대구한의대학교 향산도서관

상세정보

부가기능

Multi-antenna and mmWave Communications for Unmanned Aerial Vehicle Networks

상세 프로파일

상세정보
자료유형학위논문
서명/저자사항Multi-antenna and mmWave Communications for Unmanned Aerial Vehicle Networks.
개인저자Rupasinghe, R. A. Nadisanka Perera.
단체저자명North Carolina State University.
발행사항[S.l.]: North Carolina State University., 2019.
발행사항Ann Arbor: ProQuest Dissertations & Theses, 2019.
형태사항190 p.
기본자료 저록Dissertations Abstracts International 81-03B.
Dissertation Abstract International
ISBN9781085646284
학위논문주기Thesis (Ph.D.)--North Carolina State University, 2019.
일반주기 Source: Dissertations Abstracts International, Volume: 81-03, Section: B.
Advisor: Hughes, Brian
이용제한사항This item must not be sold to any third party vendors.
요약Unmanned aerial vehicle (UAV) base stations (BSs) can be a promising solution to provide connectivity and quality of service (QoS) guarantees during temporary events and after disasters. However, due to the limited power resources available on board of a UAV, it is of paramount importance to achieve spectral and energy efficient transmission in an UAV aided communication network. To that end, non-orthogonal multiple access (NOMA) can be a promising technology which can facilitate spectral and energy efficient communication in such a network. Hence, within the first three chapters of this dissertation, considering a UAV aided communication network serving users in a hotspot scenario, we introduce NOMA transmission in millimeter (mmWave) frequencies employing multi-antenna techniques. Subsequently, we investigate in detail how to enhance the performance of such a network especially focusing on realistic deployment aspects. Impact of UAV mobility and physical constraints of the UAV antenna array are important aspects to be considered for realistic deployments of a UAV aided communication network. Hence, first we study how the anticipated region to be covered on the ground is affected by UAV-BS hovering altitude especially when the physical vertical beamwidth of the beam generated by UAV-BS is limited. In particular, we show that the region where users are distributed (user region) may not be covered entirely at some UAV-BS hovering altitudes. During such situations we propose a beam scanning approach to identify the optimal area to be radiated within the user region. We accordingly propose a hybrid transmission strategy serving all or some of the NOMA users within the user region. The feedback overheads associated with NOMA are critical for realistic deployment of UAV aided communication networks with NOMA transmission. Hence, next we study the impact of limited feedback schemes and user ordering criteria for NOMA. Two limited feedback schemes are assumed as practical alternatives to full channel state information (CSI) feedback: 1) user distance, and 2) user angle. Based on these feedback schemes users are ordered with respect to their distance and angle (considering both Fej쨈er kernel and absolute angle) during NOMA formulation. Importantly, to the best of our knowledge, user angle as a feedback scheme for NOMA has not been studied in the literature before. The numerical results imply that users should be ordered based on a channel quality measure, on which users become more distinguishable. Physical layer security (PLS) can enable emerging wireless communication networks to maintain the confidentiality of the information of legitimate users. Hence, as the final aspect of our UAV-NOMA study, we shed light upon how to enhance the PLS in a UAVNOMA network. In particular, we introduce a protected zone based approach to achieve PLS against potential eavesdropper attacks in a UAV aided communication network. However, limited resource availability refrain protected zone being extended to cover the entire region where potential eavesdroppers can exist. To overcome this issue, we propose an approach to optimize the protected zone shape (for fixed area). Efficient placement of UAV-BSs is equally important to enhance the spectral and energy efficient communication. Hence, next we study the UAV placement problem and propose a hovering optimization technique which implicitly achieves user separation in the angular domain. In particular, we make use of UAV mobility and multi-antenna techniques within the proposed scheme. The numerical results show that under interference limited conditions, the proposed scheme provides sum rate performance comparative to linear zero force beamforming without the requirement of global CSI. Integrating a large antenna array to a UAV-BS may not be practically feasible at all frequencies due to the limited space availability on board of a UAV. This necessitates the importance of properly quantifying the amount of gains a UAV-BS can expect when equipped with moderately large antenna arrays. To address this issue, we derive an exact analytical expression for achievable rates in multi-cell multi-input-multi-output (MIMO) systems under pilot contamination considering eigen-beamforming (EBF), focusing on moderately large antenna array regime. We further evaluate achievable rates with regularized zero forcing (RZF) beamforming through extensive computer simulations. With mmWave transmission, it is possible to accommodate more antenna elements within a given form factor. This is especially useful for UAV-BSs since this allows to have a large antenna array within a limited space. However, with frequency division duplexing (FDD) transmission, the overheads associated with feedback-based channel acquisition can greatly compromise the achievable rates of FDD based massive MIMO systems. We address this issue in this dissertation by proposing a graph theoretic approach to reduce training over heads associated with FDD massive MIMO systems. The proposed graphtheoretic approach exploits knowledge of the angular spectra of the BS-mobile station (MS) channels to construct downlink (DL) training protocols with reduced overheads.
일반주제명Computer engineering.
Aerospace engineering.
Electrical engineering.
언어영어
바로가기URL : 이 자료의 원문은 한국교육학술정보원에서 제공합니다.

서평(리뷰)

  • 서평(리뷰)

태그

  • 태그

나의 태그

나의 태그 (0)

모든 이용자 태그

모든 이용자 태그 (0) 태그 목록형 보기 태그 구름형 보기
 
로그인폼