進階搜尋


下載電子全文  
系統識別號 U0026-0408201615031900
論文名稱(中文) 使用低耗電藍牙裝置與樸素貝式分類法之室內定位技術
論文名稱(英文) Indoor Positioning Using Bluetooth Low Energy Devices and Naive Bayes Classification
校院名稱 成功大學
系所名稱(中) 工程科學系
系所名稱(英) Department of Engineering Science
學年度 104
學期 2
出版年 105
研究生(中文) 王博威
研究生(英文) Po-Wei Wang
學號 n96034154
學位類別 碩士
語文別 英文
論文頁數 68頁
口試委員 指導教授-侯廷偉
口試委員-王明習
口試委員-鄧維光
口試委員-席家年
中文關鍵字 室內定位  低耗電藍牙  樸素貝式分類器 
英文關鍵字 Indoor Positioning  BLE  naive Bayes classifier 
學科別分類
中文摘要 室內定位系統的評比,除了考量定位精準度外,建造成本亦在衡量範圍內,以現行常見的定位技術而言,特徵定位法精準度高、佈建時間成本高,三角定位法精準低,但佈建時間成本低,本研究目標為結合特徵定位法的優點與三角定位法的優點,提出一套能降低時間成本但提高定位精準度之室內定位演算法,本研究採用iBeacon作為訊號源,iBeacon由低耗電藍牙(BLE, Bluetooth Low Energy)發展而成,由於BLE頻寬窄,導致訊號衰減快速,造成傳統距離轉換模型沒辦法使用在BLE訊號源上,離訊號源愈遠所造成的轉換誤差愈大,本研究以混和濾波器減少訊號震盪,並利用樸素貝式分類器對距離藉由訊號強度進行分類,最後在使用最小平方估計法進行多點定位,得到定位結果,實驗結果顯示,比較傳統距離轉換模型,更能提升角落的定位精準度,使用樸素貝式分類器之準確率於1.5公尺內次數可達80%,而使用LAM(Linear Approximation Model)之準確率於1.5公尺內次數僅有48%,利用本研究所提出之室內定位演算法進行小區塊分類結果其準確率與特徵定位演算法相近。
英文摘要 The efficiency of an indoor positioning system can be evaluated by its positioning accuracy and building cost. There are currently two representative indoor positioning techniques. Fingerprinting has high positioning accuracy and high building time cost, and triangulation has less positioning accuracy and less building time cost. The purpose of our research is to combine the advantages of both the fingerprinting and triangulation techniques in order to propose a new indoor positioning algorithm that can reduce building time and can enhance positioning accuracy. Our research takes iBeacon, which is extended from BLE (Bluetooth Low Energy) as transmitters. Due to BLE’s narrow channel space, BLE signal strength descends rapidly. This makes Linear Approximation Model not feasible in distance translation. The farther from transmitter, the more distance translating error. In this research, the hybrid filter is used to converge oscillating signal values, and the naive Bayes classifier is used to classify distance according to signal strength. Finally, more than three transmitters are used to calculate the location of objects using least square estimation. The experiment results show the proportion of positioning error using the naive Bayes classifier at 1.5 meters is about 80%, and the proportion of translating distance by LAM is only 48%. Compared to LAM, the proposed indoor positioning algorithm can increase positioning accuracy at corners, and compared to the fingerprinting method, the proposed indoor algorithm has nearly the same positioning accuracy.
論文目次 摘要.................... ................I
Abstract................................II
致謝....................................III
Table of Contents.......................IV
List of Tables..........................VI
List of Figures.........................VII
Chapter 1 Introduction..................1
1.1 Background..........................1
1.2 Motivation and Purpose..............3
1.3 Organization of Thesis..............4
Chapter 2 Related Work..................5
2.1 Indoor Positioning System...........5
2.2 Signal Filter.......................6
2.2.1 Moving Average Filter.............7
2.2.2 Weight Filter.....................8
2.2.3 Window Selection Filter...........9
2.2.4 Kalman Filter.....................9
2.3 Triangulation.......................10
2.3.1 Angle of Arrival..................10
2.3.2 Distance Method...................12
2.3.2.1 Time of Arrival (TOA)...........12
2.3.2.2 Time Difference of Arrival......14
2.3.2.3 Received Signal Strength........15
2.3.2.4 Free Space Model................17
2.3.2.5 Two Ray Ground Model............18
2.3.2.6 Shadowing Model.................20
2.3.2.7 Linear Approximation Model......20
2.4 Linear Regression...................21
2.5 Least Square Estimation.............23
2.6 Fingerprinting......................24
2.7 Naive Bayes.........................27
2.8 Multi-Floor Positioning.............29
Chapter 3 Proposed Algorithm............30
3.1 Proposed Approach...................30
3.2 Wireless Source.....................32
3.3 Linear Approximation Model VS. Naive Bayes Classifier 38
3.4 Positioning Algorithm...............40
Chapter 4 Experiments and Results.......44
4.1 Experimental Devices................44
4.2 Indoor Positioning Experiment.......45
4.2.1 Area one..........................45
4.2.2 Area two..........................49
4.2.3 Block three.......................53
4.3 Tracking Experiment.................57
4.4 Discussion..........................60
Chapter 5 Conclusion and Future Work....61
5.1 Conclusion..........................61
5.2 Future Work.........................62
References..............................63
Appendix A Multi-floor Positioning Method.......67

