增強 MQTT-base物聯網安全架構及建構之智慧自動化溫室系統

Enhancement on the MQTT-Based IoT System Sericurity and a Green House IoT System

陳奕叡(Yi-Jui Chen)

指導教授 ： 簡宏宇

國立暨南國際大學/管理學院/資訊管理學系/碩士(2018年)

摘要

隨著物聯網技術快速發展下，生活中充斥著大量的物聯網應用，本研究關心台灣農業面臨轉型精緻化栽種的過程以及面對大量設備在網路上的訊息傳輸安全性問題。近年來因應政府補助措施，溫室如雨後春筍般大量建置，溫室具有隔離外界環境影響以及提供適合農作物生長的環境等優點，可以有效的提高農作物的產量以及生長品質。但是傳統的溫室建置需要昂貴的成本以及傳統控制界面不夠友善(或者只是普通的手動排程設定)，大大降低了溫室的方便性，隨著雲時代的來臨、低成本的物聯網(IoT)建置越來越成熟，我們可以用這些低成本高效率的建置來收集監控植物生長資訊，再透過這些收集的大量資料(例：天氣變化數據、農作物生長數據、病蟲害分析等…。)做進一步的植物生長分析，將可大大的改善溫室種植的效益及維護成本。本研究針對中小型溫室設計開發出一套方便操控的低成本溫室自動化監控系統並將收集到的數據透過日誌檔的方式將重要資訊做儲存，並收集紅外線照片的每一像素透過 Normalized Difference Vegetation Index(NDVI)運算計算出一筆數值來判斷是否為光合作用旺盛的區域，然後將收集到的數據使用數據分析軟體( Splunk )進行初步的分析。
由於物聯網設備大多擁有較少的計算資源及暴露於較易被攻擊的環境，因此物聯網系統的佈建也帶來極大的被攻擊危機。在眾多的物聯網安全挑戰下，此研究偏重在設備的認證(這也是目前眾多以Message Queuing Telemetry Transport ，簡稱MQTT)為通訊架構的物聯網系統的大弱點。本研究提出基於MQTT協議為基礎之設備身分鑑別管理系統及安全物聯網，此系統可以整合管理及認證各項物聯網設備，並與 Splunk 系統整合以進一步作資料分析與管控。

關鍵字

MQTT ； 設備身分鑑別 ； 數據分析 ； 溫室自動化 ； NDVI

並列摘要

Various Internet of Things (IoT) deployments have become more and more popular in our daily life. It not only enriches our life experiences but also enhances the convenience of many services. However, IoT devices are usually resource-limited and are deployed in many hostile environments. It incurs many potential attacks and threats. Agriculture is one of many fields that might benefit from the IoT technologies. In this thesis, we concern two main challenges： one is secure IoT framework and the other is a Green-House IoT system under this framework.
In Taiwan agriculture practice, each farmer usually has a very limited space and limited facilities. In such conditions, a green-house system augmented with effective IoT systems could improve the productivity and the quality of the produce. The traditional control interface of a green-house system is not friendly (some are just manual setting), costly and not data-driven. With the advent of the cloud computing and low-cost Internet-of-Things (IoT) platforms, we can apply these low-cost and effective technologies to monitor/control the environment conditions and the facilities. And, a real-time data analysis is crucial to the success of an agriculture IoT system. With RethinkDB, raspyberry pi, tornado, and Splunk, we develop an efficient-and-effective greenhouse system to achieve the above goals.
Regarding the secure IoT framework, we especially focus on the device management and authentication of those Message Queuing Telemetry Transport (MQTT)-based IoT systems, because they are very popular. We propose a secure MQTT-based IoT framework and design an effective device management and authentication mechanism. With this framework and mechanism, we can effective manage/authenticate the devices. One advantage of our design is that the mechanism can be easily integrated with existent MQTT standards and platforms.

並列關鍵字

MQTT ； device authentication ； data analysis ； Greenhouse automation ； NDVI

參考文獻

[1] ETSI TS 102 690 v2.1.1, "Machine-to-Machine communications (M2M)： Functional architecture", 2013

[2] G. Tiantian,Z. Weizhu. "Design and implementation of the span greenhouse agriculture Internet of Things system." Fluid Power and Mechatronics (FPM), 2015 International Conference on. IEEE, 2015, pp. 398-401.

[3] 陳希煌.2014.農業經濟情勢與政策調適－台灣農業發展研討.台灣.財團法人台灣經濟研究院.237pp

[4] L. Sanbo. "Application of the internet of things technology in precision agriculture irrigation systems." Computer Science & Service System (CSSS), 2012 International Conference on. IEEE, 2012, pp. 1009-1013.

[5] S. Vatari, A. Bakshi, and T. Thakur. "Green house by using IOT and cloud computing." Recent Trends in Electronics, Information & Communication Technology (RTEICT), IEEE International Conference on. IEEE, 2016, pp. 246-250.