Research

Topic details:

My work is focussed on Machine Learning applications for Physical Layers of 5G and Beyond 5G Networks. 

5G is the latest technology in cellular communication which primarily has three prominent use cases. These are eMBB (high-speed connectivity), uRLLC (Ultra-Reliable and Low Latency), and mMTC (a large number of users). These use cases include applications such as cellular communication, the Internet of Things, and self-driving cars. As a result, the number of devices is growing, and overall traffic is also increasing. However, designing and optimizing wireless networks with such a level of complexity is tedious. Some challenges in designing such wireless networks are interference from other UE or BS, high energy usage, and signaling overheads. Conventional mathematical models often may not exist for such complex systems; even if they do, they may be intractable. Thus, the networks have to self-optimize - which is where ML comes in. Furthermore, ML models are very good at extracting meaningful mapping within the data more effectively and accurately than humans. Thus we are motivated to develop ML-based solutions to complex 5G and B5G physical layer applications. 

There are two ways of looking at physical layer optimization with Machine Learning. The first approach is in line with traditional system models that include a chain of signal processing blocks, each optimized separately. Under this approach, there are several areas, such as modulation classification, sequence detection, channel estimation, etc., where ML is naturally a good fit

Publications:

1. R. Singh, A. K. Yerrapragada and R. K. Ganti, “Physical Layer Design for Ambient IoT” –Accepted at IEEE Communication Standards Magazine June 2025 edition.

2. R. Singh, A. K. Yerrapragada and R. K. Ganti, “A Machine Learning based Hybrid Receiver for 5G NR PRACH” - Accepted at IEEE International Conference on Machine Learning for Communication and Networking (ICMLCN) 2025.

3. R. Singh, A. K. Yerrapragada, J. K. S and R. K. Ganti, "Enhancements for 5G NR PRACH Reception: An AI/ML Approach," 2024 Wireless Telecommunications Symposium (WTS), Oakland, CA, USA, 2024, pp. 1-6, doi: 10.1109/WTS60164.2024.10536687. 

4. J. K. S et al., "Physical Layer Design of a 5G NR Base Station," 2024 National Conference on Communications (NCC), Chennai, India, 2024, pp. 1-6, doi: 10.1109/NCC60321.2024.10485845.