近年來，隨著物聯網的快速發展，個人、工業和農業…等都已經開始應用這種技術來解決產業上遇到的困難。物聯網使數據收集更加方便，讓我們能夠隨時監控環境數據，並且能遠程操作控制器，不用親自到現場就能控制，更進階的控制是能夠自訂規則或自動學習，根據環境收集的數據自動決定是否要啟動控制器，達到智慧控制的效果。
台灣對農業的物聯網需求也有所增加。我們可以利用農場收集的數據進行多種分析和應用，解決農民在工作中遇到的問題。然而，當感測器數據異常時，我們無法立即知道感測器數據已經發生異常。這種情況可能真的是農場環境中的異常情況，需要及時處理，否則可能會對農場造成危害。要隨時檢查收集的環境數據會耗費大量的時間及人力，但是目前並沒有一個整合異常狀況偵測功能的物聯網平台系統。
因此在此篇論文中，我們設計符合農業環境數據的異常數據偵測方法來隨時監控環境數據，當有異常狀況發生時，幫助使用者找出有異常現象的感測器，並且由我們的物聯網平台發送通知到使用者手機app，告知發生的時間和可能造成的原因，來幫助使用者找出農場的問題。通過偵測異常數據，可以幫助減少由於環境或人為引起的農場損失。

In recent years, with the rapid development of the Internet of Things, individuals, industry and agriculture have begun to apply this technology to solve industrial difficulties. IoT makes data collection more convenient, allowing us to monitor environmental data at any time, and can remotely operate the controller without having to go to the site to control it.
The demand for IoT in agriculture has also increased in Taiwan. We can use the data collected by the farm to do many analysis and application to solve the problems encountered by farmers at work. However, when the sensor data is abnormal, we cannot immediately know that the sensor data has been abnormal. This situation may be an abnormal situation in the farm environment and needs to be dealt with in time, otherwise it may cause harm to the farm. But monitoring environmental data at any time is time consuming and labor intensive. However, there is currently no IoT platform that integrates abnormal condition detection.
Therefore, in this thesis, we design some anomaly data detection methods that conforms to agricultural environmental data to monitor environmental data at any time. When an abnormal situation occurs, the user is helped to find an abnormal sensor, and by our IoT platform sends notifications to the user's mobile app to inform them of the time of the event and the possible causes to help the user identify the farm's problems. By detecting anomalous data, it can help reduce farm losses due to the environment or man-made.

Abstract I
中文摘要 II
List of Figures V
List of Tables VII
Chapter 1 Introduction 1
Chapter 2 Related Works 4
2.1 Basic Concepts of Anomaly Detection 4
2.1.1 Different Types of Outliers 5
2.1.2 The Feature of Anomaly Detection 7
2.2 LoRa and LoRaWAN 8
2.2.1 LoRa 9
2.2.2 LoRaWAN Architecture 10
2.3 Related Works of Anomaly Detection System 12
Chapter 3 System Design 14
3.1 Anomaly Detection System Architecture 14
3.1.1 API Server 16
3.1.2 Database Management 21
3.1.3 Anomaly Detection Module 23
3.1.4 Web Server 24
3.2 System Implementation Discussion 25
3.2.1 Front-end Implementation Discussion 25
3.2.2 Back-end Implementation Discussion 26
3.3 Detection Algorithms 27
3.3.1 Type 1: Sensing data from stable to drastic 31
3.3.2 Type 2: Sensing data from drastic to stable 37
3.3.3 Type 3: Sensing data has clear abnormal standards 39
3.4 Calculate Detection Period 40
Chapter 4 System Implementation and Experiment 42
4.1 Data Collection Environment 42
4.2 Experimental Results 44
4.2.1 Selecting thresholds 44
4.2.2 Visualization results and instant notification 48
Chapter 5 Conclusion and Future Works 52
References 55

