1 Research work
Feedback Reduction using Type II CSI in 5G NR MIMO Systems In a cellular system, the uplink channel carries the feedback bits indicating channel state information (CSI) in order to enable an appropriate modulation and coding scheme (MCS) along with scheduling data to multiple users in the downlink. This means there is a trade-o between the feedback information and the data the uplink channel can carry. Orthogonal frequency division multiplexing (OFDM) based fth-generation (5G) cellular systems use downlink scheduling, adaptive modulation and coding (AMC) and multiinput multi-output (MIMO) to increase spectral eciency. Reduced feedback schemes are a crucial component of such systems. These schemes ensure that the overhead of feeding back the CSI does not overwhelm the limited-bandwidth uplink feedback channel. In this work, rst, we try to understand and develop 3GPP aligned MIMO Type II CSI feedback scheme as specied in the technical specication of 5G NR. While Type II CSI generation is specied in the standard, some details related to it, such as how to generate the wideband and subband-specic components, are left open for interpretation. We propose to develop methods that address these open gaps and ensure feedback generation that is as close to perfect CSI as possible. Second, we study QuaDRiGa channel simulator to generate MIMO channel traces for single-user (SU-MIMO) scenario and corresponding Type II traces using algorithm developed in the rst step. My role in the project was to implement the Type II CSI generation algorithm for SUMIMO single and multi-rank transmission by incorporating dual-polarized uniform planar array (UPA) antennas at the base station. To evaluate the eectiveness of regenerated dual-stage precoder using Type II CSI, we consider reconstruction error and rate as the two performance metrics with rate being the key performance indicator (KPI). Cell-wide simulation results shows that the algorithm could generate the Type II precoder close to the ideal precoder.
2 Contribution to CNI
Networking Module Hands-on Laboratory I had got the opportunity to get involved with CNI and a few other colleagues in developing a networking module based on congestion control, a part of the hands-on laboratory course. This lab course was designed for the master’s students in the ECE department at IISc. The work was primarily based on creating the content and structuring the module for the rst-year students.