Table 1-1 Comparison of Indoor Positioning Technologies [3] [5].....2
Table 3-1 Standard Deviation from one to eight meters.....37
Table 4-1 Details of Transmitters in area one.....46
Table 4-2 Comparison of the proposed algorithm to fingerprinting in area one.....49
Table 4-3 Details of transmitters in area two.....50
Table 4-4 Comparison of the proposed algorithm to fingerprinting in area two.....53
Table 4-5 Details of transmitters in area three.....54
Table 4-6 A comparison of the proposed algorithm to fingerprinting in area three.....57

Figure 2.1 Indoor positioning technologies [3].....6
Figure 2.2 Multipath propagation.....7
Figure 2.3 Kalman filter [10].....10
Figure 2.4 AOA algorithm.....11
Figure 2.5 TOA algorithm.....14
Figure 2.6 TDOA algorithm.....15
Figure 2.7 Flow of proposed algorithm.....16
Figure 2.8 TRGM.....19
Figure 2.9 Same trend line with different data [18].....22
Figure 2.10 The positioning of four transmitters.....24
Figure 2.11 Signal conflict in center.....25
Figure 2.12 Flow of fingerprinting.....26
Figure 3.1 Flow of proposed algorithm.....31
Figure 3.2 BLE channels and the three most commonly used Wi-Fi channels [9].....33
Figure 3.3 iBeacon packet format [22].....33
Figure 3.4 1-D Kalman filter.....34
Figure 3.5 RSSI measured at one meter with Kalman filter.....35
Figure 3.7 RSSI measured at one meter with the hybrid filter.....37
Figure 3.8 Signal strength from one to eight meters.....38
Figure 3.9 LAM : RSS=-19.3258×logd-69.4657.....39
Figure 3.10 Naïve Bayes classifier classified result.....40
Figure 3.11 The concept of the correction method.....43
Figure 4.1 Experimental devices.....44
Figure 4.2 Floor plan of the laboratory.....46
Figure 4.3 Positioning error and standard deviation in area one.....47
Figure 4.4 Proportion of position error within 1 and 1.5 meters in area one.....48
Figure 4.5 Floor plan of the meeting room.....50
Figure 4.6 Positioning error and standard deviation in area two.....51
Figure 4.7 The proportion of position error within 1 and 1.5 meters in area two.....52
Figure 4.8 Floor plan of the public space.....54
Figure 4.9 Positioning error and standard deviation in area three.....55
Figure 4.10 Proportion of position error within 1 and 1.5 meters in area three.....56
Figure 4.11 The results of the first tracking experiment.....58
Figure 4.12 The results of the second tracking experiment.....60




