Graduation Year

2018

Document Type

Dissertation

Degree

Ph.D.

Degree Name

Doctor of Philosophy (Ph.D.)

Degree Granting Department

Electrical Engineering

Major Professor

Nasir Ghani, Ph.D.

Committee Member

Ismail Uysal, Ph.D.

Committee Member

Mahshid Naeini, Ph.D.

Committee Member

Yao Liu, Ph.D.

Committee Member

Chadi Assi, Ph.D.

Keywords

5G, Beamforming, sparse channels, Meta-heuristic, link recovery

Abstract

Millimeter wave technologies present an appealing solution for increasing data throughputs as they provide abundant contiguous channel bandwidths as compared to conventional microwave networks. However, millimeter wave technologies suffer from severe propagation limitations and channel impairments such as atmospheric attenuation and absorption, path and penetration losses, and blockage sensitivity. Therefore, phased arrays and beamforming technologies are necessary to compensate for the degraded signal levels due to the aforementioned factors. Namely, base stations and mobile stations utilize directional transmission in the control- and data- plane for an enhanced channel capacity, which results in initial access challenges due to the absence of omni-directional transmission. Here the base station and mobile station are compelled to exhaustively search the entire spatial domain, i.e., in order to determine the best beamforming and combining vectors that yield the highest received signal level.

Overall, a wide range of studies have looked at the initial beam access challenges in millimeter wave networks, with most efforts focusing on iterative and exhaustive search procedures, as well as subarrays schemes and out-of-band beam access. However, these studies suffer from significant signaling overhead attributed to the prolonged beam scanning cycle. In particular, access times here are excessively high that exceed control plane latencies and coherence times. Furthermore, existing work suffer from high computational complexity, power consumption, energy inefficiency, as well as low directivities and high outage probabilities.

In light of the above, the contributions in this dissertation propose fast initial beam access schemes based upon novel meta-heuristic search schemes and beamforming architectures. These contributions include modified search procedures inspired by Nelder Mead, Luss-Jaakola, divide-and-conquer with Tabu search, generalized pattern search, and Hooke Jeeves methods.

Furthermore, efficient and highly-directive access schemes are also developed in this dissertation levering sidelobe emissions, grating lobes and Hamming codes. The overall performance of the proposed solutions here is extensively evaluated versus traditional access schemes and incorporating different channel and path loss models.

Finally, this dissertation addresses the problem of link sensitivity and blockage effects in millimeter wave networks, a subsequent stage to beam access and link association. Nevertheless, a novel link recovery procedure is proposed here that features instantaneous link-recovery and high signal levels.

Download

Share

COinS