Presentation Title

MATLAB Simulation Plan for 5G Beamforming

Faculty Mentor

Thomas Ketseoglou

Start Date

18-11-2017 10:00 AM

End Date

18-11-2017 11:00 AM

Location

BSC-Ursa Minor 100

Session

Poster 1

Type of Presentation

Poster

Subject Area

engineering_computer_science

Abstract

In fifth generation of wireless communications (5G), a large number of base station antennas is anticipated resulting in massive Multiple-Input Multiple-Output (MIMO) systems. Within the context of massive MIMO, beamforming techniques offer the possibility of dramatically increasing the downlink throughput, by employing ``the best path’’ possible for transmission of high rate data. In this undergraduate research work, we address the interesting problem of a comparison of 5G beamformers with Quadrature Phase Shift Keying (QPSK) Modulation and ones with Quadrature Amplitude Modulation (QAM). Although the latter ones can offer dramatically higher throughput at high signal-to-noise ratio (SNR), their precoding complexity is exponentially higher than the one required by QPSK, therefore the QPSK ones can be much simpler, offering implementation advantages. In addition, QPSK can offer better leverage of antennas by using 3X3 groups in the precoding process. Thus, the issue of comparing QPSK with QAM becomes an interesting, relevant one. The presentation will cover the methodology employed in this study which comprises: a) A plan for simulating the performance of the two beamforming systems, b) A trade-off analysis of the performance, and c) The impact of the future simulation results to the downlink beamformer design in 5G. Furthermore, future work will also be discussed.

Summary of research results to be presented

The problem of beamforming design is a central one in 5G. In this presentation, we will show how optimized beamformers are designed for 5G wireless, based on MATLAB code and how these can be used to develop a simulation comparison of QPSK and QAM employing the optimized precoders, toward efficient system design. This work is part of our ongoing work on 5G beamforming, pioneered by Dr. Ketseoglou and Dr. Ayanoglu, and published in, e.g., IEEE Transactions on Communications.

This document is currently not available here.

Share

COinS
 
Nov 18th, 10:00 AM Nov 18th, 11:00 AM

MATLAB Simulation Plan for 5G Beamforming

BSC-Ursa Minor 100

In fifth generation of wireless communications (5G), a large number of base station antennas is anticipated resulting in massive Multiple-Input Multiple-Output (MIMO) systems. Within the context of massive MIMO, beamforming techniques offer the possibility of dramatically increasing the downlink throughput, by employing ``the best path’’ possible for transmission of high rate data. In this undergraduate research work, we address the interesting problem of a comparison of 5G beamformers with Quadrature Phase Shift Keying (QPSK) Modulation and ones with Quadrature Amplitude Modulation (QAM). Although the latter ones can offer dramatically higher throughput at high signal-to-noise ratio (SNR), their precoding complexity is exponentially higher than the one required by QPSK, therefore the QPSK ones can be much simpler, offering implementation advantages. In addition, QPSK can offer better leverage of antennas by using 3X3 groups in the precoding process. Thus, the issue of comparing QPSK with QAM becomes an interesting, relevant one. The presentation will cover the methodology employed in this study which comprises: a) A plan for simulating the performance of the two beamforming systems, b) A trade-off analysis of the performance, and c) The impact of the future simulation results to the downlink beamformer design in 5G. Furthermore, future work will also be discussed.