本文提出一個多用途可擴充之無線分散式負載感測平台，旨在作為需要多個負載節點共同協作之應用程式的基礎。我們的方法是將傳統的負載感測平台分解成兩部分：一個主機子系統與一個節點子系統。一個節點子系統包含多個負載感測節點。感測節點負責量測待測物之重量並將數位化後之讀數透過低功耗藍芽技術傳送給與此節點連線之主機子系統，而主機子系統著重於資料收集與資料分析處理。
我們實作了兩個應用程式，磅秤應用程式以及使用者辨識程式，用於展示我們所提出平台之通用性。在磅秤應用程式中，實驗結果顯示，使用單個負載感測節點之平均絕對誤差範圍為0.452至3.147公克；另外，使用四個節點共同協作之平均絕對誤差則降低為0.586公克。然而，使用者辨識程式則致力於利用最近鄰居法模型並透過不同的使用者腳步姿態，從一小群人中識別出一個特定使用者。交叉驗證結果顯示出，此應用之精準度達到98.33%。我們的實驗結果證實此平台之機動性、靈活性、可擴充性與多用途性。

We propose a general-purpose scalable distributed wireless load-sensing platform that serves as infrastructure for applications that necessitate the collaboration among multiple load-sensing units. The load-sensing platform consists of a host subsystem and a node subsystem. A node subsystem comprises one or multiple nodes that perform load-sensing, i.e., weight measurement, and transmit the data to the host via BLE (Bluetooth Low Energy). The host subsystem performs computation for data analysis.
We demonstrate the generality of our proposed platform by building a weight scale application and a user-identification application. For the weight application, experimental results show that a four-node collaborative weight scale achieves a lower average absolute error of 0.586 grams than individual nodes, whose absolute errors range from 0.452 to 3.147 grams. In the user-identification application, which uses a kNN model for identifying a person among a small group of people based on the gait, results show the accuracy of up to 98.33%. The results confirm the mobility, flexibility, scalability, and versatility of our proposed load-sensing platform.

Contents i
Acknowledgments vi
1 Introduction 1
1.1 Motivation . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 Contributions . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.3 Thesis Organization . . . . . . . . . . . . . . . . . . . . . . 3
2 Related Work 4
2.1 Ubiquitous Computing . . . . . . . . . . . . . . . . . . . . . . 4
2.2 Motion Detection . . . . . . . . . . . . . . . . . . . . . . . . 5
3 Background Theory 7
3.1 Strain Gauge Load Cell . . . . . . . . . . . . . . . . . . . . . 7
3.2 Wheatstone Bridge . . . . . . . . . . . . . . . . . . . . . . . 10
4 System Architecture and Implementation 12
4.1 System Architecture . . . . . . . . . . . . . . . . . . . . . . 12
4.2 Node Subsystem . . . . . . . . . . . . . . . . . . . . . . . . 12
4.2.1 Load Cell . . . . . . . . . . . . . . . . . . . . . . . . . . 14
4.2.2 PGA and ADC . . . . . . . . . . . . . . . . . . . . . . . . . 14
4.2.3 EcoMini . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
4.3 Host Subsystem . . . . . . . . . . . . . . . . . . . . . . . . 19
4.3.1 Data Collection . . . . . . . . . . . . . . . . . . . . . . . 19
4.3.2 Calibration . . . . . . . . . . . . . . . . . . . . . . . . . 19
5 Evaluation 22
5.1 Experimental Setup . . . . . . . . . . . . . . . . . . . . . . 22
5.2 Calibration . . . . . . . . . . . . . . . . . . . . . . . . . . 23
5.3 Scale Application . . . . . . . . . . . . . . . . . . . . . . . 27
5.3.1 Single-Node Subsystem . . . . . . . . . . . . . . . . . . . . 27
5.3.2 Multiple-Node Subsystem . . . . . . . . . . . . . . . . . . . 28
5.4 User Identification Application . . . . . . . . . . . . . . . . 31
6 Conclusions and Future Work 34
6.1 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . 34
6.2 Future Work . . . . . . . . . . . . . . . . . . . . . . . . . . 34

[1] “24-Bit Analog-to-Digital Converter (ADC) for Weight Scales.” https://cdn.sparkfun.com/datasheets/Sensors/ForceFlex/hx711_english.pdf.
[2] N. A. Streitz, J. Geißler, T. Holmer, S. Konomi, C. Müller-Tomfelde, W. Reischl, P. Rexroth, P. Seitz, and R. Steinmetz, “i-land: An interactive landscape for creativity and innovation,” in Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, CHI ’99, (New York, NY, USA), pp. 120–127, ACM, 1999.
[3] M. D. Addlesee, A. Jones, F. Livesey, and F. Samaria, “The orl active floor [sensor system],” IEEE Personal Communications, vol. 4, pp. 35–41, Oct 1997.
[4] R. J. Orr and G. D. Abowd, “The smart floor: A mechanism for natural user identification and tracking,” in CHI ’00 Extended Abstracts on Human Factors in Computing Systems, CHI EA ’00, (New York, NY, USA), pp. 275–276, ACM, 2000.
[5] A. Schmidt, M. Strohbach, K. Van Laerhoven, A. Friday, and H.-W. Gellersen, “Context acquisition based on load sensing,” UbiComp 2002: Ubiquitous Computing, pp. 161–192, 2002.
[6] J. M. Choi, B. H. Choi, J. W. Seo, R. H. Sohn, M. S. Ryu, W. Yi, and K. S. Park, “A system for ubiquitous health monitoring in the bedroom via a bluetooth network and wireless lan,” in The 26th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, vol. 2, pp. 3362–3365, Sept 2004.
[7] A. M. Adami, M. Pavel, T. L. Hayes, and C. M. Singer, “Detection of movement in bed using unobtrusive load cell sensors,” IEEE Transactions on Information Technology in Biomedicine, vol. 14, pp. 481–490, March 2010.
[8] Z. T. Beattie, C. C. Hagen, M. Pavel, and T. L. Hayes, “Classification of breathing events using load cells under the bed,” in 2009 Annual International Conference of the IEEE Engineering in Medicine and Biology Society, pp. 3921–3924, Sept 2009.
[9] M. Safizadeh and S. Latifi, “Using multi-sensor data fusion for vibration fault diagnosis of rolling element bearings by accelerometer and load cell,” Information Fusion, vol. 18, no. Supplement C, pp. 1 – 8, 2014.
[10] C. Slattery and M. Nie, “A reference design for high-performance, low-cost weigh scales,” Analog Dialogue, vol. 39, no. 4, pp. 13–18, 2005.