臺灣博碩士論文加值系統

隨著科技的不斷發展，專注度已經成為提升個人效率和企業生產力的關鍵因素。人們常常面臨許多分心的誘因，這可能對工作、學習和日常生活產生負面影響。為了應對這個挑戰，一個能夠有效監控專注度的系統變得至關重要。在這個背景下，生成對抗網路（Generative Adversarial Networks , GANs）是一個引人注目的工具，特別是在醫學領域的資料庫擴充中。透過將真實樣本與生成的偽裝樣本結合，GANs能夠增強資料集的多樣性，從而提高模型的泛化能力和性能。本篇論文將充分展示生成對抗網路在資料庫增強中的潛力，並且將其應用於室內定位與專注度監控系統中。本論文旨在建立一個創新的腦波燈控系統，透過結合腦波與定位資訊，實現智慧化的燈光調控。主要利用生成對抗網路進行腦波和定位資料集的增強，同時運用主成分分析（Principal Components Analysis , PCA）方法對資料進行維度檢測，以提高數據處理效能。透過機器學習（Machine Learning）與深度學習（Deep Learning）技術，將增強後的資料集訓練成分類模型。此系統將能夠根據用戶的腦波狀態和位置，動態調整燈光的範圍、照度與色溫，以提供更舒適、個性化的照明體驗。本篇論文將使用改良後的GANs （Generative Adversarial Networks）來增強腦波與定位資料集，分別將原資料集與增強後資料集丟入機器學習模型中學習並比較最後的評估指標（Evaluation Metrics），透過增強資料集任務來對GANs進行參數微調，增強資料集任務完成後再分別使用NN（Neural Network）、KNN（K-Nearest Neighbors）、SVM（Support Vector Machine）和LSTM（Long Short-Term Memory）進行專注度與定位分類比較。其中我們將引入灰狼演算法（Grey Wolf Optimizer, GWO）、模糊邏輯（Fuzzy Logic）與Fuzzy-PD來進一步提升定位和燈控系統的穩定性。最後再將訓練好的模型結合Raspberry Pi4等硬體完成一套「基於灰狼演算法的室內定位深度學習專注度監控系統」。最後實驗結果顯示，本系統在定位誤差方面可以達到15公分的公分級定位精度，並且專注度有顯著提升。

In the age of technological advancement, attention is pivotal for individual productivity and organizational efficiency. However, distractions pose significant challenges. An effective attention monitoring system is crucial. GANs (Generative Adversarial Networks) augment medical databases by synthesizing diverse data samples. This paper explores GANs' potential in enhancing datasets for indoor positioning and attention monitoring. The primary aim is to develop a brainwave-controlled lighting system adjusting settings based on users' brainwave states and positions. GANs enhance brainwave and positioning datasets, followed by dimensionality reduction via Principal Components Analysis for improved data processing efficiency. Machine learning and deep learning techniques train classification models on augmented datasets. The system dynamically adjusts lighting parameters, offering a personalized experience. Improved GANs augment datasets, fine-tuned through parameter adjustments. Various machine learning models including NN (Neural Networks), KNN (K-Nearest Neighbors), SVM (Support Vector Machines), and LSTM (Long Short-Term Memory) classify attention and positioning. Fuzzy Logic enhances system stability. Finally, the trained models will be integrated with hardware components such as Raspberry Pi 4 to complete an "Indoor Positioning Deep Learning Attention Monitoring System based on the Grey Wolf Optimizer Algorithm." The experimental results show that this system can achieve a positioning accuracy of 15 centimeters and significantly improve attention monitoring.

