Research Vision
My research vision is to make cellular networks predictable, measurable, and controllable to enable reliable next-generation latency-sensitive applications, including AR/VR, cloud services, and connected autonomy. I combine large-scale real-world measurements, cross-layer performance analysis, and programmable 5G/6G experimentation to understand and address challenges in network reliability, latency, and performance. My goal is to develop lightweight, standards-compliant, and practical solutions that leverage AI-driven analytics and closed-loop network control to improve performance, energy efficiency, and reliability in real-world deployments.
Education
Northeastern University
Ph.D. in Computer Engineering
2021 - 2026
I am part of NuWINS lab and also affiliated with the Northeastern Institure for the Wireless Internet of Things (WIoT). As part of the WIoT I am involved in developing next generation cellular networks/testbeds. Besides, my other projects involve ensuring end to end performance, mobility, reliability and consistency of 5G networks.
Southern Illinois University Carbondale
Master's in Computer Engineering
2018 - 2020
I was part of the Ubiquitous Networking & Computing lab where I investigated Multi-Path TCP (MPTCP) performance in heterogenous networks.
Bangladesh University of Engineering and Technology (BUET)
B.Sc. in Electrical and Electronics Engineering
Bangladesh University of Engineering and Technology (BUET) is one of the top enginnering schools in Bangladesh.
Publications
* Equal contribution
Experience
*Developed a production-ready anomaly detection pipeline for E911 network KPIs using a lightweight Isolation Forest model, enriched with SIP/CCDR protocol context, including message types, cause codes, and failure indicators. Engineered historical baseline features using z-scores, rolling means, and rolling standard deviations to identify KPI degradation, achieving ~90% anomaly correlation with known network incidents.
*Benchmarked graph-based SSB optimization approaches using Google OR-Tools and SCIP on production network inventory data, incorporating neighbor weights to evaluate optimization performance in terms of execution time, number of configuration changes, and iterations required to reach optimality.
Investigated performnce of low-latency video streaming over cellular networks in a joint project with Facebook. The purpose of this study was to correlate RAN side KPIs with application layer metrics to enhance user Quality of Experience (QoE).
Skills
- Python, C/C++, Android (Java/Kotlin), HTML, CSS
Dataset | Software
- Dataset : for Cross-Sectional, Year-Long Study of 5G Uplink Performance. [IFIP 2024]
- Dataset : for Predictability of Fine-grained Cellular Network Throughput using Machine Learning Models. [IEEE MASS 2024]
- X5G : Our first ORAN-Compliant Multi-Vendor 5G testbed.
- Dataset for 5G network performance on the wheels. [ACM IMC 2023]
- Android app NextG-UP app for a crowd-sourcing based measurement study of 5G uplink performance evolution.
- MPTCP source code for an ROG Phone 2, from our WiNTECH 2021 paper.
- MPTCP source code for Pixel-5 Phone.
CV
- Here is my CV.
Awards and Professional Service
- 2026: Reviewer, COMCOM; Computer Networks; IEEE Transactions on Green Communications and Networking
- 2025: Reviewer, COMCOM; IEEE VTC; IEEE Transactions on Green Communications and Networking
- 2024: Reviewer, Internet of Things; Reviewer, Computer Networks; Web Chair, IEEE LANMAN; TPC Member, IEEE WCNC; Reviewer, IEEE ICC
- 2023: Web Chair, IEEE WoWMoM
- 2021: NSF-funded student travel grant for MobiCom; Reviewer, IEEE Globecom; Reviewer, SIGCOMM 5G-MEMU Workshop
- 2019: Selected for NSF-funded POWDER Network and Wireless Week, Salt Lake City, Utah