Debopriya Roy Dipta

Debopriya Roy Dipta

he/him

Ph.D. Candidate | Graduate Research Assistant | Computer Engineering

Publications

You can also find my articles on my Google Scholar profile.

Conference Papers

Dynamic Frequency-Based Fingerprinting Attacks against Modern Sandbox Environments

Published in 2024 IEEE 9th European Symposium on Security and Privacy (EuroS&P), 2024

We demonstrate that CPU dynamic frequency scaling can be exploited to fingerprint and identify web content executing inside modern sandbox environments such as V8 and Cloudflare Workers, bypassing isolation guarantees.

Recommended citation: D. R. Dipta, T. Tiemann, B. Gulmezoglu, E. Marin and T. Eisenbarth, "Dynamic Frequency-Based Fingerprinting Attacks against Modern Sandbox Environments," 2024 IEEE 9th European Symposium on Security and Privacy (EuroS&P), pp. 327-344, Vienna, Austria, 2024.
Download Paper

DefWeb: Defending User Privacy against Cache-based Website Fingerprinting Attacks with Intelligent Noise Injection

Published in Proceedings of the 39th Annual Computer Security Applications Conference (ACSAC '23), 2023

DefWeb is a browser-level defense that injects carefully crafted noise into cache-based side-channel signals to mislead website fingerprinting classifiers while maintaining low overhead for legitimate users.

Recommended citation: S. Son, D. R. Dipta and B. Gulmezoglu, "DefWeb: Defending User Privacy against Cache-based Website Fingerprinting Attacks with Intelligent Noise Injection," Proceedings of the 39th Annual Computer Security Applications Conference (ACSAC '23), Association for Computing Machinery, New York, NY, USA, 379–393, 2023.
Download Paper

DF-SCA: Dynamic Frequency Side Channel Attacks are Practical

Published in Proceedings of the 38th Annual Computer Security Applications Conference (ACSAC '22), 2022

We show that CPU dynamic frequency scaling — a ubiquitous performance optimization — creates a practical side channel that leaks sensitive information across privilege boundaries without requiring privileged access or precise timers.

Recommended citation: D. R. Dipta and Berk Gulmezoglu, "DF-SCA: Dynamic Frequency Side Channel Attacks are Practical," Proceedings of the 38th Annual Computer Security Applications Conference (ACSAC '22), Association for Computing Machinery, New York, NY, USA, 841–853, 2022.
Download Paper

Comparison of Different Hybrid Renewable Energy Systems With Optimized PV Configuration to Realize the Effects of Multiple Schemes

Published in 2019 1st International Conference on Advances in Science, Engineering and Robotics Technology (ICASERT), 2019

We compare different hybrid renewable energy system configurations with optimized photovoltaic setups to evaluate performance trade-offs and the effects of various integration schemes on energy yield and reliability.

Recommended citation: F. I. Bappy, M. Jahirul Islam, A. K. Podder, D. R. Dipta, H. M. Resalat Faruque and E. Hossain, "Comparison of Different Hybrid Renewable Energy Systems With Optimized PV Configuration to Realize the Effects of Multiple Schemes," 2019 1st International Conference on Advances in Science, Engineering and Robotics Technology (ICASERT), Dhaka, Bangladesh, 2019.

Design of a Convolutional Neural Network Based Smart Waste Disposal System

Published in 2019 1st International Conference on Advances in Science, Engineering and Robotics Technology (ICASERT), 2019

We design a smart waste disposal system using a convolutional neural network to automatically classify and sort waste materials, enabling automated recycling and reducing manual sorting effort.

Recommended citation: M. S. Haque Sunny, D. R. Dipta, S. Hossain, H. M. Resalat Faruque and E. Hossain, "Design of a Convolutional Neural Network Based Smart Waste Disposal System," 2019 1st International Conference on Advances in Science, Engineering and Robotics Technology (ICASERT), Dhaka, Bangladesh, 2019.

Efficient Hardware Implementation of 256-bit ECC Processor Over Prime Field

Published in 2019 International Conference on Electrical, Computer and Communication Engineering (ECCE), 2019

We present an efficient FPGA-based hardware implementation of a 256-bit Elliptic Curve Cryptography (ECC) processor optimized for speed and area over prime fields.

Recommended citation: M. S. Rahman, M. S. Hossain, E. H. Rahat, D. R. Dipta, H. M. R. Faruque and F. K. Fattah, "Efficient Hardware Implementation of 256-bit ECC Processor Over Prime Field," 2019 International Conference on Electrical, Computer and Communication Engineering (ECCE), Bangladesh, 2019.

Dynamic Time Warping Assisted SVM Classifier for Bangla Speech Recognition

Published in 2018 International Conference on Computer, Communication, Chemical, Material and Electronic Engineering (IC4ME2), 2018

We propose a Bangla speech recognition system that combines Dynamic Time Warping (DTW) for feature alignment with a Support Vector Machine (SVM) classifier, achieving robust recognition performance on a collected Bangla speech dataset.

