Wireless Indoor Localization.pdf

发布时间：2022-06-10 发布人：admin 分类：说明书资料大小：8.84M 资料格式：pdf 举报版权申诉

第1页 / 共225页

weixin_38290023-11207350-Wireless Indoor Localization.pdf-第2页.png

第2页 / 共225页

weixin_38290023-11207350-Wireless Indoor Localization.pdf-第3页.png

第3页 / 共225页

weixin_38290023-11207350-Wireless Indoor Localization.pdf-第4页.png

第4页 / 共225页

weixin_38290023-11207350-Wireless Indoor Localization.pdf-第5页.png

第5页 / 共225页

weixin_38290023-11207350-Wireless Indoor Localization.pdf-第6页.png

第6页 / 共225页

weixin_38290023-11207350-Wireless Indoor Localization.pdf-第7页.png

第7页 / 共225页

weixin_38290023-11207350-Wireless Indoor Localization.pdf-第8页.png

第8页 / 共225页

Preface

Organization of the Book

Anticipated Audience

Acknowledgments

Contents

Acronyms

Part I Getting Started

1 Background and Overview

1.1 Wireless Indoor Localization

1.2 State-of-the-Art Approaches

1.2.1 Infrastructure

1.2.2 General Architecture

1.2.3 Historical Stages

1.3 WiFi Fingerprint-Based Approach

1.3.1 General Frameworks

1.3.2 Challenges for Ubiquitous Applications

1.4 Book Organization

References

2 Mobile Crowdsourcing and Inertial Sensing

2.1 Harnessing Crowdsourcing in Mobile Computing

2.2 Embracing Mobility via Inertial Sensing

2.2.1 Sensors Types

2.2.2 Mobility Information

2.2.2.1 Displacement Information

2.2.2.2 Direction Information

2.2.2.3 Integrated Information

References

Part II Boosting Deployment: Making It Available

3 Radio Map Construction Without Site Survey

3.1 Introduction

3.2 Related Work

3.3 Overview

3.3.1 Data Collection

3.3.2 System Architecture

3.4 Stress-Free Floor Plan

3.5 Fingerprint Space

3.5.1 Fingerprint Collection

3.5.2 Pre-processing

3.5.3 Fingerprint Space Construction

3.6 Mapping

3.6.1 Feature Extraction

3.6.1.1 Corridor Recognition

3.6.1.2 Room Recognition

3.6.1.3 Reference Point Mapping

3.6.2 Space Transformation

3.6.2.1 Floor-Level Transformation

3.6.2.2 Room-Level Transformation

3.7 Experiments

3.7.1 Experiment Design

3.7.2 Performance Evaluation

3.7.2.1 Fingerprint Space Generation

3.7.2.2 Mapping Performance

3.7.2.3 Localization Error

3.8 Conclusion

References

4 Building Tomography: Automatic Floor Plan Generation

4.1 Introduction

4.2 Related Works

4.3 System Overview

4.4 Trace Collection

4.4.1 User Data Collection

4.4.2 Dead-Reckoning

4.5 Trace Realization

4.5.1 Entrance Discovery

4.5.2 Reference Point Extraction

4.5.3 Reference Point Matching

4.5.4 Drift Fixing

4.5.5 Multi-floor Case

4.6 Map Generation

4.6.1 Space Regionalization

4.6.2 Functionality Recognition

4.7 Experiments

4.7.1 Experiment Setup

4.7.2 Reference Point Matching Performance

4.7.3 Regionalization Performance

4.7.4 Functionality Recognition

4.8 Conclusion

References

Part III Facilitating Maintenance: Making It Sustainable

5 Adaptive Radio Map Updating

5.1 Introduction

5.2 Related Works

5.3 Preliminaries and Measurements

5.3.1 Measurements of RSS Dynamics

5.3.2 Radio Map Updating with Reference Points

5.4 Overview

5.5 Method Design

5.5.1 Pin Data Collection

5.5.1.1 Pin Data Specification

5.5.1.2 Mobility Monitoring

5.5.2 Reference Point Estimation

5.5.3 Map Updating

5.5.3.1 Learning the Regression Model

5.5.3.2 Updating the Radio Map

5.6 Implementations and Evaluation

5.6.1 Experimental Methodology

5.6.2 Performance Evaluation

5.6.2.1 Performance of Trajectory Matching

5.6.2.2 Performance of Map Updating

5.6.2.3 Localization Performance

5.7 Discussions and Limitations

5.8 Conclusions

References

6 Self-Deployable Peer-to-Peer Navigation

6.1 Introduction

6.2 Related Works

6.3 Motivations and Challenges

6.3.1 Model Description

