검색 상세

Robust Localization under Anchor Uncertainty in 5G/6G mmWave Systems

초록/요약

통합 센싱 및 통신(ISAC: Integrated Sensing and Communication) 기술은 5G/6G무선네트워크의핵심기술로주목받고있다.특히대규모안테나배열 을기반으로하는대규모다중입출력(massive MIMO)과고주파대역(mmWave) 의 결합은 신호의 희소성(sparsity)을 활용한 고정밀 위치 추정(localization)을 가능하게 한다. 기존 거리 기반 중심의 위치 추정 방식과 달리 5G/6G ISAC 은 무선 채널로부터 고해상도의 각도 정보를 추출할 수 있으며, 전파의 강한 직진 특성을 활용하여 가시선(LOS: Line-of-Sight) 경로뿐만 아니라 비가시선 (NLOS: Non-Line-of-Sight)경로까지위치추정에활용한다. 또한 ISAC은 거리 및 각도 정보를 동시에 활용함으로써 기준점 역할을 수 행하는앵커(anchor)의수를크게줄일수있으며,단일 anchor환경에서도위치 추정이가능하다.이에따라지상고정형기지국(BS: Base Station)뿐만아니라, 필요 시 임의의 위치에 배치 가능한 무인기(UAV: Unmanned Aerial Vehicle) 나재구성지능형반사면(RIS: Reconfigurable Intelligent Surface)과같은동적 (anchor) 기반의 위치 추정 기술이 중요한 연구 주제로 부상하고 있다. 동적 anchor는 기하 구조를 능동적으로 최적화할 수 있어 위치 추정 성능을 향상시 킬 수 있으나, 동시에 anchor의 위치 및 방향 불확실성이라는 새로운 문제를 초래한다. 실제 환경에서는 센서 정밀도 한계, 플랫폼 진동, 보정 오차 등의 요인으 로 anchor의 상태 정보(위치 및 방향)를 정확히 파악하기 어렵다. 기존 연구는 대부분 anchor가정확히알려져있다고가정하거나, anchor불확실성을고려하 더라도위치오차또는방향오차중하나만을다루는한계가있었다.특히단일 동적 anchor기반의 3차원위치 ·방향동시추정문제는기존연구에서충분히 다뤄지지않았다. 이에 본 논문은 5G/6G ISAC 환경에서 anchor의 위치 및 방향 불확실성을 모두 고려한 강인한 단일 anchor 기반 Localization 기법을 제안한다. 먼저, 위 치 오차(3차원)와 방향 오차(2차원, 방위각 ·고도각)를 갖는 단일 지상 BS 환 경에서사용자 3차원위치를추정하는가중최소자승법(WLS: Weighted Least Squares)기반닫힌형해(closed-form)추정기를제안하고,크라메르-라오하한 (CRLB)을 통해 성능 한계를 분석한다. 이어, 위치 및 방향 오차가 모두 존재 하는 단일 UAV-BS 환경을 고려하여 사용자의 3차원 위치와 3차원 방향, 즉 6 차원측위를동시에추정하는기법을제안한다.이를위해 QUEST(QUaternion ESTimator) 기반 방향 추정과 WLS 기반 위치 및 동기화 추정 알고리즘을 설 계하고, CRLB를통해성능을이론적으로검증한다.시뮬레이션결과,제안기 법은 기존 기법 대비 anchor 상태 오차에 대해 높은 강인성을 보이며 우수한 Localization성능을달성함을확인하였다. 중심어 : mmWave, ISAC,위치추정,앵커불확실성

