Millimeter wave MIMO communications : high-resolution angle acquisition and low-resolution time-frequency synchronization




Zhu, Dalin

Journal Title

Journal ISSN

Volume Title



Knowledge of the propagation channel is critical to exploit the full benefit of multiple-input multiple-output (MIMO) techniques in millimeter wave (mmWave) cellular systems. Obtaining accurate channel state information in mmWave systems, however, is challenging due to high estimation overhead, high computational complexity and on-grid setting. It is also desirable to reduce the analog-to-digital converters (ADCs) resolution at mmWave frequencies to reduce power consumption and implementation costs. The use of low-precision ADCs, though, brings new design challenges to practical cellular networks.

In the first part of this dissertation, we develop several new methods to estimate and track the mmWave channel's angle-of-departure and angle-of-arrival with high accuracy and low overhead. The key ingredient of the proposed strategies is custom designed beam pairs, from which there exists an invertible function of the angle to be estimated. We further extend the proposed algorithms to dual-polarized MIMO in wideband channels, and angle tracking design for fast-varying environments. We derive analytical angle estimation error performance of the proposed methods in single-path channels. We also use numerical examples to characterize the robustness of the proposed approaches to various transceiver settings and channel conditions.

In the second part of this dissertation, we focus on improving the low-resolution time-frequency synchronization performance for mmWave cellular systems. In our system model, the base station uses analog beams to send the synchronization signal with infinite-resolution digital-to-analog converters (DACs). The user equipment employs a fully digital front end to detect the synchronization signal with low-resolution ADCs. For low-resolution timing synchronization, we propose a new multi-beam probing based strategy, targeting at maximizing the minimum received synchronization signal-to-quantization-plus-noise ratio among all serving users. Regarding low-resolution frequency synchronization, we construct new sequences for carrier frequency offset (CFO) estimation and compensation. We use both analytical and numerical examples to show that the proposed sequences and the corresponding metrics used for retrieving the CFOs are robust to the quantization distortion.


LCSH Subject Headings