摘要...i
Abstract...ii
誌謝...iii
目錄...iv
表目錄...vi
圖目錄...viii
第一章 緒論...1
1.1研究動機與目的...1
1.2相關研究...2
1.3章節安排...5
第二章 文獻回顧...6
2.1腦電波圖（Electroencephalography, EEG）...6
2.2資料集增強（Datasets Augmentation）...7
2.2.1 Generative Adversarial Networks（GANs）...7
2.2.2 Principal Component Analysis（PCA）...8
2.3分類模型（Classification Models）...9
2.3.1 Neural Network（NN）...9
2.3.2 K-Nearest Neighbors（KNN）...11
2.3.3 Support Vector Machine（SVM）...11
2.3.4 Recurrent Neural Network（RNN）...14
2.3.5 Long Short-Term Memory（LSTM）...14
2.4模糊控制（Fuzzy Control）...15
2.4.1 Fuzzification...15
2.4.2 Fuzzy Rule-Base...17
2.4.3 Inference Engine...17
2.4.4 Defuzzification...18
2.5室內定位方法...19
2.5.1訊號傳播通道模型...19
2.5.2 Fingerprint...20
2.5.3 Trilateration...21
2.6灰狼優化演算法（Grey Wolf Optimizer Algorithm, GWO）...22
2.7 Message Queuing Telemetry Transport（MQTT）...23
2.8 Node-RED...25
第三章 室內定位...26
3.1室內空間發射台擺放拓樸...26
3.2第一階段定位...27
3.2.1矩形劃分法...28
3.2.2區域訊號接收強度資料集建立...28
3.2.3指紋法之第一階段定位...31
3.2.4第一階段資料庫增強...34
3.3第二階段定位...42
3.3.1訊號接收強度與距離轉換方法...43
3.3.2灰狼優化演算法目標定位方法...48
3.3.3模糊邏輯處理雜訊干擾方法...51
3.3.4訊號浮動處理...55
第四章 腦波專注度...61
4.1資料收集與標籤處理...61
4.1.1腦波儀...61
4.1.2資料收集...62
4.1.3資料標籤化處理...63
4.2腦波資料庫增強實驗...64
4.2.1原始資料集模型分類...64
4.2.2增強資料集模型分類...66
4.3腦波燈控系統設計...71
4.3.1燈控模糊控制設計...72
4.3.2燈控模糊PD控制設計...78
4.3.3燈控定位系統設計...85
第五章 模擬與實驗結果...94
5.1模擬結果...97
5.1.1定位系統模擬結果...97
5.1.2腦波專注度監控系統模擬結果...103
5.1.3燈控系統模擬結果...105
5.1.4 Local Interpretable Model-agnostic Explanations（LIME）模型分析...106
5.2 實驗結果...109
5.2.1定位系統實驗結果...109
5.2.2腦波專注度監控系統實驗結果...115
5.2.3燈控系統實驗結果...117
5.2.4人機介面設計...118
5.2.5實驗結果討論...119
第六章 結論與未來展望...121
參考文獻...122
附錄一...126
附錄二...128
Extended Abstract...129

[1]Goodfellow, I., Pouget-Abadie, J., Mirza, M., Xu, B., Warde-Farley, D., Ozair, S., Courville, S. and Bengio, Y. “Generative adversarial nets.” Advances in neural information processing systems, 2014, pp. 2672-2680.
[2]A. M. Shaker, M. Tantawi, H. A. Shedeed and M. F. Tolba, “Generalization of Convolutional Neural Networks for ECG Classification Using Generative Adversarial Networks,” IEEE Access, vol. 8, pp. 35592-35605, 2020, doi: 10.1109/ACCESS.2020.2974712.
[3]W. Cao, N. Czarnek, J. Shan and L. Li, “Microaneurysm Detection Using Principal Component Analysis and Machine Learning Methods,” IEEE Transactions on NanoBioscience, vol. 17, no. 3, pp. 191-198, July 2018, doi: 10.1109/TNB.2018.2840084.
[4]Radford, A., Metz, L., and Chintala, S. “Unsupervised representation learning with deep convolutional generative adversarial networks.” arXiv preprint arXiv:1511.06434, 2015.
[5]P. Chakraborty and S. Rafiammal, “EEG based Neurocognitive Metrics for Determination of Concentration Level in Young Adults,” 2020 International Conference on Communication and Signal Processing (ICCSP), Chennai, India, 2020, pp. 0411-0414, doi: 10.1109/ICCSP48568.2020.9182248.
[6]R. M. Mehmood and H. J. Lee, “Emotion classification of EEG brain signal using SVM and KNN,” 2015 IEEE International Conference on Multimedia & Expo Workshops (ICMEW), Turin, Italy, 2015, pp. 1-5, doi: 10.1109/ICMEW.2015.7169786.
[7]N. -D. Mai, N. M. Hoang Long and W. -Y. Chung, “1D-CNN-based BCI system for detecting Emotional states using a Wireless and Wearable 8-channel Custom-designed EEG Headset,” 2021 IEEE International Conference on Flexible and Printable Sensors and Systems (FLEPS), Manchester, United Kingdom, 2021, pp. 1-4, doi: 10.1109/FLEPS51544.2021.9469818.
[8]R. K. Jeevan, V. M. R. S.P., P. Shiva Kumar and M. Srivikas, “EEG-based emotion recognition using LSTM-RNN machine learning algorithm,” 2019 1st International Conference on Innovations in Information and Communication Technology (ICIICT), Chennai, India, 2019, pp. 1-4, doi: 10.1109/ICIICT1.2019.8741506.
[9]Z. Liu, M. Tong, X. Liu, Z. Du and W. Chen, “Research on Extended Image Data Set Based on Deep Convolution Generative Adversarial Network,” 2020 IEEE 4th Information Technology, Networking, Electronic and Automation Control Conference (ITNEC), Chongqing, China, 2020, pp. 47-50, doi: 10.1109/ITNEC48623.2020.9085221.
