DronePrint

Kolamunna, Harini; Dahanayaka, Thilini; Li, Junye; Seneviratne, Suranga; Thilakaratne, Kanchana; Zomaya, Albert Y.; Seneviratne, Aruna

doi:10.1145/3448115

Cited by 28 publications

(9 citation statements)

References 43 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As a pre-processing step, we performed peak normalization on test set audio files, and normalized the MFCC vectors to have a mean of 0, and a standard deviation of 1. The pre-processing step was performed to mitigate the amplitude (loudness) dependency in the reference and target audio, as well as to bring the time-domain signal frames into comparable/similar ranges [42] .…”

Section: Methodsmentioning

confidence: 99%

Improving Dysarthric Speech Segmentation With Emulated and Synthetic Augmentation

Naeini,

Simmatis,

Jafari

et al. 2024

IEEE J. Transl. Eng. Health Med.

View full text Add to dashboard Cite

Background: Acoustic features extracted from speech can help with the diagnosis of neurological diseases and monitoring of symptoms over time. Temporal segmentation of audio signals into individual words is an important pre-processing step needed prior to extracting acoustic features. Machine learning techniques could be used to automate speech segmentation via automatic speech recognition (ASR) and sequence to sequence alignment. While state-of-the-art ASR models achieve good performance on healthy speech, their performance significantly drops when evaluated on dysarthric speech. Methods and procedures: Fine-tuning ASR models on impaired speech can improve performance in dysarthric individuals, but it requires representative clinical data, which is difficult to collect and may raise privacy concerns. This study explores the feasibility of using two augmentation methods to increase ASR performance on dysarthric speech: 1) healthy individuals varying their speaking rate and loudness (as is often used in assessments of pathological speech) 2) synthetic speech with variations in speaking rate and accent (to ensure more diverse vocal representations and fairness). Results: Experimental evaluations showed that fine-tuning a pre-trained ASR model with data from these two sources outperformed a model fine-tuned only on real clinical data and matched the performance of a model fine-tuned on the combination of real clinical data and synthetic speech. When evaluated on held-out acoustic data from 24 individuals with various neurological diseases, the best performing model achieved an average word error rate of 5.7% and a mean correct count accuracy of 94.4%. In segmenting the data into individual words, a mean intersectionover-union of 89.2% was obtained against manual parsing (ground truth). Clinical impact: It can be concluded that emulated and synthetic augmentations can significantly reduce the need for real clinical data of dysarthric speech when fine-tuning ASR models and, in turn, for speech segmentation.

show abstract

Section: Methodsmentioning

confidence: 99%

Improving Dysarthric Speech Segmentation With Emulated and Synthetic Augmentation

Naeini,

Simmatis,

Jafari

et al. 2024

IEEE J. Transl. Eng. Health Med.

View full text Add to dashboard Cite

show abstract

“…On the other hand, the identification model also needs fine-grained signal identification capability, which involves recognizing the specific drone type. This capability allows for obtaining detailed parameters of the drone, such as its physical attributes and flight speed [10], thus providing essential data support to the IoD system.…”

Section: Overview Of the Proposed Drone Rf Signal Identification Fram...mentioning

confidence: 99%

“…Active ADI technology primarily involves detecting drone targets through active radar echoes [8,9]. In contrast, non-cooperative ADI technology passively detects drone targets based on physical mediums such as acoustic signals [7,10], optical signals [11,12], and radio frequency (RF) signals emitted by drones [13][14][15]. Compared to other technologies, non-cooperative ADI technology based on drone RF signals offers a wider surveillance range and higher identification accuracy.…”

Section: Introductionmentioning

confidence: 99%

SignalFormer: Hybrid Transformer for Automatic Drone Identification Based on Drone RF Signals

Yan,

Han,

et al. 2023

Sensors

View full text Add to dashboard Cite

With the growing integration of drones into various civilian applications, the demand for effective automatic drone identification (ADI) technology has become essential to monitor malicious drone flights and mitigate potential threats. While numerous convolutional neural network (CNN)-based methods have been proposed for ADI tasks, the inherent local connectivity of the convolution operator in CNN models severely constrains RF signal identification performance. In this paper, we propose an innovative hybrid transformer model featuring a CNN-based tokenization method that is capable of generating T-F tokens enriched with significant local context information, and complemented by an efficient gated self-attention mechanism to capture global time/frequency correlations among these T-F tokens. Furthermore, we underscore the substantial impact of incorporating phase information into the input of the SignalFormer model. We evaluated the proposed method on two public datasets under Gaussian white noise and co-frequency signal interference conditions, The SignalFormer model achieved impressive identification accuracy of 97.57% and 98.03% for coarse-grained identification tasks, and 97.48% and 98.16% for fine-grained identification tasks. Furthermore, we introduced a class-incremental learning evaluation to demonstrate SignalFormer’s competence in handling previously unseen categories of drone signals. The above results collectively demonstrate that the proposed method is a promising solution for supporting the ADI task in reliable ways.

show abstract

“…This will maximize drone detection while filtering out other acoustic sources. More details on these types of acoustic sources can be found in [49][50][51][52][53]. Figure 15a,b shows the typical acoustic signatures of pistol gunfire and a small fourrotor drone, respectively, measured with a broadband microphone (Piezotronics Model 378A21).…”

Section: Operation In Airmentioning

confidence: 99%

“…This will maximize drone detection while filtering out other acoustic sources. More details on these types of acoustic sources can be found in [49][50][51][52][53].…”

Section: Operation In Airmentioning

confidence: 99%

Directional Resonant MEMS Acoustic Sensor and Associated Acoustic Vector Sensor

Ivancic,

Karunasiri,

Alves

2023

Sensors

View full text Add to dashboard Cite

This paper reports on the design, modeling, analysis, and evaluation of a micro-electromechanical systems acoustic sensor and the novel design of an acoustic vector sensor array (AVS) which utilized this acoustic sensor. This research builds upon previous work conducted to develop a small, lightweight, portable system for the detection and location of quiet or distant acoustic sources of interest. This study also reports on the underwater operation of this sensor and AVS. Studies were conducted in the lab and in the field utilizing multiple acoustic sources (e.g., generated tones, gun shots, drones). The sensor operates at resonance, providing for high acoustic sensitivity and a high signal-to-noise ratio (SNR). The sensor demonstrated a maximum SNR of 88 dB with an associated sensitivity of −84.6 dB re 1 V/μPa (59 V/Pa). The sensor design can be adjusted to set a specified resonant frequency to align with a known acoustic signature of interest. The AVS demonstrated an unambiguous, 360-degree, in-plane, azimuthal coverage and was able to provide an acoustic direction of arrival to an average error of within 3.5° during field experiments. The results of this research demonstrate the potential usefulness of this sensor and AVS design for specific applications.

show abstract

DronePrint

Cited by 28 publications

References 43 publications

Improving Dysarthric Speech Segmentation With Emulated and Synthetic Augmentation

Improving Dysarthric Speech Segmentation With Emulated and Synthetic Augmentation

SignalFormer: Hybrid Transformer for Automatic Drone Identification Based on Drone RF Signals

Directional Resonant MEMS Acoustic Sensor and Associated Acoustic Vector Sensor

Contact Info

Product

Resources

About