Acoustic Sensing From a Multi-Rotor Drone

Wang, Lin; Cavallaro, Andrea

doi:10.1109/jsen.2018.2825879

Cited by 57 publications

(77 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Proposed the ego-noise might be erroneously detected as target sound, thus degrading the noise suppression performance [17].…”

Section: Hybrid Methodsmentioning

confidence: 99%

“…Unsupervised approaches reduce the ego-noise using only the microphone array through beamforming [34]- [39], timefrequency spatial filtering [16], [17], or blind source separation [8], [16]. Delay-and-sum fixed beamforming has limited performance in improving the SNR [34], [35].…”

Section: A Ego-noise Reductionmentioning

confidence: 99%

“…Template-based [20] Reference-based [3], [28], [29] Unsupervised Beamforming [34]- [39] Blind source separation [8], [16] Time-frequency spatial filtering [16], [17]…”

Section: Supervisedmentioning

confidence: 99%

“…Time-frequency spatial filtering (TFS) [15]- [17] and blind source separation (BSS) [8], [16] represent the state of the Manuscript received: July 20, 2020 The authors are with Centre for Intelligent Sensing, Queen Mary University of London, London, UK (e-mail: {lin.wang, a.cavallaro}@qmul.ac.uk) art for ego-noise reduction on drones. Assuming the direction of arrival (DOA) of the target sound to be known, TFS works robustly in low-SNR scenarios by exploiting the timefrequency sparsity of the acoustic signals.…”

Section: Introductionmentioning

confidence: 99%

“…Assuming the direction of arrival (DOA) of the target sound to be known, TFS works robustly in low-SNR scenarios by exploiting the timefrequency sparsity of the acoustic signals. A drawback of TFS is its remarkable drop in performance when the DOA of the target sound is close to that of the source of egonoise [17]. The application of BSS to ego-noise reduction is straightforward as most BSS algorithms are based on independent component analysis (ICA) and thus do not require the knowledge of the locations of the microphones and the target sound source [18], [19].…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

A Blind Source Separation Framework for Ego-Noise Reduction on Multi-Rotor Drones

Wang

Cavallaro

2020

IEEE/ACM Trans. Audio Speech Lang. Process.

Self Cite

View full text Add to dashboard Cite

Acoustic sensing from a multi-rotor drone is heavily degraded by the strong ego-noise produced by the rotating motors and propellers. To address this problem, we propose a blind source separation (BSS) framework that extracts a target sound from noisy multi-channel signals captured by a microphone array mounted on a drone. The proposed method addresses the challenging problem of permutation alignment, in extremely low signal-to-noise-ratio scenarios (e.g. SNR <-15 dB), by performing clustering on the time activities of the separated signals across frequencies. Since initialization plays an important role to the success of clustering, we propose a pre-processing algorithm which uses time-frequency spatial filtering (TFS) to generate a reference to pre-align the permutation. The pre-alignment not only improves the performance of clustering and permutation alignment, but also solves the target-channel selection problem for BSS. The proposed method integrates the advantages of both TFS and BSS. Experimental results with real-recorded data show that the proposed method is capable of processing the audio stream continuously in a blockwise manner and also remarkably outperforms the state of the art.

show abstract

“…Proposed the ego-noise might be erroneously detected as target sound, thus degrading the noise suppression performance [17].…”

Section: Hybrid Methodsmentioning

confidence: 99%

Section: A Ego-noise Reductionmentioning

confidence: 99%

“…Template-based [20] Reference-based [3], [28], [29] Unsupervised Beamforming [34]- [39] Blind source separation [8], [16] Time-frequency spatial filtering [16], [17]…”

Section: Supervisedmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

A Blind Source Separation Framework for Ego-Noise Reduction on Multi-Rotor Drones

Wang

Cavallaro

2020

IEEE/ACM Trans. Audio Speech Lang. Process.

Self Cite

View full text Add to dashboard Cite

show abstract

Research on Optimal Control Method of Four Rotor UAV Based on BP Neural Network

Dai

Zhou

Zhao

et al. 2023

Lecture Notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering

View full text Add to dashboard Cite

Extraction of UAV Sound from a Mixture of Different Sounds

Ghani

Khan

2020

Acoust Aust

View full text Add to dashboard Cite

With the rapid advancement of technology of unmanned aerial vehicles (UAVs), security and safety of military and civil infrastructure have been jeopardized. By exploiting the unreliable capabilities of radar to detect low-flying UAVs with small radar cross section, they can be utilized for malicious purposes, e.g., unauthorized surveillance. To detect UAVs, therefore, various other techniques including audio sniffing/analysis of environment have been investigated. It has been shown recently that a UAV can be differentiated from other sound generating objects based on the various features extracted from the sound captured by a single or multiple acoustic sensors. However, features extraction and classification process can only give reliable results if it is fed with a sound generated by a single source. In practice, the captured sound may be a mixture of contribution of two or more different sources. In this paper, we investigate a well-known blind source separation technique, known as projection pursuit, to separate the constituent sounds in a mixture. We have considered a scenario when the different mixed unvoiced sounds are independent and non-Gaussian. The results show that in the given scenario, projection pursuit can be applied successfully to separate UAV sounds from various other unvoiced sounds.

show abstract

Acoustic Sensing From a Multi-Rotor Drone

Cited by 57 publications

References 35 publications

A Blind Source Separation Framework for Ego-Noise Reduction on Multi-Rotor Drones

A Blind Source Separation Framework for Ego-Noise Reduction on Multi-Rotor Drones

Research on Optimal Control Method of Four Rotor UAV Based on BP Neural Network

Extraction of UAV Sound from a Mixture of Different Sounds

Contact Info

Product

Resources

About