Virginia Tech :
Stochastic Geometry-based Uplink Analysis of Massive MIMO Systems with Fractional Pilot Reuse: In this work, using Johnson-Mehl cells, we modeled and analyzed a massive MIMO system that incorporates fractional pilot reuse to mitigate the effect of pilot contamination for the cell-edge users. We presented a pilot partitioning rule that improves the cell-edge user performance with minimal impact on average cell-center users throughput.
Pilot Allocation Schemes for Cell-Free Massive MIMO systems: In this work, we proposed a distributed pilot allocation scheme to mitigate the effect of pilot contamination among a set of co-pilot users in a cell-free massive MIMO system. The proposed distributed algorithm is inspired by the random sequential adsorption (RSA) process in the stochastic geometry literature. We further proposed two centralized pilot allocation schemes to benchmark the performance of the RSA-inspired scheme.
Stochastic Geometry-based analysis of Cell-Free Massive MIMO System: In this work, using tools from stochastic geometry we analyzed the downlink performance of a cell-free massive MIMO system with limited fronthaul capacity. First, we provide design guidelines for a finite network where each AP serves all the users in the network. Next, we also provide useful system insights for a user-centric cell-free system.
Multilayer Random Sequential Adsorption: Inspired by our pilot allocation scheme of the cell-free massive MIMO system, we proposed a new variant of the multilayer RSA process. For this variant, we derived the density result for
Stochastic geometry-based modeling and analysis of the performance of licensed and unlicensed operators that coexist in the 3.5-GHz shared CBRS band.
IIT Kharagpur, India:
During my stay at IIT Kharagpur, I have carried out the following research activities in the capacity of an MS student and as a research consultant for the project titled "Development of Interference mitigation methods for
Next-Generation Mobile Communication Networks". (PI: Dr. Suvra Sekhar Das)
Performance analysis of VoIP and Best Effort traffic under LTE – A specified diversity MIMO modes.
Developed a down-link framework for a narrow-band MIMO multi-cellular system using the sum of log-normal approach. The framework is valid for a wide and practical range of shadowing standard deviation, number of antennas, and activity of the interfering base stations. Moreover, using the proposed framework, we have analyzed the performance of different MIMO schemes in terms of average ergodic spectral efficiency, average outage probability, and average outage spectral efficiency.
Proposed a difference of convex programming-based power allocation scheme for the downlink of non-cooperative multi-cellular OFDM-Non Orthogonal Multiple Access (NOMA) systems. The proposed power allocation algorithm along with a greedy user selection scheme is found to achieve near-optimal performance in terms of the geometric average user throughput of the system.
Proposed a centralized resource allocation algorithm for a cooperative multi-cellular OFDM-NOMA system. The proposed algorithm is iterative in nature that provides considerable improvement in the cell edge user throughput with a marginal reduction in the mean cell throughput compared to a non-cooperative multi-cellular OFDM-NOMA system.
Developed an LTE-A link-level simulator with the incorporation of different MIMO modes along with convolutional and turbo codes for error correction.
Performance evaluation of different static coordination schemes (e.g. FFR, SFR), and dynamic coordination schemes (centralized, distributed) for Fourth-generation cellular systems in the presence of both channel aware schedulers (e.g. proportional fair) and channel blind scheduler (e.g. round-robin). The performance comparison is done in terms of cell edge user throughput, mean cell throughput, and Jain's fairness index.
Undergraduate Project:
Development of a simulation testbed for analyzing packet data performance of WCDMA network in the presence of both LDPC codes (as PHY layer FEC) and Raptor Codes (as application layer FEC) in the absence of HARQ.
Evaluation of packet data performance in terms of packet error rate, throughput, and delay in the WCDMA network in the presence of LDPC codes, soft handoff, and HARQ.