6.3.2 Usage Examples

6.3.3 Design Challenges

6.4 Overview

6.5 Fingeram Specification

6.5.1 Definition and Generation

6.5.2 Radio Features

6.5.3 Visual Features

6.6 ppNav Design

6.6.1 Trace Generation

6.6.2 Walking Progress Estimation

6.6.3 Deviation Detection

6.6.4 Path Locking-On

6.7 Implementation and Experiments

6.7.1 Experimental Setup

6.7.2 Performance Evaluation

6.7.2.1 Performance of Motion Event Detection

6.7.2.2 Performance of Trace Synchronization

6.7.2.3 Performance of Deviation Detection

6.7.2.4 Performance of Lock on Direction

6.8 Discussion

6.9 Conclusion

References

Part IV Enhancing Accuracy: Making It Reliable

7 Exploiting Spatial Awareness via Fingerprint Spatial Gradient

7.1 Introduction

7.2 Related Works

7.3 ViVi Design

7.4 Understanding Fingerprint Spatial Gradient

7.4.1 Limitations of Existing RSS Fingerprints

7.4.2 Fingerprint Spatial Gradient

7.4.2.1 FSG Specification

7.4.2.2 FSG Advantages

7.5 Exploiting Fingerprint Spatial Gradient

7.5.1 Profiling Fingerprint Spatial Gradient

7.5.1.1 Profiling a Reference Location

7.5.1.2 Profiling a Query Fingerprint

7.5.2 Comparing Fingerprint Spatial Gradient

7.5.2.1 Cosine Similarity

7.5.2.2 Turning Function Distance

7.5.2.3 Discrete Fréchet Distance

7.6 Implementations and Evaluation

7.6.1 Experimental Methodology

7.6.1.1 Experimental Scenarios

7.6.1.2 Comparative Methods

7.6.2 Performance Evaluation

7.6.2.1 Performance in Different Areas

7.6.2.2 Performance Comparison

7.6.2.3 Impacts of Device Diversity

7.6.2.4 Impacts of RSS Fingerprint Similarity (RFS) Measures

7.6.2.5 Impacts of Neighbour Number

7.6.2.6 Impacts of AP Number

7.6.2.7 Impacts of k

7.6.2.8 System Latency

7.7 Conclusions

References

8 Enhancing WiFi Fingerprinting with Visual Clues

8.1 Introduction

8.2 Related Works

8.3 Measurements and Observations

8.3.1 Measurements

8.3.2 Observations

8.4 System Overview

8.4.1 Working Flow from User's Perspective

8.4.2 Working Flow from Server's Perspective

8.5 Image Processing Pipeline

8.5.1 Image Matching

8.5.2 Relative Geometric Constraint Extraction

8.5.3 Supporting Photo Set Selection

8.6 WiFi Processing Pipeline

8.6.1 Device Calibration

8.6.2 WiFi Fingerprinting

8.7 Joint Location Estimation

8.7.1 Problem Formulation

8.7.2 Problem Hardness

8.7.3 Two-Tier Shape Matching Heuristic

8.8 Evaluation

8.8.1 Experimental Methodology

8.8.2 Micro Benchmarks

8.8.2.1 Effectiveness of Image Matching

8.8.2.2 Effectiveness of Geometric Constraint Extraction

8.8.2.3 Effectiveness of Photo Selection

8.8.2.4 Effectiveness of Device Calibration

8.8.3 Overall System Performance

8.8.3.1 Overall Localization Accuracy

8.8.3.2 System Latency

8.8.3.3 Energy Overhead

8.8.4 Impacting Factors

8.8.4.1 Impact of Distance to POIs

8.8.4.2 Impact of Quantity of Photos

8.8.4.3 Impact of Illumination

8.8.4.4 Impact of Photo Resolution

8.8.4.5 Impact of Obstacles in Photos

8.9 Discussion

8.10 Conclusion

References

9 Mitigating Large Errors in Practice

9.1 Introduction

9.2 Related Works

9.3 Preliminary and Measurements

9.3.1 Problem Statement

9.3.2 Observations

9.4 Design Methodology

9.4.1 Phantom Fingerprints

9.4.2 Robust Fingerprinting

9.4.3 Discriminatory Policy

9.4.4 Localization

9.5 Experiments and Evaluation

9.5.1 Experimental Methodology

9.5.2 Performance Evaluation

9.5.2.1 Overall Performance

9.5.2.2 Effect of Individual Modules

9.6 Conclusions

References

Part V Conclusions

10 Research Summary and Future Directions

10.1 Research Summary

10.2 Future Directions

Chenshu Wu Zheng Yang Yunhao Liu Wireless Indoor Localization A Crowdsourcing Approach

