About
PhD Student & Graduate Research Associate
Performance and Resource Optimization in Networks (PROTON) Lab, Arizona State University
Interests
City
Tempe, Arizona, United States
krumman@asu.edu
I am a Ph.D. student in the School of Electrical, Computer and Energy Engineering at Arizona State University and a member of the Performance and Resource Optimization in Networks (PROTON Lab), advised by Dr. Eirini Eleni Tsiropoulou. I received my M.Sc. in Applied Statistics and Data Science in 2024 from Jahangirnagar University (JU), Bangladesh, and B.Sc. in Electrical and Electronic Engineering in 2016 from Islamic University of Technology (IUT), Bangladesh.
My research focuses on the integration of deep learning with integrated sensing and communications.
Education
PhD in Electrical Engineering
Arizona State University | Sep 2025 - Present
Expected graduation: Fall 2027
M.Sc. in Statistics and Data Science
Jahangirbagar University | Dec 2022 - Apr 2024
GPA: 3.67/4.00
B.Sc. in Electrical and Electronic Engineering
Islamic University of Technology | Dec 2012 - Dec 2016
GPA: 3.88/4.00
Research Experience
Graduate Research Associate
PROTON Lab, Arizona State University | Sep 2025 - Present
Supervisor: Dr. Eirini Eleni Tsiropoulou, Associate Professor
- Designed an agentic ISAC-aware emergency landing framework for drones, formulated as a constrained MDP and solved with Lagrangian PPO. It jointly optimizes descent and the sensing--communication power split under crash, SINR-outage, secrecy, and energy constraints, achieving 86% landing success and 95% secrecy-constraint satisfaction and outperforming Constrained SAC and non-ISAC baselines.
- Developed FusionLander, a multi-task Swin Transformer model for joint semantic segmentation and monocular depth estimation in emergency landing-zone detection, reaching 88.31% mIoU and 0.086 AbsRel at 0.11s per frame and outperforming U-Net and DeepLabv3+.
- Built a Maskable PPO scheduler for drone parcel pickup with a physics-based aerodynamic energy model and action masking for payload and battery limits, achieving about 2.6x the reward of heuristic baselines with roughly 47% lower energy per unit reward.
- Proposed RECON, a regret-matching framework for distributed localization in GNSS-denied networks, achieving near brute-force accuracy at a fraction of the computation time with sub-second convergence for 30 nodes.
Graduate Research Assistant
MENTIS Lab, Northeastern University | Jan 2024 - Aug 2025
Supervisor: Dr. Francesco Restuccia, Associate Professor
- Led SHRINK, a learning-based channel sounding protocol for Wi-Fi MIMO, implemented on commercial routers through Nexmon firmware modification of Broadcom chipsets. It cut feedback airtime and data overhead by 81% versus IEEE 802.11ac and raised throughput 24.5% over state-of-the-art methods, with continual learning to adapt to new environments using under 5 s of data.
- Co-developed MAGIC, a mmWave CSI-based gesture recognition system using temporal convolutional networks and meta-learning, achieving 99.24% accuracy and 98.82% in unseen environments and outperforming state-of-the-art domain adaptation methods by 14%.
- Contributed to m3MIMO, a fully digital 8x8 mmWave MU-MIMO testbed (57-64 GHz, 1 GHz bandwidth) built on RFSoC SDRs, where digital beamforming improved BER by 13.3x and SNR by 2.05x.
Professional Experience
Executive, Validation and Calibration Department
Square Pharmaceuticals PLC, Gazipur, Bangladesh | Jan 2022 - Dec 2023
- Completed 50+ equipment and system validations in compliance with cGMP guidelines and the company's Validation Master Plan.
- Took part in 15 internal quality management audits and served as the main point of contact for external auditors on validation and calibration activities.
Lecturer, Electrical and Electronic Engineering Department
European University of Bangladesh, Dhaka, Bangladesh | Sep 2018 - Dec 2021
- Taught theory and laboratory courses in Signals and Systems, Digital Signal Processing, Wireless Communications, and Electrical Circuits I, developing course materials and lab manuals.
- Supervised 15+ undergraduate thesis groups (3--4 students each) and student projects, guiding them from problem definition through design, implementation, and final defense.
Executive, Engineering Department
Square Toiletries Limited, Narayanganj, Bangladesh | Aug 2017 - Sep 2018
- Installed and commissioned new machinery in two projects, coordinating with suppliers and production teams, and maintained and calibrated production equipment to reduce downtime and material waste.
- Led cost-reduction process improvement projects on schedule and within budget, and brought production systems into compliance with safety and regulatory standards.
Publications
Journal Articles
Agentic Edge Intelligence for Low Altitude Economy Drone Safe Landing with Secure ISAC and Energy Aware Control
Authors: K M Rumman, Dimitrios Kaltsogiannis, Eirini Eleni Tsiropoulou
October 2026
Journal: IEEE Transactions on Consumer Electronics
A Deep Learning Framework for Emergency Drone Landing Zone Detection
Authors: K M Rumman, Andre Riker, Eirini Eleni Tsiropoulou
March 2026
Journal: IEEE Access
Intelligent Drone Pickup Scheduling via Deep Reinforcement Learning in Low Altitude Economy Networks
Authors: K M Rumman, Dimitrios Kaltsogiannis, Eirini Eleni Tsiropoulou
February 2026
Journal: Elsevier, Internet of Things
Fingerprint Based Biometric Attendance System
Authors: Md Saidur Rahman, K M Rumman, Rubab Ahmed, Abdur Rahman, Md Asaduzzaman Sarker
March 2023
Journal: European Chemical Bulletin, volume 12, pages 184–190
Design and Implementation of Intelligent Railway System
Authors: Md Saidur Rahman, Md Abdullah Kawser, K M Rumman, Fahrin Rahman, Rubab Ahmed
September 2022
Journal: Journal of Image Processing and Intelligent Remote Sensing, volume 2, pages 53–62
Conference Papers
RECON: Regret-Matching for Precise Cooperative Node Localization in GNSS-Denied Wireless Networks
Authors: K M Rumman, Eirini Eleni Tsiropoulou
December 2026
Conference: IEEE Global Communications Conference (GLOBECOM) {Accepted}
Shrink: Reducing MIMO Feedback Overhead in Wi-Fi with Dynamic Data-Driven Channel Sounding
Authors: K M Rumman, Francesca Meneghello, Khandaker Foysal Haque, Francesco Gringoli, Francesco Restuccia
October 2025
Conference: ACM MobiHoc 2025, pages 11–20
SignNet-Nano: Efficient Sign Language Recognition for Real-Time Edge Deployment
Authors: Rifat Zabin, Md. Riasat Tanjim Hossain, Khandaker Foysal Haque, K M Rumman
October 2025
Conference: IEEE COMPAS 2025, pages 1–6
Magic: Meta-Learning Adaptive Gesture Recognition with mmWave MIMO CSI
Authors: Khandaker Foysal Haque, K M Rumman, Arman Elyasi, Francesca Meneghello, Francesco Restuccia
May 2025
Conference: IEEE WoWMoM 2025, pages 131–140
m3mimo: An 8×8 mmWave Multi-User MIMO Testbed for Wireless Research
Authors: Khandaker Foysal Haque, Francesca Meneghello, K M Rumman, Francesco Restuccia
November 2024
Conference: ACM MobiCom 2024, pages 1922–1929