Mobile Millimeter Wave Channel Acquisition, Tracking, and Abrupt Change Detection

Chuang Zhang,Dongning Guo,Pingyi Fan

DOI: https://doi.org/10.48550/arxiv.1610.09626

2016-01-01

Abstract:Millimeter wave provides a promising approach for meeting the ever-growing traffic demand in next generation wireless networks. It is crucial to obtain relatively accurate channel state information so that beamforming/combining can be performed to compensate for severe path loss in this band. In contrast to lower frequencies, a typical mobile millimeter wave channel consists of a few dominant paths. It is generally sufficient to estimate the path gains, angles of departure (AoD), and angles of arrival (AoA) of those paths. In this paper, multiple transmit and receive antennas and beamforming with a single baseband processing chain are assumed. We propose a framework for estimating millimeter wave channels with intermittent abrupt changes (e.g., blockage or emergence of dominant paths) and slow variations of AoDs and AoAs. The solution consists of three components: tracking of the slow channel variations, detection of abrupt changes, followed by (re-)acquisition of channel (and back to the tracking stage). For acquisition, we formulate a least squares problem and find its solution based on the Levenberg-Marquardt algorithm. To track slow variations of AoDs and AoAs, we propose a new approach using Kalman filtering. Finally, an algorithm based on a likelihood test is devised for detecting abrupt changes. Simulation results show that, with moderate signal-to-noise ratios, the proposed scheme can achieve more than 8 dB higher estimation accuracy than several other methods using the same number of pilots.