Analysis of Commercial Drone Sounds and Its Identification

EasyChair Preprint 4168

Abstract

The usage of quadcopter types of drones is now on a mature and practical stage. Numbered companies are manufacturing them and expanding their application. Considerable characteristics of this type of flying object that has the maneuverability and practicality is now being focused on how we control this among our urban life from its offensive liabilities. They are small enough to avoid many current airborne detection systems and cheap enough to use them as disposable. In this paper, we tried to analyze the sounds of a subset of commercial drones as quadcopter types and also built a trained simple non-linear neural network filter to classify them among the given sound samples. We borrowed Mel-frequency cepstral coefficients as the well-known methodology of sound analysis but including some of the parameter adjustments, and applied LeNet neural network filter structure for its simple classification. In order to maintain the information of adjacent samples among the series of wave samples, 2-D spectrogram planning was applied as the input signal of the filter. Most of the frequencies from drones were observed as gathered around 3 to 5Khz, up to 10Khz, and adjusted LeNet architecture could classify over 10 types of drone categories with over 95% of accuracy.

Keyphrases: Convolutional Neural Network, Drone sound, LeNet, Neural Network Capacity, Quadcopter, commercial drone sound, mel frequency kepstrum

@booklet{EasyChair:4168,
  author    = {Sinwoo Yoo and Hyukjun Oh},
  title     = {Analysis of Commercial Drone Sounds and Its Identification},
  howpublished = {EasyChair Preprint 4168},
  year      = {EasyChair, 2020}}