[10]T. Louis, “A Tutorial on Variational Autoencoders with a Concise Keras Implementation, ” , https://tiao.io/post/tutorial-on-variational-autoencoders-with-a-concise-keras-implementation/, 2022.
[11]S. Chanthakit and C. Rattanapoka, “MQTT Based Air Quality Monitoring System using Node MCU and Node-RED,” 2018 Seventh ICT International Student Project Conference (ICT-ISPC), Nakhonpathom, Thailand, 2018, pp. 1-5, doi: 10.1109/ICT-ISPC.2018.8523891.
[12]Y. Yang et al., “An attention grading of students’ attention in online learning under different light environments,” 2023 7th International Conference on Machine Vision and Information Technology (CMVIT), Xiamen, China, 2023, pp. 149-159, doi: 10.1109/CMVIT57620.2023.00036.
[13]P. D. Tinh and T. T. N. Mai, “Mobile Indoor Positioning System Utilizing WiFi RSSI and Motion Data,” 2020 4th International Conference on Recent Advances in Signal Processing, Telecommunications & Computing (SigTelCom), Hanoi, Vietnam, 2020, pp. 98-102, doi: 10.1109/SigTelCom49868.2020.9199021.
[14]X. Li, Z. Deng, F. Yang, X. Zheng, L. Zhang and Z. Zhou, “WiFi Indoor Location Method Based on RSSI,” 2021 11th IEEE International Conference on Intelligent Data Acquisition and Advanced Computing Systems: Technology and Applications (IDAACS), Cracow, Poland, 2021, pp. 1036-1040, doi: 10.1109/IDAACS53288.2021.9660916.
[15]S. Mirjalili, S. M. Mirjalili, and A. Lewis, “Grey Wolf Optimizer,” Advances in Engineering Software, vol. 69, pp. 46-61, 2014, ISSN 0965-9978, doi: 10.1016/j.advengsoft.2013.12.007.
[16]A. S. Haron, Z. Mansor, I. Ahmad and S. M. M. Maharum, “The Performance of 2.4GHz and 5GHz Wi-Fi Router Placement for Signal Strength Optimization Using Altair WinProp,” 2021 IEEE 7th International Conference on Smart Instrumentation, Measurement and Applications (ICSIMA), Bandung, Indonesia, 2021, pp. 25-29, doi: 10.1109/ICSIMA50015.2021.9526299.
[17]H. Hesham, Y. Ayman and M. Ashour, “Investigating antenna positioning, beam width and direction in indoor DAS systems to meet coverage,” 2019 International Conference on Innovative Trends in Computer Engineering (ITCE), Aswan, Egypt, 2019, pp. 304-309, doi: 10.1109/ITCE.2019.8646410.
[18]R. Zhang et al., “Effect of Indoors Artificial Lighting Conditions on Computer-Based Learning Performance,” International Journal of Environmental Research and Public Health, vol. 17, no. 7, p. 2537, Apr. 2020, doi: 10.3390/ijerph17072537.
[19]R. Pascanu, T. Mikolov, and Y. Bengio, “On the difficulty of training recurrent neural networks,” Proceedings of the 30th International Conference on Machine Learning, vol. 28, no. 3, pp. 1310-1318, 2013.
[20]Y. -M. Huang, Y. -P. Cheng, S. -C. Cheng and Y. -Y. Chen, “Exploring the Correlation Between Attention and Cognitive Load Through Association Rule Mining by Using a Brainwave Sensing Headband,” IEEE Access, vol. 8, pp. 38880-38891, 2020, doi: 10.1109/ACCESS.2020.2975054.
[21]J. Ye, Y. Lin, Z. Li, J. Lee, A. -A. Abdulrahman and M. Jin, “A Non-Invasive Sleep Analysis Approach Based on a Fuzzy Inference System and a Finite State Machine,” IEEE Access, vol. 7, pp. 2664-2676, 2019, doi: 10.1109/ACCESS.2018.2886205.
[22]L. A. Zadeh, “Fuzzy sets,” Information and Control, vol. 8, no. 3, pp. 338-353, Jun. 1965.
[23]S. Hochreiter and J. Schmidhuber, “Long Short-Term Memory,” Neural Computation, vol. 9, no. 8, pp. 1735-1780, 15 Nov. 1997, doi: 10.1162/neco.1997.9.8.1735.
[24]R. Jabeur, N. Ouaaline and A. Lakrim, “A Fuzzy Logic Controller Controls a Smart Lighting System for Energy Savings,” 2021 9th International Renewable and Sustainable Energy Conference (IRSEC), Morocco, 2021, pp. 1-6, doi: 10.1109/IRSEC53969.2021.9741196.
[25]H. Liao, X. Sheng and D. Yang, “Design of adaptive lighting system based on fuzzy PID,” 2018 Chinese Control And Decision Conference (CCDC), Shenyang, China, 2018, pp. 1543-1547, doi: 10.1109/CCDC.2018.8407372.
