An Eigenfunction Approach to Parameter Estimation for 1D Diffusion Problems

Schäfer M, Rabenstein R, Ruderer A (2019)

Publication Language: English

Publication Type: Conference contribution, Conference Contribution

Publication year: 2019

Pages Range: 3786 - 3789

Event location: Neapel IT

DOI: 10.23919/ecc.2019.8795692

Abstract

The behavior of a distributed parameter system can be represented by an expansion into eigenfunctions. It allows to calculate the output signal in dependence on the parameters of the system. This contribution considers the inverse problem: Estimate the system parameters from noisy measurements of the output. To this end, an eigenfunction expansion serves to establish a state-space description of the distributed parameter system. The corresponding state-space matrices define an extended Kalman filter to perform estimation and tracking of parameter values. This approach is shown here for a diffusion problem in one spatial dimension. The value of the diffusion parameter is estimated from a simulated particle concentration under varying conditions. Problems of this kind arises for example in the emerging field of molecular communications.

Authors with CRIS profile

Maximilian Schäfer Lehrstuhl für Multimediakommunikation und Signalverarbeitung Rudolf Rabenstein Lehrstuhl für Multimediakommunikation und Signalverarbeitung Alexander Ruderer Lehrstuhl für Multimediakommunikation und Signalverarbeitung

How to cite

APA:

Schäfer, M., Rabenstein, R., & Ruderer, A. (2019). An Eigenfunction Approach to Parameter Estimation for 1D Diffusion Problems. In Proceedings of the European Control Conference (ECC) (pp. 3786 - 3789). Neapel, IT.

MLA:

Schäfer, Maximilian, Rudolf Rabenstein, and Alexander Ruderer. "An Eigenfunction Approach to Parameter Estimation for 1D Diffusion Problems." Proceedings of the European Control Conference (ECC), Neapel 2019. 3786 - 3789.

BibTeX: Download