This research primarily centers on exploring the implementation of Quantum Key Distribution (QKD) for secure communication. QKD utilizes quantum mechanical properties to establish secure channels, ensuring utmost confidentiality in data transmission through quantum principles.

The project is dedicated to developing novel and efficient optimization frameworks tailored for resource allocation in advanced communication systems. Our emphasis is on creating solutions that not only enhance the efficiency of resource allocation but also address specific challenges unique to advanced communication environments. By leveraging cutting-edge optimization techniques, we aim to contribute to the advancement of communication networks and ensure optimal resource utilization.

The project aims to create a learning-based framework for identifying heart disease by exploiting electrocardiogram (ECG) data. This framework will focus on the detection, classification, and prediction of heart disease.  We concentrate on developing the framework for prediction with long-term forecasting capability and multi-variant generalization capability, as well as detection and classification with high accuracy and efficient generalization capability under intra-patient or/and inter-patient paradigms.  By leveraging cutting-edge techniques in signal processing and artificial intelligence, we aim to contribute to advancing the heart disease identification framework and ensuring its efficiency as a computer-assisted system for cardiologists. Currently, we have published novel works related to the intra-patient ECG classification and the ECG time series forecasting.

The project focuses on developing end-to-end learning frameworks within space optical communications, addressing both point-to-point and multiple access channels. This initiative resulted in significant publications, including an IEEE GLOBECOM paper centered around a point-to-point autoencoder model tailored for space optical communications. This model was specifically designed to operate within realistic fading channels simulated by a system toolkit.

Furthermore, the project expanded its scope within IEEE Transactions on Machine Learning and Communications Networks. The extended research delved into exploring diverse fading channels and varied code rates while aiming to reduce computational complexity without compromising robustness.

Another contribution emerged in an IEEE Communications Letters publication, introducing an innovative approach for examining an autoencoder model in multi-user space optical communications. This pioneering approach introduced a layered framework integrating batch normalization in the encoder and layer normalization in the decoders. This framework showcased superior performance compared to existing state-of-the-art learning frameworks, signifying promising advancements in this domain