Improved radiation detection algorithm using wavelet packet decomposition

Abstract

Easy-to-pilot drones are now readily available off the shelf, which have caused many security risks in public due to the unauthorised drones. However, limited works can be achieved to detect the drones, especially under strong interferences. This work considers the detection of drones via utilising a remote sensing algorithm developed in this work to extract the communication signals of the drone. The clutters including the static and/or non-static noises in the collected communication signals are eliminated based on the effective background suppression algorithm and linear trend suppression technique. Then, principal parts extraction method is utilised to remove the low-value parts of the radio-frequency signals, which can also improve via the automatic gain control method. Gaussian noise is suppressed using wavelet transform. Three errors of the acquired estimated frequency estimates from the above algorithms can be utilised to judge whether there are unauthorised drones or not in the monitoring airspace. The obtained detection results indicate the better performance and reliability of the detection algorithm developed in this study.