Abstract—Drones, or Unmanned Aerial Vehicles (UAVs), have become much cheaper to buy and may become a security risk due to their small size and ability to fly autonomously. Many traditional detection techniques (radar, radio frequency, and visual system) do not work as well when trying to detect small UAVs. Another option has been to use an acoustic method of detection, as drones produce identifiable sounds as a result of their motors and propellers. A low-cost acoustic drone detection system will be developed using machine learning and deep learning. Data were obtained through audio recordings of seven types of drones from publicly available GitHub datasets and by collecting environmental noise through the British Broadcasting Corporation (BBC) Sound Library and YouTube. In order to classify drones versus non-drones, 26 Mel-Frequency Cepstral Coefficients (MFCC) features (13 static and 13 delta) are extracted from the audio signals. Additionally, clustering analysis is performed to confirm the uniqueness of the drone’s acoustic signatures before training the models with the data. Using a convolutional recurrent neural network classifier yielded an accuracy of 97%, F1‒Score of 0.97, and Receiver Operating Characteristic (ROC), Area Under ROC Curve (AUC) of 0.99, outperforming both the Balanced Random Forest (BRF) and Multilayer Perceptron (MLP) classifiers. Furthermore, it demonstrates considerable robustness when tested under noisy conditions.

Index Terms—acoustic signal, Convolutional Recurrent Neural Network (CRNN), deep learning, drone detection, machine learning, Mel-Frequency Cepstral Coefficients (MFCC), Unmanned Aerial Vehicle (UAV)

Cite: SShahad W. Jasim  and Saad S. Hreshee, "Mel-Frequency Cepstral Coefficients (MFCC)-based Acoustic Drone Detection Using Machine Learning and Deep Learning Algorithms," International Journal of Electrical and Electronic Engineering & Telecommunications, vol. 15, no. 4, pp. 267-275, 2026. doi: 10.18178/ijeetc.15.4.267-275

Copyright © 2026 by the authors. This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited (CC BY 4.0).