Hi all, I'm a Ph.D. student, ECE@University of Southern California, working in Autonomous Networks Research Lab with Dr. Bhaskar Krishnamachari . I love working on exciting problem statements involving Distributed Systems and Embedded IoT. Know more about my past research experience through my work🚀
At Autonomous Networks Research Lab advised by Dr. Krishnamachari.
For courses EE597 (Wireless Networks) and EE250 (Distributed Systems and IoT)
Worked on various Entrepreneurship and Technology Innovation startups and worked as an IoT-ML intern at Citris Labs, Berkeley Engineering.
Ranked top 3%(First Class with Distinction) in the entire batch, CGPA:3.7/4, Dept. Rank~ 8/125.
Using a carefully designed testbed of 51 Raspberry Pi (RPi) devices, we mimic the complexities of a large-scale IoT system. To the best of our knowledge, this work is the first to deploy Edge Machine Learning for DDoS detection in a testbed of such a size replicating some of the complexities and intricacies of real-world IoT networks. We have designed the flow-level traffic and complexities of networked devices to closely mirror real-world scenarios. Furthermore, we introduce real-time inference capabilities by deploying the trained LSTM/MM-WC model on each Rpi device. This enables each node to autonomously detect DDoS attacks using real-time traffic, further enhancing the practicality of our framework.
Accepted in [SenSys'23: The 21st ACM Conference on Embedded Networked Sensor Systems]
Published in [SenSys'23: The 21st ACM Conference on Embedded Networked Sensor Systems]
Published in [IEEE Transactions on Networking]
We introduce DARSAN, a decentralized review system designed for Non-Fungible Token (NFT) marketplaces, to address the challenge of verifying the quality of highly resalable products with few verified buyers by incentivizing unbiased reviews.
Under the supervision of Dr. Sudip Misra CSE Dept., developed a distributed framework for enabling Transfer Learning in B5G Powered Industrial-IoT(first of its kind).
A demand based delivery mechanism was demonstrated for distributing models among fog nodes.
An Optimization algorithm using Integer-Linear-Programming was developed for minimization of Transmission timein B5G networks.
Published in [IEEE Transactions on Industrial Informatics,October 2021]
Developed an IoT sports wearable through Body Area Network, WSN with fog nodes in constrained environments.
Implemented AI on the edge, developed a novel algorithm to predict anomalies in regular athlete metrics.
Designed a storage based Resource Allocation algorithm, a Path Hopping Optimization algorithm for transmissionof files from cloud to end user via fog nodes, a Transmission Latency Optimization algorithm in 5G Network.
Aims to help eradicate language barriers for English as foreign language speakers, AR glasses with real-time subtitles.
Developed a Neural Network model for Voice Recognition and real-time language translation.
The entire system was processed on a low power Raspberry Pi Zero edge computer.
Implemented pre-trained Mask R-CNN using COCO dataset with 94% accuracy.
Implemented Image Captioning, OCR in the model and designed a text-to-speech feedback mechanism after imageclassification through audio output(I2S) for the visually impaired to navigate.
Worked under Dr. P. Venkateswaran on a ML framework involving cloud-fog architecture for handling high volume Industrial-IoT data from multiple industrial sensors.
Developed of Neural Networks, Markov Decision Processes on FPGAs, for establishing model randomness with probabilistic distributions. Worked on Convex Optimizations towards reduction in communication, prediction latency.
Computer Science Department
Best Hardware Hack
Hackathon Winner
UC Berkeley
School of Computing
SCET, UC Berkeley
ETHDenver 2019
19,900USD for studying@UCB
Phi Kappa Sigma Berkeley Associates
From Tamoghna's work under my supervision at IIT-Kharagpur, I can attest to his integrity and ambition towards research. The persistence and goal achieving nature of his will definitely be valuable to any group that he joins. I have every confidence in Tamoghna and give my highest recommendations. Good luck! 🔥
Looking at Tamoghna's journey till today from his early school, undergraduate days I can clearly envision him as an upcoming prolific researcher in his field. His profile reflects hardwork, determination and the steps he climbed to reach where he is today. I can comment that with the kind of extra miles he puts into his research, Tamoghna has in him the potential to churn out the finest results.
Tamoghna was the youngest member in all his teams but was the top excelling student in his classes at UC Berkeley. I've seen him grow throughout the last 3 years. Be it in-depth research or practical application, Tamoghna is your guy! 🔥
A perfect blend of Research, Application and Innovation = Tamoghna. Throughout his work at Quasistatics Inc, Tamoghna set deadlines for himself and sincerely met every deadline and thus completed the assigned targets for the projects on time. He worked with us on 2 MVPS for 7 months and looking at his dedication in IoT and Wearables, I'm sure about his bright future as an impactful Researcher in the community. I wish him all the best for his Doctoral studies at USC.
Tamoghna was among the top 2 percent of his batch and completed his bachelors with the highest grade. I will remember his stubborn, obsessive nature to troubleshoot and take a deeper look into any problem. Guiding Tamoghna throughout the last 2 years, I am certain that he is ready to make an impact in the research community. A versatile individual with a plethora of application-based skills along with cutting-edge research. All the best! 👍