臺灣博碩士論文加值系統

本文提出了一種新的工業物聯網系統框架，提供關鍵設備能具備低延遲特點。由於目前大多數物聯網系統中，主要議題在提供大規模應用或能耗問題做討論，比較少研究考慮物聯網的延遲性能。而且，由於感測器行為無法預測，還有不可抗拒的環境變因，若須時時刻刻監控延遲性能，將提高整體建置成本。為了降低封包並提升服務質量，我們所採用的路管理框架，可以保證聚合器的網路性能。本論文將關注聚合器和感測器之間的配對方法，針對網路中不同的優先級感測器，我們將透過調整配對感測器與聚合器來提高延遲性能並保證傳輸質量。除此之外，透過預測尚未發生的任務關鍵事件，使傳輸資料在不同的環境變化下更可靠。為此，我們設計一套基於群組的工業物聯網的實驗研究方法，整合現有的消息隊列遙測傳輸協議（MQTT），用於控制物聯網設備並動態更新與集群相關的參數，並通過WiFi與行動網路，從我們的物聯網系統中設置並提取關鍵數據。最後，通過深度學習網路生成的人工數據來評估配對的延遲性能。結果證明所提出的系統框架的可行性，並透過所提出的配對模型可以減少40%-60%之端到端延遲。

This thesis presents a novel system framework for industrial Internet of Things to support mission-critical applications with delay constraints. While most IoT systems aim at providing massive connectivity for delay-tolerant applications, few works address delay-sensitive IoT applications. Besides, due to the unplanned stop and the unpredictable sensor behaviors by the various environment, it is costly to improve performance by tracking the IoT system. To re-duce the drop rate and improve the quality of service, we adopt Network of Thing (NoT) framework. In this thesis, we focus on the association issue between an aggregator and sensors. For sensors with different priorities in the network, we design the association rule to improve the delay performance and ensure the transmission quality. For flexible associating decision, we predict the possible association scheme with the different environments by generating an artificial event. We implement NoT system by utilizing message queuing telemetry transport (MQTT) protocol to dynamically update the association scheme. We describe how the delay data from the Industrial IoT system can be collected and the mission-critical data can be extracted. We evaluate the delay performance of different artificial events, which are generated by neural network with corresponding association scheme. Our results demonstrate that the proposed system framework can improve the delay performance of industrial IoT by 40% to 60%.

1 Introduction 1
1.1 Industrial Internet of Things . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 Network of Things: Design issue and Challenge . . . . . . . . . . . . . . . 2
1.3 Propose Solution and Contribution . . . . . . . . . . . . . . . . . . . . . . 4
2 Background 5
2.1 General Association Approach . . . . . . . . . . . . . . . . . . . . . . . . . 5
2.2 Learning-based Association with Insufficient Data . . . . . . . . . . . . . . 6
2.3 Conditional GAN for Identifying Different Priority Traffic . . . . . . . . . . 8
2.4 Related Work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.4.1 Association of Aggregator In Industrial IoT . . . . . . . . . . . . . 9
2.4.2 Decision Making with Industrial IoT . . . . . . . . . . . . . . . . . 9
3 System Model and Problem 12
3.1 Model and Assumption . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
3.2 Delay Analysis Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
3.3 Problem Statement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
4 Predictive Delay-aware Association 16
4.1 Predictive Event Generator . . . . . . . . . . . . . . . . . . . . . . . . . . 16
4.1.1 The Scheme of Conditional GAN with Bayesian latent . . . . . . . 18
4.2 Delay-aware Association . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
4.2.1 Delay Performance Indicator . . . . . . . . . . . . . . . . . . . . . . 22
4.2.2 Procedure of the Delay-aware Association . . . . . . . . . . . . . . 22
5 Implementation of IIoT System with Delay Measurement System 27
5.1 Architecture Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
5.2 Network Stack Description . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
6 Experimental Results 33
6.1 Testbed . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
6.2 Scenario . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
6.3 Traffic Evolution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
6.3.1 Generative Data from cGAN Models . . . . . . . . . . . . . . . . . 38
6.3.2 Delay performance Indicator . . . . . . . . . . . . . . . . . . . . . . 43
6.3.3 Association Result . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
7 Conclusion 47
Bibliography 48

