TY - GEN
T1 - Thresholded Wirtinger Flow for Fast Millimeter Wave Beam Alignment
AU - Gan, Chao
AU - Yang, Jing
AU - Shen, Cong
N1 - Funding Information:
The work of C. Gan and J. Yang was partially supported by the US National Science Foundation (NSF) under Grants ECCS-1650299, CNS-1956276. The work of C. Shen was partially supported by the US National Science Foundation (NSF) under Grant CNS-2002902.
Publisher Copyright:
© 2020 IEEE.
PY - 2020/11/1
Y1 - 2020/11/1
N2 - The objective of this paper is to estimate millimeter wave (mmWave) channels based on the magnitude of the received complex measurements. This is motivated by the observation that in practice, a random carrier frequency offset usually occurs for each transmission and corrupts the phase of the measurement. The problem is cast into a non-convex optimization framework, and solved through a thresholded Wirtinger flow (TWF) based approach. Due to the complex signals involved in the problem, and the constraints imposed by phase shifters, existing design and analysis of TWF are not directly applicable. To overcome these challenges, a structured design of the combiner (precoder) is investigated, which naturally leads to a novel initialization scheme for the TWF algorithm. Simulation results indicate that the proposed approach can successfully recover multiple paths in an mmWave channel and the required number of measurements scales mildly with the dimension of the array size.
AB - The objective of this paper is to estimate millimeter wave (mmWave) channels based on the magnitude of the received complex measurements. This is motivated by the observation that in practice, a random carrier frequency offset usually occurs for each transmission and corrupts the phase of the measurement. The problem is cast into a non-convex optimization framework, and solved through a thresholded Wirtinger flow (TWF) based approach. Due to the complex signals involved in the problem, and the constraints imposed by phase shifters, existing design and analysis of TWF are not directly applicable. To overcome these challenges, a structured design of the combiner (precoder) is investigated, which naturally leads to a novel initialization scheme for the TWF algorithm. Simulation results indicate that the proposed approach can successfully recover multiple paths in an mmWave channel and the required number of measurements scales mildly with the dimension of the array size.
UR - http://www.scopus.com/inward/record.url?scp=85107725654&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85107725654&partnerID=8YFLogxK
U2 - 10.1109/IEEECONF51394.2020.9443380
DO - 10.1109/IEEECONF51394.2020.9443380
M3 - Conference contribution
AN - SCOPUS:85107725654
T3 - Conference Record - Asilomar Conference on Signals, Systems and Computers
SP - 32
EP - 36
BT - Conference Record of the 54th Asilomar Conference on Signals, Systems and Computers, ACSSC 2020
A2 - Matthews, Michael B.
PB - IEEE Computer Society
T2 - 54th Asilomar Conference on Signals, Systems and Computers, ACSSC 2020
Y2 - 1 November 2020 through 5 November 2020
ER -