Speaker: Han Yan
Affiliation: Ph.D. Candidate
Via Zoom: https://ucla.zoom.us/j/192134955
Abstract: Millimeter-wave (mmW) communications is the key technology in 5G and beyond mobile networks. The vast spectrum in mmW band enables unprecedented data rate which enhances many applications such as automated intelligence, autonomous vehicles, and virtual/augmented reality. The increased propagation loss in mmW band requires massive antenna arrays to meet link budget. However, practical realizations of mmW communication face challenges in the array hardware design and physical layer signal processing design. We present cross-disciplinary research progress in addressing these challenges.
In the first part, we study a physical layer procedure in mmW mobile networks: directional link establishment via initial access and beam training. A novel signal processing algorithm is proposed to achieve a joint initial cell discovery, synchronization, and fine beam alignment. Discovery rate and theoretical performance bound of beam training accuracy is derived. Using simulation with 5G compliant frame structure and realistic 28GHz 3D channel model, we show that the proposed method provides order of magnitude access latency and overhead saving as compared to existing solutions. A proof-of-concept in mmW testbed and implementation concerns are then discussed.
While 5G mmW utilizes analog phased array, more advanced array architecture will certainly be adopted in the evolution of 5G. However, there is a lack of understanding in performance and power consumption among candidates. In the second part, we study the power efficient mmWave array architectures. We model hardware blocks of three array architectures. Their spectral efficiency and power consumption are compared. Our analysis explains the reason why fully-digital arrays would be the optimal solution even in mmW band.
Biography: Han Yan is a Ph.D. candidate at the Cognitive Reconfigurable Embedded Systems (CORES) Lab in the Electrical & Computer Engineering Department at UCLA. He received the B.E. degree from Zhejiang University, Hangzhou, China, in 2013, and the M.S. degree in Electrical Engineering from UCLA in 2015. He has broad research interests in signal processing and communication system design for millimeter-wave mobile networks, cooperative unmanned aerial vehicles networks, and dynamic spectrum sharing radios. He has 15 publications (10 as the first author) in conferences and journals. His research resulted in grants funded by governmental agencies including NSF, SRC, and DARPA. He was a recipient of the UCLA Dissertation Year Fellowship in 2018 and Qualcomm Innovation Fellowship in 2019.
Date(s) - May 22, 2020
9:30 am - 11:30 am