceedings Volumes | 2009 | 6 Pages |
A sensor location problem for monitoring network with stationary nodes used for estimating unknown parameters of distributed-parameter system is addressed. In particular, the situation is considered, when the actual spatial positions of sensor nodes at the experimentation stage may be uncertain to some extent and randomly fluctuate around some locations specified at the configuration stage. In the presented approach, some results from experimental design theory for dynamic systems with random regressors are extended for the purpose of configuring a sensor network. Then, a simple algorithm based on the notion of approximate near minimum from statistical learning theory is adapted to select the most informative sensor locations. The delineated approach is illustrated by numerical example on a sensor network design for a two-dimensional convective diffusion process.
