Address: 406 Hardy Road, Simrall Electrical Engineering, MS 39762, USA
I am Sabyasachi Biswas, a Postdoctoral Researcher in Electrical and Computer Engineering at Mississippi State University, having successfully defended my Ph.D. dissertation in Summer 2025. I hold a strong academic foundation from the Department of Electrical and Electronic Engineering at Bangladesh University of Engineering and Technology (BUET). With over four years of focused experience in digital and radar signal processing, my research lies at the intersection of signal processing, real-time RF sensing, and deep learning, particularly in developing interpretable and computationally efficient neural networks for human activity recognition, wireless signal classification, and autonomous systems. I have worked extensively with radar, LiDAR, and Wi-Fi-based sensing platforms, building end-to-end machine learning pipelines that process raw sensor data using custom-designed layers rooted in DSP principles.
My professional experience includes a six-month Co-Op at Bose Corporation, where I designed physics-informed machine learning algorithms for immersive audio systems, enabling automatic calibration and precise IMU-head axis mapping. As a graduate research assistant at the IMPRESS Lab, I led multiple interdisciplinary projects involving sensor fusion, real-time data acquisition, multi-target classification, and angle-aware radar data conditioning using embedded platforms. I have also contributed to the design and deployment of RF waveform datasets using software-defined radios and developed high-resolution time-frequency analysis tools for efficient classification in noisy environments. Additionally, I have worked on several funded projects supported by agencies such as NSF, AFRL, ERDC, and USDA.
To date, I have authored six peer-reviewed journal articles and eighteen conference papers, with several contributions published in top-tier IEEE journals. I have received multiple accolades for my research, including the ECE Best Graduate Researcher Award and consecutive recognitions at the Mississippi State University Graduate Student Research Symposium. With a passion for integrating theory with real-world applications, I am committed to driving innovations in radar systems, embedded AI, and spectrum-efficient sensing technologies that address both industrial and environmental challenges.
Radar Signal Processing
Software-Defined Radio
Wi-Fi-Based Sensing
Human Activity Recognition
Wireless Communication
Deep Learning and Explainable AI
Ph.D. in Electrical and Computer Engineering Aug 2021 – Present
Mississippi State University, MS, USA
GPA: 4.00/4.00
Dissertation Title: Complex-Valued Parameterized Learnable Filter Banks for Time-Frequency Domain Based Classification
Major Professor: Dr. John E. Ball
Supervisor: Dr. Ali C. Gurbuz
Dissertation Defended on July 22nd, 2025
Bachelor of Science in Electrical and Electronics Engineering Feb 2015 – Apr 2019
Bangladesh University of Engineering and Technology (BUET), Dhaka, Bangladesh
Thesis Title: Optimization of Electric Vehicle Charging to Shave Peak Load for Integration in Smart Grid
Supervisor: Dr. Md. Forkan Uddin
September 2025:
September 26, 2025, I received the James Worth Bagley College of Engineering Best Graduate Student Researcher award for 2024-25 Academic year.
Starting September 1, 2025, I will serve as a Postdoctoral Research Associate in Electrical and Computer Engineering at Mississippi State University.
As a Postdoctoral Research Associate, I will contribute to developing advanced signal processing, machine learning, and sensor fusion algorithms for multi-modal threat detection. In addition to supervising graduate and undergraduate researchers and leading experimental data collection with radar, acoustic, lidar and camera systems.
I will also strengthen my grant writing skills under the mentorship of Dr. Ball and Dr. Ali C. Gurbuz.
Publication Alert!!
“Interpretable CNN models for computationally efficient bearing fault diagnosis using learnable Gaussian/Sinc filters” published in Manufacturing Letters (Vol. 44, Suppl., 2025, pp. 110–120, https://doi.org/10.1016/j.mfglet.2025.06.015). This work, in collaboration with Dr. Mahathir Mohammad Bappy (Louisiana State University) and Dr. Abdullah Al Mamun (Old Dominion University), presents interpretable CNN architectures using parameterized Gaussian and Sinc filters for real-time bearing fault diagnosis.
July 2025:
Big News: I have successfully defended my PhD Dissertation titled, "Complex-Valued Structured Parameterized Learnable Filter Banks for Time-Frequency Domain Based Classification" on 22nd July.
Our paper, 'A Multimodal Video and Radar Fusion Framework for High-Accuracy Isolated Sign Language Recognition,' by Sultan Mohammad Monjur, Sabyasachi Biswas, and Ali C. Gurbuz, has been accepted to the IEEE/CVF International Conference on Computer Vision (ICCV) 2025.
Our paper, 'RF-ChessSIGN: Radar-enabled Human-Computer Interaction in a Real-Time Sign Language-Controlled Game,' by Kenneth DeHaan, Emre Kurtoglu, Sabyasachi Biswas, Caroline Kobek Pezzarossi, Darrin J. Griffin, Chris Crawford, Ali C. Gurbuz, Evie A. Malaia, Abraham Glasser, Raja Kushalnagar, Sevgi. Z Gurbuz, has been accepted to the IEEE/CVF International Conference on Computer Vision (ICCV) 2025.
June 2025:
Our team, Sultan Mohammad Monjur and Sabyasachi Biswas, was named 2nd Runner-Up in the Kaggle competition ‘1st Multimodal Italian Sign Language Recognition’ [Link], under the guidance of Dr. Ali Cafer Gurbuz.
Publication Alert!! Three papers have been published in the SPIE Conference,
Bruce Hicks, Sabyasachi Biswas, John E. Ball, Ali C. Gurbuz, "Automatic classification of radar and communication waveforms through interpretable deep learning," Proc. SPIE 13471, Radar Sensor Technology XXIX, 134710M (29 May 2025); https://doi.org/10.1117/12.3054111
Kester Nucum, Sabyasachi Biswas, John E. Ball, "Micro-Doppler models of drones, birds, and bird-like drones," Proc. SPIE 13471, Radar Sensor Technology XXIX, 134710D (29 May 2025); https://doi.org/10.1117/12.3052657
Sifat Zina Karim, Sabyasachi Biswas, John Ball, "Learnable 2D Gaussian filters for computationally efficient abdominal organ classification," Proc. SPIE 13458, Real-Time Image Processing and Deep Learning 2025, 1345802 (28 May 2025); https://doi.org/10.1117/12.3053222
May 2025:
Attended and presented at Radar Conference 2025, Atlanta, Georgia, USA.
Our paper, ‘Learnable Gaussian Filter-Based Automatic RF Waveform Modulation Recognition,’ by Bruce Hicks, Sabyasachi Biswas, John E. Ball, and Ali C. Gurbuz, was selected as a Student Paper Competition finalist at the Radar Conference.
April 2025: I was named the Bagley College of Engineering’s 2024-25 Best Graduate Researcher at Mississippi State University.
February 2025: I have been awarded the ECE's 2024-25 Best Graduate Researcher at Mississippi State University.