5G/6G-Centric AI

Project: Future Internet Architecture Model  

Duration: 03.2011 - 02.2012 

Industry: KCC Korea Communication Communications Society and ETRI

Description: The current Internet suffers from various issues, including ossification, inflexibility, security, and quality of service, among others. To address these issues, there have been several initiatives to develop new network architectures, known as Future Network Architectures, Next Generation Network Architecture, and the Future Internet. These initiatives have taken either a clean-slate approach, in which the system is redesigned from scratch based on new core principles, or an evolutionary approach, in which a new architecture is developed that can accommodate the current Internet until it fully evolves into a Future Network. 

This project focuses on the design and development of real-time software and embedded systems for the implementation of concept-proving prototypes in a timely manner. The project approach is also extended to cover end-to-end user applications, including Internet connectivity over small embedded devices for global communication between machine-to-machine, user-to-user, and user-to-machine. The project covers three major topics: architectural design, embedded systems design, and real-time software design at the systems level. 

Publications: 

Dhananjay Singh*,"Developing an Architecture: Scalability, Mobility, Control, and Isolation on Future Internet Services", Second International Conference on Advances in Computing, Communications and Informatics (ICACCI-2013), Mysore, India, August 22-25, 2013. 

Project: Massive MIMO  Algorithm  

Duration: 03.2016 - 02.2018  extended to 02. 20 20

Industry: Joint Resesearch and Development project Korea-Pakistan Gov.  

Description: In this project, Massive MIMO (multiple-input-multiple-output) is a key technology in 5G mobile networks that utilizes a large number of antennas at the cell base station to form a huge antenna array. This technology has gained significant attention in the field of wireless communications in recent years due to its ability to improve channel capacity and spectrum utilization. However, one challenge in the implementation of massive MIMO systems is the high complexity of channel estimation algorithms. In order to address this issue, a new, low complexity channel estimation algorithm has been proposed that is based on the inherent sparsity of wireless communication channels. The algorithm uses the traditional discrete Fourier transform (DFT) to separate the channel taps from the noise space, allowing the channel estimation to be calculated only for a part of the channel taps and reducing computational complexity. The simulation results demonstrate that this proposed algorithm can achieve nearly minimum mean square error (MMSE) performance while maintaining low complexity. Additionally, the bit error rate and inter-cell interference results indicate that the proposed improved algorithm has better overall performance compared to conventional algorithms, making it a suitable choice for practical applications. 

Publications: 

• Imran Khan, Madhusudan Singh and Dhananjay Singh*, "Compressive Sensing-based Sparsity Adaptive Channel Estimation for 5G Massive MIMO Systems", Applied Sciences 8(5),754, May 2018. https://doi.org/10.3390/app8050754 [SCIE-2.67] 

• Imran Khan, Mohammad Haseeb Zafar, Mohammad Tariq Jan, Jaime Lloret, Mohammed Basheri, and Dhananjay Singh*, "Spectral and Energy Efficient Low-Overhead Up-link and Downlink Channel Estimation for 5G Massive MIMO Systems", Entropy, Vol. 20(2), 92, January 30, 2018, https://doi.org/10.3390/e20020092 [SCIE-2.49]

• Imran Khan, Dhananjay Singh*, "Efficient compressive sensing based sparse channel estimation for 5G massive MIMO systems", AEU - International Journal of Electronics and Communications, Elsevier, V 89, March 2018, Pages 181–190, https://doi.org/10.1016/j.aeue.2018.03.038 [SCIE-3.1]