請用此 Handle URI 來引用此文件：
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/101697完整後設資料紀錄
| DC 欄位 | 值 | 語言 |
|---|---|---|
| dc.contributor.advisor | 施吉昇 | zh_TW |
| dc.contributor.advisor | Chi-Sheng Shih | en |
| dc.contributor.author | 周君哲 | zh_TW |
| dc.contributor.author | Jyun-Jhe Chou | en |
| dc.date.accessioned | 2026-02-26T16:48:01Z | - |
| dc.date.available | 2026-02-27 | - |
| dc.date.copyright | 2026-02-26 | - |
| dc.date.issued | 2025 | - |
| dc.date.submitted | 2026-01-29 | - |
| dc.identifier.citation | “Infrared Array Sensor Grid-EYE (AMG88) Datasheet.” [Online]. Available: https://datasheet.octopart.com/AMG8833-Panasonic-datasheet-62338626.pdf
“FLIR Lepton® 3.5 Datasheet.” [Online]. Available: https://mm.digikey.com/Volume0/opasdata/d220001/medias/docus/5491/Lepton_Series_7-28-23.pdf S. Studenski, S. Perera, K. Patel, C. Rosano, K. Faulkner, M. Inzitari, J. Brach, J. Chandler, P. Cawthon, E. B. Connor et al., “Gait speed and survival in older adults,” Jama, vol. 305, no. 1, pp. 50–58, 2011. A. Rezaei, M. C. Stevens, A. Argha, A. Mascheroni, A. Puiatti, and N. H. Lovell, “An unobtrusive human activity recognition system using low resolution thermal sensors, machine and deep learning,” IEEE Transactions on Biomedical Engineering, vol. 70, no. 1, pp. 115–124, 2022. S. Singh and B. Aksanli, “Non-intrusive presence detection and position tracking for multiple people using low-resolution thermal sensors,” Journal of Sensor and Actuator Networks, vol. 8, no. 3, p. 40, 2019. Y.-D. Lee and W.-Y. Chung, “Wireless sensor network based wearable smart shirt for ubiquitous health and activity monitoring,” Sensors and Actuators B: Chemical, vol. 140, no. 2, pp. 390–395, 2009. R. Gravina, C. Ma, P. Pace, G. Aloi, W. Russo, W. Li, and G. Fortino, “Cloudbased activity-aaservice cyber–physical framework for human activity monitoring in mobility,” Future Generation Computer Systems, vol. 75, pp. 158–171, 2017. G. Fortino, S. Galzarano, R. Gravina, and W. Li, “A framework for collaborative computing and multi-sensor data fusion in body sensor networks,” Information Fusion, vol. 22, pp. 50–70, 2015. M. Alizadeh, G. Shaker, J. C. M. De Almeida, P. P. Morita, and S. Safavi-Naeini, “Remote monitoring of human vital signs using mm-wave FMCW radar,” IEEE Access, vol. 7, pp. 54 958–54 968, 2019. F. Lemic, J. Martin, C. Yarp, D. Chan, V. Handziski, R. Brodersen, G. Fettweis, A. Wolisz, and J. Wawrzynek, “Localization as a feature of mmWave communication,” in 2016 international wireless communications and mobile computing conference (IWCMC). IEEE, 2016, pp. 1033–1038. P. Zhao, C. X. Lu, J. Wang, C. Chen, W. Wang, N. Trigoni, and A. Markham, “mid: Tracking and identifying people with millimeter wave radar,” in 2019 15th International Conference on Distributed Computing in Sensor Systems (DCOSS). IEEE, 2019, pp. 33–40. T. G. Reddy, S. C. Sai, B. P. Kumar, R. T. Venkatesh, K. Sathwik, and K. Singh, “Face recognition door lock system using Raspberry Pi,” in 2023 Third International Conference on Secure Cyber Computing and Communication (ICSCCC), 2023, pp. 218–221. J. Zhang, Z. Wang, and N. Verma, “In-memory computation of a machine-learning classifier in a standard 6T SRAM array,” IEEE Journal of Solid-State Circuits, vol. 52, no. 4, pp. 915–924, 2017. Z. Chen, X. Chen, and J. Gu, “15.3 a 65nm 3T dynamic analog RAM-based computing-in-memory macro and CNN accelerator with retention enhancement, adaptive analog sparsity and 44TOPS/W system energy efficiency,” in 2021 IEEE International Solid-State Circuits Conference (ISSCC), vol. 64. IEEE, 2021, pp. 240–242. Y.