Applying the majority voting rule in acoustic detection and classification of drones

Abstract

This paper discusses an approach to target detection and classification based on acoustic signals collected using one single microphone. This study has applications to sonar or any other sound event classification system. We divide the problem into two parts, namely feature extraction and target detection and classification. We use an optimization step based on human auditory uncertainty. We employ a majority voting rule for every set of feature vectors, i.e., an estimate is only performed if the majority agrees. We conducted experiments using a single channel of the AIRA-UAS dataset, a public database of raw drone noises collected with an array of microphones mounted on a drone. This dataset comprises many different kinematics, with different spectra. The features we used are based on the Mel-Frequency Cepstral Coefficients (MFCC) and the Short-Time Fourier Transform of raw signals. We used the K-Nearest Neighbors algorithm for classification and adopted the crossvalidation strategy to evaluate the method. We observed that the use of MFCC results in less biased estimations, which favors the voting strategy. The detection in the proposed method reached a probability of false positive near 0%, even with a small set of votes, and a classification accuracy of 99.1%. These metrics satisfy the requirements of most civilian and military applications.

Simulation-Based Enhancement of SNR in Drone Communication through Uniform Linear Array Configurations

“…The signal-to-noise ratio represents the ratio between the desired signal and unwanted noise in a communication system. A high signal-to-noise ratio is paramount for clear and reliable communication between drones and ground control stations, as well as for accurate data transmission [1]. However, achieving a high signal-to-noise ratio in drone networks presents several challenges [2].…”

Enhancing TDE-based Drone DoA Estimation with Genetic Algorithms and Zero Cyclic Sum

“…A counter-drone system comprises two stages, e.g., threat evaluation and weapon assignment [13]. A significant challenge is the drone threat evaluation step, i.e., drone detection [14]- [16] and parameter estimation such as DoA [17]- [19], localization [20], [21], model [22], and payload additional weight [23].…”

“…Using different signals, a drone threat evaluation system can detect and estimate drone parameters; for instance, radar [24], radio frequency [25], optical [26]- [29], and acoustic signals [22], [30], [31]. Other works employ two or more sensors to detect drones, e.g., optical and acoustic [32], acoustic and RF [33], video and acoustics collected from a drone [34].…”