[1] Koustav Routh, Tannistha Pal, “A survey on technological, business and societal aspects of Internet of Things by Q3, 2017”, 2018 3rd International Conference On Internet of Things: Smart Innovation and Usages (IoT-SIU), Bhimtal, India, pp. 1-4, 23-24 Feb. 2018.
[2] Timothy Malche, Priti Maheshwary, “Internet of Things (IoT) for building smart home system”, 2017 International Conference on I-SMAC (IoT in Social, Mobile, Analytics and Cloud) (I-SMAC), Palladam, India, pp. 65-70, 10-11 Feb. 2017.
[3] Rahul Dagar, Subhranil Som, Sunil Kumar Khatri, “Smart Farming – IoT in Agriculture”, 2018 International Conference on Inventive Research in Computing Applications (ICIRCA) , Coimbatore, India, pp. 1052-1056, 11-12 July 2018.
[4] Preeti Yadav, Sandeep Vishwakarma, “Application of Internet of Things and Big Data towards a Smart City”, 2018 3rd International Conference On Internet of Things: Smart Innovation and Usages (IoT-SIU), Bhimtal, India, pp. 1-5, 23-24 Feb. 2018.
[5] LV Yaqiong, TU Lei, C.K.M LEE, TANG Xin, “IoT based Omni-Channel Logistics Service in Industry 4.0”, 2018 IEEE International Conference on Service Operations and Logistics, and Informatics (SOLI), Singapore, pp. 240-243, 31 July-2 Aug. 2018.
[6] Ericsson, “Ericsson mobility report,” White paper, 2017.
[7] LoRa Alliance, “LoRaWAN specification,” Jul. 2016, Available: https://lora-alliance.org/
[8] “Semtech,” [Online]. Available: https://www.semtech.com/lora
[9] Yuanyan Luo, Xuehui Du, Yi Sun, “Survey on Real-time Anomaly Detection Technology for Big Data Streams”, 2018 12th IEEE International Conference on Anti-counterfeiting, Security, and Identification (ASID), Xiamen, China, pp. 26-30, 9-11 Nov. 2018.
[10] Chandola, V., Benerjee, A., Kumar, V.: Anomaly Detection, A Survey (2007).
[11] “Different Types of Outliers,” [Online].
Available: https://www.anodot.com/blog/quick-guide-different-types-outliers/
[12] Chandola, V., Banerjee, A., and Kumar, V. 2016. Anomaly Detection. in Encyclopedia of Machine Learning and Data Mining, C. Sammut and G. I. Webb, Eds. Boston, MA: Springer US, 1–15.
[13] “Positioning of LPWAN technology,” [Online].
Available: https://behrtechnologies.com/blog/tag/utilities/
[14] H. Lu et al., “Motor anomaly detection for unmanned aerial vehicles using reinforcement learning,” IEEE Internet Things J., vol. 5, no. 4, pp. 2315–2322, Aug. 2018.
[15] Chuan-Yu Chang, Yi-Ping Chang, “Application of abnormal sound recognition system for indoor environment”, 2013 9th International Conference on Information, Communications & Signal Processing, Tainan, Taiwan, 10-13 Dec. 2013.
[16] Jae-Jung Lee, Gyu-Jin Kim, Moon-Hyun Kim, “Trajectory extraction for abnormal behavior detection in public area”, 2012 9th International Conference & Expo on Emerging Technologies for a Smarter World (CEWIT), Incheon, South Korea, 5-6 Nov. 2012.
[17] “MQTT 3.1.1 specification”. OASIS. December 10, 2015. Retrieved April 25, 2017.
[18] “JSON Web Token,” [Online].
Available: https://segmentfault.com/a/1190000005783306
[19] “JWT acquisition and verification,” [Online].
Available: https://www.itread01.com/p/838503.html
[20] “Node.js,” [Online]. Available: https://nodejs.org/en/
[21] “Express,” [Online]. Available: https://expressjs.com/
[22] “Bootstrap,” [Online]. Available: https://getbootstrap.com/
[23] “jQuery,” [Online]. Available: https://jquery.com/
[24] “Highcharts,” [Online]. Available: https://www.highcharts.com/
[25] “Amazon DynamoDB,” [Online].
Available: https://aws.amazon.com/tw/dynamodb/
[26] “well water samples ph levels,” [Online]. Avaiable: https://www.sciencefriday.com/educational-resources/diy-groundwater-model/
[27] “decibel detection data,” [Online]. Available: http://sustainability.ntu.edu.tw/action_info.aspx?ID=5
[28] Luminto, Harlili, “Weather analysis to predict rice cultivation time using multiple linear regression to escalate farmer's exchange rate”, 2017 International Conference on Advanced Informatics, Concepts, Theory, and Applications (ICAICTA), Denpasar, Indonesia, pp. 1-4, 16-18 Aug. 2017.
[29] S. Nagini, T. V. Rajini Kanth, B. V. Kiranmayee, “Agriculture yield prediction using predictive analytic techniques”, 2016 2nd International Conference on Contemporary Computing and Informatics (IC3I), Noida, India, pp. 783-788, 14-17 Dec. 2016.
[30] S. Swain, P. Patel, S. Nandi, “A multiple linear regression model for precipitation forecasting over Cuttack district, Odisha, India”, 2017 2nd International Conference for Convergence in Technology (I2CT), Mumbai, India, pp. 355-357, 7-9 April 2017.
[31] Y. Zhao et al., “Data-driven correction approach to refine power curve of wind farm under wind curtailment,” IEEE Trans. Sustain. Energy, vol. 9, no. 1, pp. 95–105, Jan. 2018.
[32] X. Shen, X. Fu, and C. Zhou, ‘‘A combined algorithm for cleaning abnormal data of wind turbine power curve based on change point grouping algorithm and quartile algorithm,’’ IEEE Trans. Sustain. Energy, vol. 10, no. 1, pp. 46–54, Jan. 2019.
[33] Y. Zhao, L. Ye, and Q. Zhu, “Characteristics and processing method of abnormal data cluster caused by wind curtailments in wind farms”, Autom. Electr. Power Syst., vol. 38, no. 27, pp. 39–46, 2014.
[34] “LinkIt 7697,” [Online]. Available: https://labs.mediatek.com/zh-tw/platform/linkit-7697
[35] “Gemtek GL6509 LoRa module,” [Online].
Available: https://www.gemteks.com/products/lora-iot/module
[36] “Amazon EC2,” [Online]. Available: https://aws.amazon.com/tw/ec2/
[37] “QuickDBD,” [Online]. Available: https://www.quickdatabasediagrams.com/
[38] “draw.io,” [Online]. Available: https://www.draw.io/