-T. Hsu, C.-Y. Yao, T.-Y. Wu, T.-D. Chiueh, and T.-T. Liu, “A highthroughput energy–area-efficient computing-in-memory SRAM using unified charge-processing network,” IEEE Solid-State Circuits Letters, vol. 4, pp. 146–149, 2021. A. Biswas and A. P. Chandrakasan, “CONV-SRAM: An energy-efficient SRAM with in-memory dot-product computation for low-power convolutional neural networks,” IEEE Journal of Solid-State Circuits, vol. 54, no. 1, pp. 217–230, 2018. Y.-C. Chiu, Z. Zhang, J.-J. Chen, X. Si, R. Liu, Y.-N. Tu, J.-W. Su, W.-H. Huang, J.-H. Wang, W.-C. Wei et al., “A 4-Kb 1-to-8-bit configurable 6T SRAM-based computation-in-memory unit-macro for CNN-based AI edge processors,” IEEE Journal of Solid-State Circuits, vol. 55, no. 10, pp. 2790–2801, 2020. M. Kang, S. K. Gonugondla, A. Patil, and N. R. Shanbhag, “A multi-functional inmemory inference processor using a standard 6T SRAM array,” IEEE Journal of Solid-State Circuits, vol. 53, no. 2, pp. 642–655, 2018. C.-Y. Yao, T.-Y. Wu, H.-C. Liang, Y.-K. Chen, and T.-T. Liu, “A fully bit-flexible computation in memory macro using multi-functional computing bit cell and embedded input sparsity sensing,” IEEE Journal of Solid-State Circuits, 2023. J.-J. Chou, T.-W. Chang, X.-Y. Liu, T.-Y. Wu, Y.-K. Chen, Y.-T. Hsu, C.-W. Chen, T.-T. Liu, and C.-S. Shih, “CIM-based smart pose detection sensors,” Sensors, vol. 22, no. 9, 2022. [Online]. Available: https://www.mdpi.com/1424-8220/22/9/3491 J. Shermeyer and A. Van Etten, “The effects of super-resolution on object detection performance in satellite imagery,” in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, 2019, pp. 0–0. C.-S. Shih, Y.-T. Wang, and J.-J. Chou, “Multiple-image super-resolution for networked extremely low-resolution thermal sensor array,” in 2020 IEEE Second Workshop on Machine Learning on Edge in Sensor Systems (SenSys-ML). IEEE, 2020, pp. 1–6. J.-J. Chou and C.-S. Shih, “Turning over detecting using super-resolution thermal image,” in The 26th IEEE International Conference on Embedded and Real-Time Computing Systems and Applications, Virtual Conference, Aug. 2020, pp. 1–2. F. Almasri and O. Debeir, “RGB guided thermal super-resolution enhancement.” in Cloudtech, 2018, pp. 1–5. C.-S. Shih, J.-J. Chou, C.-C. Chuang, T.-Y. Wang, Z.-Y. Chuang, K.-J. Lin, W.-D. Wang, and K.-C. Huang, “Collaborative sensing for privacy preserving gait tracking using IoT middleware,” in 2017 International Conference on Research in Adaptive and Convergent Systems (RACS 2017). Krakow, Poland: ACM, September 2017. C. Dong, C. C. Loy, and X. Tang, “Accelerating the super-resolution convolutional neural network,” in European conference on computer vision. Springer, 2016, pp. 391–407. J. L. Elman, “Finding structure in time,” Cognitive science, vol. 14, no. 2, pp. 179–211, 1990. | - |
