Adaptive filter based on NARX model for atmospheric noise removal on exo-planet observations

In Digital Signal Processing, adaptive filtering is capable of dealing with random signal noise or time-varying signal. This paper presents an approach to create a NARX-based adaptive noise filter to remove atmospheric noise from astronomical signals. We use data obtained from the transit spectroscopy exo-planet observation, data was prepare using k-means clustering method of vector quantization, using the signal light coming from the star 1214 GJ Exo-planet GJ 1214b and a reference signal light curve based on Mandel and Angol model.