[1] S. Mumtaz, A. Alsohaily, Z. Pang, A. Rayes, K. F. Tsang, and J. Rodriguez, “Massive Internet of Things for industrial applications: Addressing wireless IIoT connectivity challenges and ecosystem fragmentation,” in IEEE Industrial Electronics Magazine, vol. 11, no. 1, 2017.
[2] Z. Shelby, K. Hartke, and C. Bormann, The Constrained Application Protocol, RFC 7252, Internet Engineering Task Force, June 2014.
[3] P. Thota and Y. Kim, “Implementation and comparison of M2M protocols for Internet of Things,” in IEEE Applied Computing and Information Technology/3rd IntlConf on Computational Science/Intelligence and Applied Informatics/1st Intl Conf on Big Data, Cloud Computing, Data Science and Engineering, 2016.
[4] D.-K. Choi, J.-H. Jung, H.-W. Kang, and S.-J. Koh, “Cluster-based CoAP for message queueing in Intenet-of-Things networks,” in IEEE International Conference on Advanced Communication Technology, 2017.
[5] J. Voas, “Networks of things,” in NIST Special Publication, vol. 800, no. 183, 2016.
[6] C. Zhu, S. Wu, G. Han, L. Shu, and H. Wu, “A tree-cluster-based data-gathering algorithm for industrial WSNs with a mobile sink,” in IEEE Access, vol. 3, 2015.
[7] Q. Zhang, C. Zhu, L. T. Yang, Z. Chen, L. Zhao, and P. Li, “An incremental CFS algorithm for clustering large data in industrial Internet of Things,” in IEEE Transactions on Industrial Informatics, vol. 13, no. 3, 2017.
[8] H. Guo, J. Liu, and H. Qin, “Collaborative mobile edge computation offloading for IoT over Fiber-Wireless networks,” in IEEE Network, vol. 32, no. 1, 2018.
[9] J. Chen, L. Zhou, B. Zheng, and J. Cui, “Application-dependent frame design for the Internet of Things,” in IEEE Communications, 2013.
[10] I. Goodfellow, J. Pouget-Abadie, M. Mirza, B. Xu, D. Warde-Farley, S. Ozair, A. Courville, and Y. Bengio, “Generative adversarial nets,” in Advances in Neural Information Processing Systems, 2014.
[11] M. Mirza and S. Osindero, “Conditional generative adversarial nets,” arXiv preprint arXiv:1411.1784, 2014.
[12] D. Puschmann, P. Barnaghi, and R. Tafazolli, “Adaptive clustering for dynamic IoT data streams,” in IEEE Internet of Things Journal, vol. 4, no. 1, 2017.
[13] G. Fortino, R. Gravina, W. Russo, and C. Savaglio, “Modeling and simulating Internet-of-Things systems: A hybrid agent-oriented approach,” in Computing in Science and Engineering, vol. 19, no. 5, 2017.
[14] T. Dinh and Y. Kim, “Information centric sensor-cloud integration: An efficient model to improve wireless sensor networks’ lifetime,” in IEEE Communications, 2017.
[15] M. Antunes, R. Jesus, D. Gomes, and R. L. Aguiar, “Improve IoT/M2M data organization based on stream patterns,” in IEEE Future Internet of Things and Cloud, 2017.
[16] D. Sharma and A. P. Bhondekar, “Traffic and energy aware routing for heterogeneous wireless sensor networks,” in IEEE Communications Letters, 2018.
[17] 3GPP, “Discussion on sidelink UE-to-NW relaying,” 3rd Generation Partnership Project (3GPP), Tech. Rep. R1-1708263, May 2017.
[18] Y.-J. Chen, Z.-Q. Wang, and L.-C. Wang, “Impact of aggregation factor on delay performance in group-based machine type communications,” in IEEE Annual International Symposium on Personal, Indoor, and Mobile Radio Communications, 2017.
[19] “Message Queue Telemetry Transport,” [Online]. Available: https://mqtt.org/.
[20] P. T. Eugster, P. A. Felber, R. Guerraoui, and A.-M. Kermarrec, “The many faces of publish/subscribe,” ACM computing surveys, vol. 35, no. 2, pp. 114–131, 2003.
[21] T. Bray, The JavaScript Object Notation data interchange format, RFC 7159, Internet Engineering Task Force, Mar. 2014.
[22] “Arduino YUN,” [Online]. Available: https://store.arduino.cc/usa/arduino-yun.
[23] “OpenWrt,” [Online]. Available: https://openwrt.org/.
[24] “Huawei USB dongle E3372,” [Online]. Available: http://consumer.huawei.com/en/mobile-broadband/e3372/specs/.
[25] A. Ciuffoletti, “OCCI-IoT: An API to deploy and operate an IoT infrastructure,” in IEEE Internet of Things Journal, vol. 4, no. 5, 2017.
[26] “Mitrastar Femto Pico Cell,” [Online]. Available: http://www.mitrastar.com.tw/.
[27] “Next generation Telecommunication Technology Testing Tools,” [Online]. Available: http://www.ng4t.com/.
[28] “paho-mqtt,” [Online]. Available: https://pypi.python.org/pypi/paho-mqtt.
[29] “ntplib,” [Online]. Available: https://pypi.python.org/pypi/ntplib/.
[30] D. Mills, J. Martin, J. Burbank, and W. Kasch, Network Time Protocol Version 4: Protocol and Algorithms Specification, RFC 5905, Internet Engineering Task Force, June 2010.
[31] “Mosquitto,” [Online]. Available: https://mosquitto.org/.