Abstract
This paper presents a rigorous analysis of circumstances under which wave dynamics can be fully reconstructed from sensor networks. We prove that, in many cases, wave dynamics can be fully reconstructed from a single sensor, but that such reconstructions can be sensitive to small perturbations in sensor placement. In general, multiple sensors are necessary, and we outline a procedure for reconstructing the dynamics from them. We also show that if a network of n sensors suffers from Gaussian noise, then the reconstruction error decreases as n -1/2. We illustrate and verify these results numerically by simulating wave dynamics and then reconstructing from sensor data the initial state of the system and time series of the system at randomly chosen locations.
Original language | English (US) |
---|---|
Pages (from-to) | 2297-2313 |
Number of pages | 17 |
Journal | SIAM Journal on Applied Dynamical Systems |
Volume | 21 |
Issue number | 4 |
DOIs | |
State | Published - Dec 2022 |
Externally published | Yes |
Keywords
- noisy data
- sensor networks
- solution reconstruction
- wave dynamics
- wave equation
ASJC Scopus subject areas
- Analysis
- Modeling and Simulation