Thursday, December 05, 2019
Synchronization, Impairment Estimation and Interference Alignment for Wireless Communication Systems
Wireless communications has been under intensive research and development for decades and it has emerged as one of the largest sectors of the telecommunications industry. Wireless communication systems utilize the wireless media to perform over-the-air communications. Unfortunately, the wireless media are not ideal, which leads to a number of issues and finally causes the failure of communications. The factors that may impact the quality of wireless transmissions come from all aspects, including the medial imperfection, interfering environment, mismatch of transceivers, and so on. To solve these problems and improve the quality of service (QoS), research studies are conducted on relevant topics including synchronization techniques, impairment estimation theory and techniques, and interference alignment techniques. In this thesis, interfering systems and imperfect transmissions are investigated, interference alignment and synchronization techniques are proposed, as well as implementation of part of the proposed techniques on software defined radio platforms. We firstly present a dual link algorithm and the design of an field-programmable gate array (FPGA) prototype targeting software defined radio (SDR) platforms, to align and manage the interference of multiple-input and multiple-output (MIMO) networks. To make the proposed FPGA prototype of our dual link algorithm less hardware-consuming and compatible with larger number of antennas, we also propose a hardware-efficient singular implementation for massive MIMO. Then, we propose maximum-likelihood (ML) based synchronization approaches for coarse frequency, fine frequency and timing synchronizations, jointly with minimum mean square error (MMSE) channel estimations for MIMO systems. In addition, some of the proposed synchronization techniques are implemented on FPGA or FPGA based SDR platforms as prototype. Finally, by taking the advantage of machine learning techniques into consideration, we propose a neural network based estimator to perform coarse frequency offset estimations for single carrier (SC) systems and integer frequency offset estimations for orthogonal frequency division multiplexing (OFDM) system.
Professor Xinming Huang
ECE Department, WPI
Professor Kaveh Pahlavan
ECE Department, WPI
Professor Youjian Liu
Electrical, Computer & Energy Dept.
University of Colorado, Boulder