Channel Estimation for Millimeter Wave Massive MIMO Communication

The coronavirus (COVID-19) has greatly accelerated the demand and highlight the importance of developing high-speed networks. Due to the bandwidth shortage in microwave band, millimeter wave (mmWave) communication attracts significant attention to support future high-speed communication. Although mmWave frequency spectrum offers orders of magnitude greater spectrum, this spectrum suffers much greater attenuation compared to conventional cellular bands because of penetration losses, reflection and signal atmosphere. To overcome the high propagation lose in the mmWave band, large number of antennas can be adopted at both transmitter and receiver to provide large beamforming gains. Thanks to the short wavelength of mmWave signals, large arrays can be packed in to a small area. However, the large number of antennas makes fully digital beamforming unpractical considering the huge power consumption caused by devices operating at radio frequency (RF). To reduce the hardware costs and power consumptions, constrained architectures have been proposed. By connecting each RF chain to multiple antennas with phase shifts, hybrid architecture is able to reduce the hardware cost and power consumption with reduced number of RF chains. However, because of the constrained architecture and the large number of antennas, it is difficult to obtain the channel state information (CSI) which is of great importance for obtaining desirable beamforming gains. In this thesis, we investigate the channel estimation problem for mmWave massive multiple-input and multiple-output (MIMO) systems with hybrid architecture. Novel channel estimation algorithms with high accuracy and acceptable complexity are proposed.