Research Projects
Overview
Some of the technical courses I have enrolled in:
- Linear Algebra
- Probability and Statistics
- Signal Processing for Wireless Communication
- Information Theory
- Convex Optimization
- Machine Learning, Deep Learning and Bayesian Learning: Udemy
- MIT6.0002 - Introduction to Computational Thinking and Data Science
Development of Physical Layer Security Technology for 6G Communication Networks
In this project, we guarantee physical layer secrecy using an irregular reconfigurable intelligent surface (IRIS). Reconfigurable intelligent surface (RIS) has been considered as an effective way to enhance the performance of wireless systems, and has the ability to manipulate the wireless medium. RIS is a low cost device that has passive reflecting elements, each being able to reflect incident wireless signals at specified amplitude and phase. To ensure practicality of implementation and reduction in system complexity, we used IRIS where some reflecting elements are activated, unlike RIS which uses all elements in transmission.
Unified Design of Physical-Layer Security and Machine Learning for 5G Wireless Systems
I worked in the domain of deep learning and physical layer security. Deep learning has exhibited great potential in communication systems. Recent advances in deep learning based physical layer techniques have shown that the communication system can be modeled as an end-to-end learning task. I implemented end-to-end learning approach for physical layer security in Gaussian Multiple Access Wiretap Channel. I designed a loss function based on coset coding(traditional wiretap coding technique) and a modified kmeans algorithm. This loss function had controllable parameters that can be tuned to guarantee reliability, secrecy and user-priority. In addition to this, I designed training steps that was used on our deep learning model to guarantee model efficiency.
Publictions
- Emmanuel Obeng Frimpong, Taehoon Kim, Inkyu Bang, ’Physical-Layer Security with Irregular Reconfigurable Intelligent Surfaces for 6G Networks’ MDPI Sensors, 2023 [Link]
- Emmanuel Obeng Frimpong, Taehoon Kim, Inkyu Bang, ’Deep Learning Approach for Physical-Layer Security in Gaussian Multiple Access Wiretap Channel’ ICT Express-Elsevier, 2022. [Link]