Recommended citation: M. M. Rahman, D. R. Dipta and M. M. Hasan, "Dynamic Time Warping Assisted SVM Classifier for Bangla Speech Recognition," 2018 International Conference on Computer, Communication, Chemical, Material and Electronic Engineering (IC4ME2), Rajshahi, Bangladesh, 2018.

Journal Articles

Systematical Evasion From Learning-Based Microarchitectural Attack Detection Tools

Published in IEEE Journal on Emerging and Selected Topics in Circuits and Systems (IEEE JETCAS), 2024

We present a systematic methodology for evading machine learning-based microarchitectural attack detectors, demonstrating that adversarial perturbations to hardware performance counter traces can reliably bypass state-of-the-art detection tools.

Recommended citation: D. R. Dipta, J. Tan and B. Gulmezoglu, "Systematical Evasion From Learning-Based Microarchitectural Attack Detection Tools," IEEE Journal on Emerging and Selected Topics in Circuits and Systems (IEEE JETCAS), vol. 14, no. 4, pp. 823-833, Dec. 2024.
Download Paper

MAD-EN: Microarchitectural Attack Detection Through System-Wide Energy Consumption

Published in IEEE Transactions on Information Forensics and Security (IEEE TIFS), 2023

MAD-EN leverages system-wide energy consumption data (via Intel RAPL) as a novel signal for detecting microarchitectural attacks, achieving high detection accuracy across diverse attack classes without requiring hardware performance counters.

Recommended citation: D. R. Dipta and B. Gulmezoglu, "MAD-EN: Microarchitectural Attack Detection Through System-Wide Energy Consumption," IEEE Transactions on Information Forensics and Security (IEEE TIFS), vol. 18, pp. 3006-3017, 2023.
Download Paper

Solar Energy in the United States: Development, Challenges and Future Prospects

Published in Energies, 2021

A comprehensive review of solar energy development in the United States, examining installation trends, policy challenges, grid integration issues, and future prospects for utility-scale and distributed solar deployment.

Recommended citation: S. Tabassum, T. Rahman, A. U. Islam, S. Rahman, D. R. Dipta, S. Roy, N. Mohammad, N. Nawar, E. Hossain, "Solar Energy in the United States: Development, Challenges and Future Prospects," Energies, 14(23), Article 8142, 2021.

Metrics and Enhancement Strategies for Grid Resilience and Reliability during Natural Disasters

Published in Applied Energy, 2021

We review and propose quantitative metrics for assessing electrical grid resilience and reliability under natural disaster conditions, and evaluate enhancement strategies including microgrids, distributed energy resources, and advanced communication systems.

Recommended citation: E. Hossain, S. Roy, N. Mohammad, N. Nawar, D. R. Dipta, "Metrics and enhancement strategies for grid resilience and reliability during natural disasters," Applied Energy, Volume 290, 2021.

A Comprehensive Review of the Load Forecasting Techniques Using Single and Hybrid Predictive Models

Published in IEEE Access, 2020

A comprehensive survey of electrical load forecasting methods, covering statistical models, machine learning approaches, and hybrid techniques, with comparative analysis of their accuracy, computational requirements, and applicability across forecasting horizons.

Recommended citation: A. A. Mamun, M. Sohel, N. Mohammad, M. S. Haque Sunny, D. R. Dipta and E. Hossain, "A Comprehensive Review of the Load Forecasting Techniques Using Single and Hybrid Predictive Models," IEEE Access, vol. 8, pp. 134911-134939, 2020.

Preprints & Under Review

Exploiting and Mitigating Information Leakage at the Container-Kernel Interface

Published in Under Review, 2025

We systematically study information leakage channels that arise at the interface between containerized applications and the Linux kernel, demonstrating practical co-residence detection and cross-container side-channel attacks in serverless deployments.

Recommended citation: D. R. Dipta, E. Marin, B. Gulmezoglu and T. Eisenbarth, "Exploiting and Mitigating Information Leakage at the Container-Kernel Interface." Under review.

DimShield: Exploring Intrinsic Dimension Estimation for Enhanced Machine Learning Security

Published in Under Review, 2025

DimShield is a runtime defense framework that leverages intrinsic dimension estimation to detect and mitigate adversarial attacks against machine learning models, providing robust protection without significant performance overhead.

Recommended citation: D. R. Dipta, K. Christofferson, S. Seonghun and B. Gulmezoglu, "DimShield: Exploring Intrinsic Dimension Estimation for Enhanced Machine Learning Security." Under review.

uGen: From Gaps to Exploits — A Multi-Agent, Retrieval-Augmented System for End-to-End Microarchitectural Attack Code Generation

Published in Under Review, 2025

uGen is an LLM-assisted multi-agent framework that autonomously identifies gaps in the microarchitectural attack literature, synthesizes relevant knowledge via retrieval-augmented generation, generates exploit code, tests it, and iteratively refines both attack and defense strategies.

Recommended citation: D. R. Dipta, T. Tiemann, B. Gulmezoglu, E. Marin and T. Eisenbarth, "uGen: From Gaps to Exploits — A Multi-Agent, Retrieval-Augmented System for End-to-End Microarchitectural Attack Code Generation." Under review.