Danilo Pena

Signal processing · Machine learning · Sensing

I work on signal processing and machine learning for sensing systems, mostly with audio. I am co-founder and CTO of Ondova in Montreal, where we extract physiological and interaction signals from the speaker already inside headphones and earbuds, without added sensors.

Before Ondova I led engineering at Ohmic Technologies, was a postdoctoral researcher at the Faculty of Engineering of the University of Porto, and built modem software at Samsung's R&D institute in Brazil. My PhD, at UFRN, was on acoustic source localization under impulsive noise.

Portrait of Danilo Pena

Research interests

Experience

2025 – now
Co-founder and CTO, Ondova Montreal
2023 – 2025
Director of Engineering, Ohmic Technologies Montreal
2022 – 2023
Signal Processing Engineer, Ohmic Technologies Montreal
2021 – 2023
Postdoctoral Researcher, SYSTEC, Faculty of Engineering, University of Porto Porto
2020 – 2021
Senior Software Engineer, Sidia, Samsung R&D Institute Manaus
2015 – 2017
Assistant Professor, Department of Communications Engineering, UFRN Natal
2012 – 2013
Embedded Software Engineer, Potychip Natal

Education

2019
PhD, Federal University of Rio Grande do Norte (UFRN). Acoustic source localization under impulsive noise.
2014
MSc, Federal University of Rio Grande do Norte (UFRN). Correntropy-based similarity measures on FPGA.
2012
BSc, Electrical Engineering, Federal University of Rio Grande do Norte (UFRN).

Selected publications

Stable Hybrid Cross-Attention Fusion for Audio-Visual Event Recognition
P. B. Dehaghani, D. Pena, A. P. Aguiar
IEEE ISC2, 2026
arXiv
Investigation of Feature Selection and Pooling Methods for Environmental Sound Classification
P. Dehaghani, D. Pena, A. P. Aguiar
ISPA, 2025
arXivPDF
Smart Plant Growth Device with Digital Twin Modeling for Precision Agriculture
D. Pena, A. P. Aguiar, F. Lobo Pereira, O. Hadj-Abdelkader, H. Bouzebiba, P. Dehaghani
European patent EP4483706A1, 2025
Espacenet
Energy-Efficient IoT-Based Light Control System in Smart Indoor Agriculture
O. Hadj Abdelkader, H. Bouzebiba, D. Pena, A. P. Aguiar
Sensors, 2023
DOI
A Simulation Tool for the Analysis of TDOA Methods
A. B. F. Diniz, D. S. Pena, V. A. de Sousa Jr.
IPIN, 2021
PDF
Characterization of Noise Level Inside a Vehicle Under Different Conditions
D. Flor, D. Pena, L. Pena, V. A. de Sousa Jr., A. Martins
Sensors, 2020
DOI
Acoustic Impulsive Noise Based on Non-Gaussian Models: An Experimental Evaluation
D. Pena, C. Lima, M. Dória, L. Pena, A. Martins, V. Sousa
Sensors, 2019
DOI

Full list on Google Scholar.

Software

AFX

Audio feature extraction in Python.

Python
TDOA methods

Simulation tool for time-difference-of-arrival and angle-of-arrival estimators, including robust variants.

MATLAB
Sound classification on STM32

Environmental sound classification CNN on an STM32F411RE, trained on UrbanSound8K.

C
RT-AEP

Real-time audio effects processor on PortAudio.

C++
Highlight detection

Specular highlight detection in images from the Mahalanobis distance.

MATLAB
MSP430 course

Short course on the MSP430 microcontroller.

C