more

초록/요약

Integrated sensing and communication (ISAC) has emerged as a key enabling technology for future 5G/6G wireless networks. In particular, the combination of massive multiple-input multiple-output (MIMO) and millimeter-wave (mmWave) transmission enables high resolution wireless localization by exploiting the in- herent sparsity of high-frequency channels. Unlike conventional distance based localization approaches, 5G/6G ISAC systems can extract fine angle information from the wireless channel and utilize not only line-of-sight (LOS) paths but also non-line-of-sight (NLOS) signal components for localization. Since ISAC can jointly exploit distance and angular information, it signifi- cantly reduces the number of anchors required for localization and even enables single anchor localization. Consequently, dynamic anchor platforms such as un- manned aerial vehicles (UAVs) and reconfigurable intelligent surfaces (RIS), in addition to conventional static ground base stations (BSs), have recently drawn increasing attention in localization research. Although dynamic anchors can en- hance localization accuracy by adaptively forming favorable geometric configu- rations, they also introduce a new challenge due to uncertainties in anchor state information, including position and orientation errors. In practical systems, it is difficult to obtain accurate anchor state information because of sensor limitations, platform instability, and calibration errors. Most existing localization methods assume ideal anchor information, and even the few studies that consider anchor uncertainty address only either position or orienta- tion errors. In particular, robust three-dimensional (3D) localization and orienta- tion estimation using a single dynamic anchor under simultaneous position and orientation uncertainty remains an open problem. This dissertation proposes robust single anchor localization schemes for 5G/6G ISAC systems that jointly consider anchor position and orientation uncertain- ties. First, for a single ground BS with 3D position errors and two-dimensional (2D) orientation errors (azimuth and elevation), a closed-form estimator is devel- oped, and its performance limits are analyzed using the Cramér–Rao lower bound (CRLB). Then, a localization framework is proposed for a single UAV-BS sce- nario where both 3D position and 3D orientation uncertainties exist. The method jointly estimates the six-dimensional (6D) user state by combining the quaternion estimator for orientation estimation with WLS based position and synchroniza- tion refinement. The simulation results demonstrate that the proposed schemes significantly outperform existing benchmark methods. Keywords: mmWave, ISAC, Localization, Anchor Uncertainty

more

목차

1 Introduction 1
1.1 Research Motivation 1
1.2 Contributions 4
1.3 Thesis Overview 7
2 Background and Related Works 9
2.1 Wireless Localization in Cellular Networks 10
2.1.1 Wireless Localization before 4G 10
2.1.2 Evolution of Wireless Localization in 5G and 6G 13
2.2 High Precision mmWave Localization 16
2.2.1 MIMO-OFDM 16
2.2.2 Signal Model 19
2.2.3 Localization and Orientation Estimation Methods 22
2.2.4 Performance Analysis 26
2.2.5 Utilization of Diverse Anchors 28
2.3 Related Work 30
3 Robust Localization Algorithms Considering Single Anchor State Uncertainties 36
3.1 Robust Algorithm for 5G NLOS Localization and Synchronization with Single Anchor Uncertainties 37
3.1.1 Problem Formulation 38
3.1.2 Proposed Method 41
3.1.3 Performance Analysis 49
3.1.4 Simulation Results 52
3.1.5 Conclusion 56
3.2 UAV-BS Assisted User Localization Considering Bearing Observability in mmWave Systems 57
3.2.1 System Model 57
3.2.2 Unscented Kalman Filter for User Localization 60
3.2.3 Bearing Observability Maximization 64
3.2.4 Simulation Results 67
3.2.5 Conclusion 69
4 Robust Estimator for mmWave 6D Localization under Anchor State Uncertainties 71
4.1 System Model 72
4.1.1 System Geometry 73
4.1.2 Transmission Model 76
4.1.3 Mapping Between Channel and Geometric Parameters 77
4.2 Performance Bounds Analysis 79
4.2.1 FIM of Channel Parameters 79
4.2.2 FIM of Localization Parameters 80
4.2.3 Error Bounds for Parameters Estimation 81
4.3 Proposed Method 83
4.3.1 Orientation Estimation Using A2QUEST 83
4.3.2 Positioning Using WLS 89
4.4 Numerical Results 91
4.4.1 Simulation Setup 91
4.4.2 The Effect of Multipath on Channel Estimation 94
4.4.3 Impact of Anchor Uncertainty on PEB and OEB 95
4.4.4 Impact of Geometric Relationships on PEB and OEB 97
4.4.5 Performance Comparisons with Powers 99
4.4.6 Performance Comparisons with Anchor Uncertainty 101
4.4.7 Complexity Analysis 103
4.5 Conclusion 105
5 Dissertation Conclusions 106
References 109
Appendix 127
A Partial Derivatives for FIM 127
A.1 Channel Parameters 127
A.2 Location Parameters 128
B Weight Calculation 132
B.1 Weights for A2QUEST 132
B.2 Weights for WLS 135

more