參考文獻 [1] International Telecommunication Union. The Key 2005-2015 ICT Data for The World. [Online]. Available: http://www.itu.int/en/ITU-D/Statistics/Pages/stat/default.aspx. Accessed: Apr. 6, 2016.
[2] "Location-based service," in Wikipedia, Wikimedia Foundation, 2016. [Online]. Available: https://en.wikipedia.org/wiki/Location-based_service. Accessed: Apr. 6, 2016.
[3] M. A. Al-Ammar et al., "Comparative Survey of Indoor Positioning Technologies, Techniques, and Algorithms," 2014 International Conference on Cyberworlds (CW), Santander, Spain, 2014, pp. 245-252.
[4] "Global positioning system," in Wikipedia, Wikimedia Foundation, 2016. [Online]. Available: https://en.wikipedia.org/wiki/Global_Positioning_System. Accessed: Apr. 7, 2016.
[5] H. Liu, H. Darabi, P. Banerjee and J. Liu, "Survey of Wireless Indoor Positioning Techniques and Systems," in IEEE Transactions on Systems, Man, and Cybernetics, Part C (Applications and Reviews), vol. 37, no. 6, pp. 1067-1080, Nov. 2007.
[6] D. Čabarkapa, I. Grujić and P. Pavlović, "Comparative analysis of the Bluetooth Low-Energy indoor positioning systems," 2015 12th International Conference on Telecommunication in Modern Satellite, Cable and Broadcasting Services (TELSIKS), Nis, Serbia, 2015, pp. 76-79.
[7] "Moving average," in Wikipedia, Wikimedia Foundation, 2016. [Online]. Available: https://en.wikipedia.org/wiki/Moving_average. Accessed: Apr. 8, 2016.
[8] Z. Jianyong, L. Haiyong, C. Zili and L. Zhaohui, "RSSI based Bluetooth low energy indoor positioning," 2014 International Conference on Indoor Positioning and Indoor Navigation (IPIN), Busan, Korea, 2014, pp. 526-533.
[9] R. Faragher and R. Harle, "Location Fingerprinting With Bluetooth Low Energy Beacons," in IEEE Journal on Selected Areas in Communications, vol. 33, no. 11, pp. 2418-2428, Nov. 2015.
[10] G. Welch and G. Bishop, "An introduction to the Kalman Filter," Technical Report, University of North Carolina at Chapel Hill, 1955.
[11] "Kalman filter," in Wikipedia, Wikimedia Foundation, 2016. [Online]. Available: https://en.wikipedia.org/wiki/Kalman_filter. Accessed: Apr. 8, 2016.
[12] Y. Wang, Xu Yang, Yutian Zhao, Yue Liu and L. Cuthbert, "Bluetooth positioning using RSSI and triangulation methods," 2013 IEEE 10th Consumer Communications and Networking Conference (CCNC), Las Vegas, U.S., 2013, pp. 837-842.
[13] O. G. Adewumi, K. Djouani and A. M. Kurien, "RSSI based indoor and outdoor distance estimation for localization in WSN," 2013 IEEE International Conference on Industrial Technology (ICIT), Cape Town, South Africa, 2013, pp. 1534-1539.
[14] T. I. Chowdhury et al., "A multi-step approach for RSSi-based distance estimation using smartphones," 2015 International Conference on Networking Systems and Security (NSysS), Dhaka, Bangladesh, 2015, pp. 1-5.
[15] B. Mukhopadhyay, S. Sarangi and S. Kar, "Novel RSSI evaluation models for accurate indoor localization with sensor networks," 2014 Twentieth National Conference on Communications (NCC), Kanpur, India, 2014, pp. 1-6.
[16] Y. B. Bai, M. Williams, Falin Wu, A. Kealy and Kefei Zhang, "An accuracy enhancement algorithm for fingerprinting method," 2014 International Conference on Data Science and Advanced Analytics (DSAA), Shanghai, China, 2014, pp. 110-114.
[17] Andrew Ng, “CS229 Machine Learning Lecture notes 1: Supervised Learning, Discriminative Algorithms,” [Online]. Available: http://cs229.stanford.edu/notes/cs229-notes1.pdf

[18] "Linear regression," in Wikipedia, Wikimedia Foundation, 2016. [Online]. Available: https://en.wikipedia.org/wiki/Linear_regression. Accessed: Apr. 11, 2016.
[19] Andrew Ng, “CS229 Machine Learning Lecture notes 2: Generative Algorithms,” [Online]. Available: http://cs229.stanford.edu/notes/cs229-notes2.pdf

[20] "IBeacon," in Wikipedia, Wikimedia Foundation, 2016. [Online]. Available: https://en.wikipedia.org/wiki/IBeacon. Accessed: Apr. 11, 2016.

[21] H. K. Fard, Y. Chen and K. K. Son, "Indoor positioning of mobile devices with agile iBeacon deployment," 2015 IEEE 28th Canadian Conference on Electrical and Computer Engineering (CCECE), Halifax, Canada, 2015, pp. 275-279.
[22] “ibeacon Packet Format 分析,” in Jesse’s blog, (2014, Jun. 26). [Online]. Available: http://smlie-blog.blogspot.tw/2014/06/bluetooth-ibeacon-packet-format.html. Accessed: Apr. 11, 2016.
[23] N. A. Razak, N. H. M. Arshad, R. Adnan, M. F. Misnan, N. M. Thamrin and S. F. Mahmud, "A study of Kalman's filter in embedded controller for real-time quadrocopter roll and pitch measurement," 2012 IEEE International Conference on Control System, Computing and Engineering (ICCSCE), Penang, Malaysia, 2012, pp. 590-595.
[24] H. J. Lin, T. W. Hou, “Indoor Positioning Using Kalman Filter and Support Vector Machine Based on Bluetooth Low Energy,” unpublished.
[25] J. s. Jeon, Y. Kong, Y. Nam and K. Yim, "An Indoor Positioning System Using Bluetooth RSSI with an Accelerometer and a Barometer on a Smartphone," 2015 10th International Conference on Broadband and Wireless Computing, Communication and Applications (BWCCA), Krakow, Poland, 2015, pp. 528-531.
[26] E. S. Lohan, J. Talvitie, P. Figueiredo e Silva, H. Nurminen, S. Ali-Löytty and R. Piché, "Received signal strength models for WLAN and BLE-based indoor positioning in multi-floor buildings,"2015 International Conference on Location and GNSS (ICL-GNSS), Gothenburg, Sweden, 2015, pp. 1-6.
論文全文使用權限
  • 同意授權校內瀏覽/列印電子全文服務,於2016-08-09起公開。
  • 同意授權校外瀏覽/列印電子全文服務,於2016-08-09起公開。


  • 如您有疑問,請聯絡圖書館
    聯絡電話:(06)2757575#65773
    聯絡E-mail:etds@email.ncku.edu.tw