Extraction of Disturbance Signals Based on the Inertial Sensor with a Blind Source Separation Method
Abstract
The inertial sensor is a key component in the photoelectric tracking system mounted on the moving platform, and its disturbance information extraction is a prerequisite for decoupling the system disturbance. In this paper, to get the specific signal of fs Hz, firstly, source signal of 2xfs Hz from the inertial sensor is needed. Then, single channel signal with frequency 2xfs is constructed into 2 aliased signals with the frequency of fs by down-sampling. Finally, a BSS (Blind Source Separation, BSS) algorithm is used to extract the disturbance signal in the aliased signals. Simulation results of ship borne data show that the algorithm can accurately extract the high-frequency disturbance signal of about 33 Hz in the aliased signals. The experimental results of the platform prove that the spectrum of the disturbance signal (10 Hz) and the spectrum of the target signal (0.1 Hz) do not overlap each other, and the correlation coefficient between the perturbation signal and the perturbation source signal is as high as 0.9976, which means the extraction of disturbance signal in the inertial sensor signals is realized.
- Publication:
-
Journal of Physics Conference Series
- Pub Date:
- February 2019
- DOI:
- 10.1088/1742-6596/1169/1/012009
- Bibcode:
- 2019JPhCS1169a2009H