Wireless Indoor Localization

Chenshu Wu • Zheng Yang Yunhao Liu Wireless Indoor Localization A Crowdsourcing Approach 123

Chenshu Wu University of Maryland College Park, Maryland, USA Yunhao Liu School of Software Tsinghua University Beijing, China Zheng Yang School of Software Tsinghua University Beijing, China ISBN 978-981-13-0355-5 https://doi.org/10.1007/978-981-13-0356-2 ISBN 978-981-13-0356-2 (eBook) Library of Congress Control Number: 2018948837 © Springer Nature Singapore Pte Ltd. 2018 This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is concerned, speciﬁcally the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microﬁlms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a speciﬁc statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, express or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional afﬁliations. This Springer imprint is published by the registered company Springer Nature Singapore Pte Ltd. The registered company address is: 152 Beach Road, #21-01/04 Gateway East, Singapore 189721, Singapore

Preface Location-based technology surely ranks as one of the most important breakthrough for modern life. Localization techniques underlie the foundation of Internet of Things, while location information contributes one type of the most critical big data nowadays. With the proliferation of mobile computing, location-based services have shaped and enabled a wide range of applications from vehicle navigation to logistical supply chain management and to personal location sharing. While GPS dominates outdoor localization and becomes an essential element of the global information infrastructure like the Internet, indoor localization also attracts people’s attention. Indoor location holds important values for various applications, including resource and infrastructure management, smart home and smart building monitoring, retails and sports analytics, mall navigation, virtual reality, etc. The past decades have witnessed the fast conceptualization and development of wireless indoor localization. Among numerous technologies, WiFi ﬁngerprint-based approach has become one of the most popular solutions that attracts immense efforts from both academic and industrial communities. Not only a number of start-ups, but also grand corporations including Google, Apple, Cisco, QualComm, Huawei, etc., have invested in and have been developing indoor localization products. This book aims to provide a comprehensive and in-depth understanding of wireless indoor localization for ubiquitous applications. Speciﬁcally, focusing on WiFi ﬁngerprint based localization via crowdsourcing, the book follows a top- down view and involves the three most important issues of indoor localization: deployment, maintenance, and accuracy. Through extensively reviewing the state- of-the-art literatures, it presents the latest advances in crowdsourcing-enabled WiFi localization. Crowdsourcing is a recently invented and increasingly hot topic in not only research areas but also enterprise business. It has been widely applied in indoor localization to address several critical bottlenecks. As to WiFi ﬁngerprint based localization, crowdsourcing helps with the issues including signiﬁcant deployment costs, lack of indoor ﬂoorplans, and limited location accuracy. Despite a large number of research outputs, it lacks a representative book that provides an extensive view of crowdsourcing-based indoor localization. This book turns out to be the v