[26]C. Zhang, L. Wan and Y. Liu, “Ship Heading Control Based on Fuzzy PID Control,” 2019 34rd Youth Academic Annual Conference of Chinese Association of Automation (YAC), Jinzhou, China, 2019, pp. 607-612, doi: 10.1109/YAC.2019.8787601.
[27]J. Z. Shi, “A Fractional Order General Type-2 Fuzzy PID Controller Design Algorithm,” IEEE Access, vol. 8, pp. 52151-52172, 2020, doi: 10.1109/ACCESS.2020.2980686.
[28]C. Contreras and O. Begovich, “Automatic Lighting Control and IoT Monitoring on an Indoor-Greenhouse,” 2022 IEEE 18th International Conference on Automation Science and Engineering (CASE), Mexico City, Mexico, 2022, pp. 1232-1237, doi: 10.1109/CASE49997.2022.9926506.
[29]N. Borhan, I. Saleh, A. Yunus, W. Rahiman, D. Novaliendry and Risfendra, “Reducing UWB Indoor Localization Error Using the Fusion of Kalman Filter with Moving Average Filter,” 2023 IEEE International Conference on Automatic Control and Intelligent Systems (I2CACIS), Shah Alam, Malaysia, 2023, pp. 55-59, doi: 10.1109/I2CACIS57635.2023.10193663.
[30]M. T. Ribeiro, S. Singh, and C. Guestrin, “Why should I trust you? Explaining the predictions of any classifier,” Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 2016.
[31]H. Farhood, M. Saberi and M. Najafi, “Improving Object Recognition in Crime Scenes via Local Interpretable Model-Agnostic Explanations,” 2021 IEEE 25th International Enterprise Distributed Object Computing Workshop (EDOCW), Gold Coast, Australia, 2021, pp. 90-94, doi: 10.1109/EDOCW52865.2021.00037.
[32]M. S. Ahmed et al., “Comparative Analysis of Interpretable Mushroom Classification using Several Machine Learning Models,” 2022 25th International Conference on Computer and Information Technology (ICCIT), Cox's Bazar, Bangladesh, 2022, pp. 31-36, doi: 10.1109/ICCIT57492.2022.10055555.
[33]C.-Y. Liao, R.-C. Chen, and S.-K. Tai, “Evaluating attention level on MOOCs learning based on brainwaves signals analysis,” Int. J. Innov. Comput. Inf. Control, vol. 15, no. 1, pp. 39-51, 2019.
[34]C.-Y. Liao and R.-C. Chen, “Using EEG Brainwaves and Deep Learning Method for Learning Status Classification,” 2018 International Conference on Machine Learning and Cybernetics (ICMLC), Chengdu, China, 2018, pp. 527-532, doi: 10.1109/ICMLC.2018.8527003.
[35]S.-H. Lee, C.-H. Cheng, C.-C. Lin, and Y.-F. Huang, “PSO-Based Target Localization and Tracking in Wireless Sensor Networks,” Electronics, vol. 12, p. 905, 2023. doi: 10.3390/electronics12040905.
[36]C.-H. Cheng and Y. Yan, “Indoor Positioning System for Wireless Sensor Networks Based on Two-Stage Fuzzy Inference,” International Journal of Distributed Sensor Networks, vol. 14, no. 5, 2018. doi: 10.1177/1550147718780649.
[37]C.-Y. Liao, R.-C. Chen, and S.-K. Tai, “Emotion Stress Detection Using EEG Signal and Deep Learning Technologies,” 2018 IEEE International Conference on Applied System Invention (ICASI), Chiba, Japan, 2018, pp. 90-93, doi: 10.1109/ICASI.2018.8394414.
[38]Y. Xu, et al., “Indoor Positioning Algorithm for RFID Based on Newton Interpolation and Support Vector Regression Optimized by Hybridizing Grey Wolf Optimization,” Journal of System Simulation, vol. 29, no. 9, pp. 1921-1929, 2020.
[39]X. Jin, et al., “Hybrid Strategy Improved Grey Wolf Optimization Algorithm for Plume Tracking and Localization Method in Indoor Weak Wind Environment,” IEEE Access, vol. 10, pp. 100976-100986, 2022.
[40]Y. Meng, et al., “A Two‐Stage Wireless Sensor Grey Wolf Optimization Node Location Algorithm Based on K‐Value Collinearity,” Mathematical Problems in Engineering, vol. 2020, no. 1, p. 7217595, 2020.
[41]H. Y. Abuaddous, et al., “Repulsion-Based Grey Wolf Optimizer with Improved Exploration and Exploitation Capabilities to Localize Sensor Nodes in 3D Wireless Sensor Network,” Soft Computing, vol. 27, no. 7, pp. 3869-3885, 2023.