Multiple Antenna Signal Processing Techniques for Millimeter Wave Communications
Abstract
Mobile devices operating at millimeter wave (mmWave) frequencies are expected to comprise an integral part of fifth generation (5G) communication systems to meet increasing data rate demands. Massive multiple-input multiple-output (MIMO) and advanced signal processing techniques are required to overcome the harsh propagation environment in this spectrum. We focus on two aspects of mmWave communication systems. First, the large number of antennas creates a challenge in aligning and tracking highly directional, narrow beams. Algorithms which rapidly adapt to the changing mobile environment are required. We propose a novel beam alignment and tracking algorithm for time-varying, sparse mmWave channels using multi-armed bandit beam selection. We show our algorithm has a more rapid initial beam alignment compared to other beam selection policies and, for dynamic channel support, long-term beamforming gain commensurate to omni-directional channel training. Simulation results are accomplished using idealized and realistic mmWave channel models. Second, massive MIMO systems can generate potentially prohibitive amounts of data due to the large numbers of antennas. With modern parallel, low-rate analogto-digital converters (ADCs), the bottleneck is often not in the quantization of the received signals but, rather, in the processing of the digitized bits. Therefore, we develop an adaptive algorithm for down-selecting the digital output data to meet some required output data rate threshold while simultaneously maximizing the information between the transmitted signal and the selected output.
Degree
Ph.D.
Advisors
Michelusi, Purdue University.
Subject Area
Aerospace engineering|Artificial intelligence|Electrical engineering|Electromagnetics|Optics|Physics
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.