vi Preface ﬁrst of its kind, as far as we are aware of, which introduces WiFi ﬁngerprint based localization from a perspective of crowdsourcing. Hence, we hope the reported latest advances can promote the development of this research area and contribute to the interdisciplinary research areas as well. Organization of the Book The book begins with an introductory Part I. We ﬁrstly present the background of location-aware applications, current wireless localization problems, and represen- tative localization systems in Chap. 1. The preliminary introduction in Chap. 2 to the concepts and techniques of crowdsourcing, mobile sensing, etc., would guide the readers to understand. The subsequent parts focus on tackling the critical challenges of WiFi-based localization for ubiquitous applications, each from a distinctive aspect of deployment costs, maintenance overhead, and location errors. In these parts, we present the ideas, methods, and systems for implementing the crowdsourcing approach in localization. Speciﬁcally, rapid deployment is the ﬁrst step to make indoor localization available, which is extensively investigated in Part II. In addition, it is also important to maintain consistent quality of service over long-term running, which is carefully addressed in Part III. Moreover, accuracy acts as the most critical metric to reliable and usable indoor localization, which worths in-depth exploration in Part IV. Finally, we make a summary of this book and discuss several future directions in Part V. Part II to Part IV consist of the main content of this book, and are further elaborated into seven chapters as follows: Part II Boosting deployment: making it available In this part, we consider two deployment issues of WiFi-based localization: radio map construction (Chap. 3) and digital ﬂoorplan generation (Chap. 4). Harnessing the power of mobile crowdsourcing, we present techniques that automate these two procedures, which can only be completed by manual efforts previously, and therefore rapidly boost the deployment for WiFi-based location systems in practice. Part III Facilitating maintenance: making it sustainable WiFi ﬁngerprints are sensitive to temporal environmental dynamics and vary over time. To tackle this problem, in Chap. 5, we seek for adaptive ﬁngerprint updating techniques to adapt collected ﬁngerprints to environmental changes and to guarantee the performance of deployed location services over time. Furthermore, we introduce a self-deployable approach for indoor navigation, which can work in a peer-to-peer manner with no prior installed location services. Part IV Enhancing accuracy: making it reliable Traditional WiFi-based localization suffers from large location errors. Based on in-depth understanding of root causes of its location errors, we present novel approaches to improve accuracy by embracing spatial awareness in the form of ﬁngerprint spatial gradient in Chap. 7. We further present image-assisted WiFi ﬁngerprinting in Chap. 8 and address several practical causes of location errors in Chap. 9.

Preface Anticipated Audience vii The book considers the state-of-the-art research results in many academic journals and conferences during its preparation. Thus, readers can track trends and hot topics in the ﬁeld. With the detailed introduced techniques for practical localization systems, it would draw attention of scientists and researchers in mobile and ubiquitous computing area and other related areas. The provided techniques would also systematically contribute to the practical applications and well guide the readers. Thus, it can serve as a guide book for the technicians and practitioners in the industry of location-based service and mobile sensing. Moreover, we wish this book can genuinely beneﬁt all levels of readers. The book intends to ﬁt the audience’s different technical backgrounds and career objectives and provides a balance of localization technology and location-aware applications. It does not require very speciﬁc knowledge of location-awareness technology. We kindly suggest the readers should have some basic knowledge of computer algorithms and wireless networks. Should the readers have any questions or suggestions, please contact the authors by email via wucs32@gmail.com. College Park, USA Beijing, China Beijing, China March 2018 Chenshu Wu Zheng Yang Yunhao Liu

Acknowledgments We would like ﬁrst to thank the China Computer Federation (CCF) for the precious CCF Excellent Doctoral Dissertation Award. To my knowledge, it was this award that led to the chance of publishing a book on my research results with Springer. We are grateful to many former and present students in our research groups for their contributions to the original research content covered by multiple chapters. In particular, we are most grateful to Dr. Junliang Liu at Google, Dr. Han Xu at Huawei, Dr. Zimu Zhou at ETH Zurich, Chaowei Xiao at University of Michigan, Ann Arbor, Zuwei Yin and Jingao Xu at Tsinghua University. This book would not have been possible without their contributions. During the past years, I have learned and beneﬁted a lot from many professors. I would like to take this opportunity to record my gratitudes to them. I thank Prof. Mingyan Liu at University of Michigan, Ann Arbor, who offered freedom, trust, and valuable guidance during my visit in her group and provided helpful suggestions and comments on several of my research works. Thanks are also due to Prof. Jiannong Cao at Hong Kong Polytechnic University, who kindly provided me a short-term visit opportunity during my Ph.D. study. In particular, sincere thanks to Prof. Nicholas D. Lane at University of Oxford and Nokia Bell Labs for his kind support and advices on my research. I am also grateful to Prof. Junzhou Luo at Southeast University for his guidance and support in my academic career. I thank Prof. Kyle Jamieson at Princeton University for kindly hosting a postdoc position. I am grateful to Prof. K. J. Ray Liu at University of Maryland, College Park for providing wonderful research environments. I wish to express my gratitude to many colleagues, who worked closely with me and inspired me a lot on research: thanks to Kun Qian, Yue Zheng, Junjie Yin, Tianmeng Hang, Xu Wang, Dr. Xinglin Zhang, Dr. Wei Sun, Dr. Yiyang Zhao, Dr. Longfei Shangguan, Dr. Xiaolong Zheng, etc. I would like to thank the editor Dr. Xiaolan Yao for the careful and considerable efforts in helping with the preparation and timely publication of this book. ix

分享到：

赞收藏

资料库

Wireless Indoor Localization.pdf

相关推荐

网络技术

热门标签

最新资料