| dc.identifier.uri | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/101697 | - |
| dc.description.abstract | 隨著近年 IoT 裝置大量的應用於各類的地方，到 2024 年已經有約 200 億個 IoT 裝置，且這個數量每 6 到 7 年倍增的速度增長。
資料的收集與上傳會花費大量的電力資源。然而這類感測器通常使用電池作為供電並且僅有極少的運算能力，而仰賴智慧型手機或是個人電腦去收集偵測後的資料並從中提取資訊。這會需要耗費相當大的網路頻寬及能源用於傳遞資料串流。但是受限於有限的運算資源，要在感測器端去處理龐大的資料也非常的困難且會使得電池充電的頻率大幅上升，造成使用上的不便。 本論文中討論了不同計算平台與場景下會遇到的挑戰與解決方式。在使用多組感測器做定位系統時會面臨在感測邊緣時，感測器的定位不一致的誤差。使用類比的計算記憶體平台時，受限於輸出精度及非理想的誤差，模型的訓練會需要提出大量的修改。數位計算記憶體平台，提供了整數的矩陣乘法計算，與額外的非線性計算功能，可以用於完成模型層與層之間的計算需求，而需要微調模型來符合不同的計算模式。 本論文提出的日常健康監測系統中使用的極低解析度感應器去減少感測時的能耗，並且使用多種運算裝置分散其運算工作來達到低耗能的即時事件偵測系統。將資料於感測器端計算並僅上傳辨識結果可以大幅降低所需的網路頻寬，然而在缺乏跨感測器之間資料的情況下，辨識的結果容易有死角，而需要結合各別感測器的部分辨識依據來權衡各感測器結果的可信度與比重。低解析度感測器相較於高解析度感測器，裝置成本與耗電量都較低，在使用上若能搭配合適的人工智慧模型或是結合多個低解析度感測器，可以達到與直接使用高解析度感測器相近的結果。在部分情境下，需要從連續的資料中辨識動作變化的時序事件，此類事件由於需處理複數影像資料，所以需要較大的計算資源，而計算記憶體平台可以在感測器端提供低耗能的計算能力，但是需針對其不同的計算模式微調人工智慧模型。 在偵測步行速度時，在兩個感測器重疊的感測範圍可以減少步行速度的誤差至42.7%。在動作辨識的研究中，使用極低解析度感測器可以只使用高解析度感測器四成的成本與耗電量做到其97%相近的準確度。所提出使用數位計算記憶體平台的翻身偵測系統，相比於其他物聯網平台僅需6.3%的耗電量，且辨識結果有93.9%的近似度。 | zh_TW |
| dc.description.abstract | With the widespread use of Internet of Things (IoT) devices, their number reached 20 billion in 2024, doubling every six to seven years. Collecting and uploading data consumes significant energy and network bandwidth. However, these sensors are typically powered by batteries and employ smartphones or personal computers to process data. Continuously transmitting data quickly drains the battery. Data transmission requires a substantial network bandwidth when the number of sensors or their resolutions increase. However, processing high-resolution data on edge devices is challenging due to limited computational resources. Computing-in-memory (CIM) technology offers a highly energy-efficient computing platform for edge devices to process sensing data in real time. This work explores three computing platforms and optimizes energy consumption and sensing accuracy. When multiple sensors are employed in a location tracking system, inconsistencies in sensor-positioning errors arise at the sensing edge. Analog CIM offers a highly energy- and area-efficient computing platform but is constrained by its output precision and nonideal errors. Digital CIM offers an integer matrix-multiplication computing platform with nonlinear functions that perform calculations between model layers, requiring fine-tuning to accommodate various computational functions. This work proposes ultralow-resolution sensors with CIM for computation. The network bandwidth and energy consumption for data transfer can be significantly reduced by processing the data at the sensor. However, without fusing data from multiple nearby sensors, the results may have blind spots or inconsistencies between neighboring sensors. Therefore, the detection results and evidence could reduce errors. Low-resolution sensors have lower power consumption and cost than high-resolution sensors. Fusing low-resolution data and using suitable artificial intelligence (AI) models can realize performance similar to that of a high-resolution sensor but with better cost efficiency. When recognizing temporal events of posture change from continuous data, more computing resources are necessary to process the streaming data. The CIM platform provides highly energy-efficient computing resources; however, AI models must be fine-tuned based on the hardware design. Fusing the detection results on the overlapping area of two sensors reduced the gait velocity error to 42.7%. The accuracy of posture recognition using ultralow-resolution sensors achieved 97% of that of a high-resolution sensor at only 40% of the cost and energy consumption. The proposed turnover detection system with a digital CIM platform consumes only 6.3% of energy compared to other IoT platforms while achieving 93.9% accuracy. | en |
| dc.description.provenance | Submitted by admin ntu (admin@lib.ntu.edu.tw) on 2026-02-26T16:48:01Z No. of bitstreams: 0 | en |
| dc.description.provenance | Made available in DSpace on 2026-02-26T16:48:01Z (GMT). No. of bitstreams: 0 | en |
| dc.description.tableofcontents | 口試委員審定書 i
誌謝 iii 摘要 v Abstract vii Contents xi List of Figures xv List of Tables xvii Chapter 1 Introduction 1 Chapter 2 Background and Related Work 5 2.1 Healthcare Systems 5 2.2 Thermal Array Sensors 7 2.2.1 Panasonic Grid-EYE Thermal Sensor 7 2.2.2 FLIR Lepton 3.5 8 2.3 Computing in Memory 8 2.3.1 Static Random Access Memory Computing in Memory 10 Chapter 3 Low-power Consumption Platforms 15 3.1 Device for Data Collection and System Overview 16 3.2 Sensor Fusion on Microcontroller Unit Platform 17 3.3 Ultralow-Resolution Sensing on Analog Computing-in-Memory Platforms 17 3.4 Ultralow-Resolution Sensing on Digital Computing-in-Memory Platforms 19 3.5 Computing-in-Memory-Friendly Model Training 21 3.6 Comparison and Summary 21 Chapter 4 Sensor Fusion on the Microcontroller Unit Platform 23 4.1 Challenge of Gait Velocity Detection 23 4.2 Single Sensor Location Detection 24 4.3 Multisensor Gait Velocity Fusing 25 4.4 Result Evaluation 29 Chapter 5 Ultralow-Resolution Sensing on Analog Computing-in-Memory Platforms 33 5.1 Posture Detection Challenge on Edge Devices 33 5.2 Data Collection Devices 34 5.3 Static Posture Recognition 35 5.3.1 Resolution Enhance Model 37 5.3.2 Posture Recognition Model 38 5.4 Analog Computing-in-Memory-Friendly Model Training 39 5.4.1 Nonlinear Errors on Analog Computing-in-Memory Devices 41 5.4.2 Uncertainty-Aware Convolutional Neural Network Training 43 5.4.3 Computing-in-Memory-Friendly Posture Recognition Model 44 5.4.4 Analog Computing-in-Memory-Friendly Model Training 47 5.4.4.1 Activation Function: ReLU vs. LeakyReLU 48 5.4.4.2 Customized Normalization Function 48 5.4.4.3 Error-Mapping Model 49 5.4.4.4 Training with Gaussian Noise 50 5.4.4.5 Computing-in-Memory-Friendly Model Results 52 5.4.5 Energy Consumption 54 Chapter 6 Ultralow-Resolution Sensing on Digital Computing-in-Memory Platforms 57 6.1 Turnover Detection Challenge 57 6.2 Turnover Dataset 58 6.2.1 Temperature Trend Method 59 6.2.2 Neural Network Method 61 6.3 Turnover Detection on Digital CIM 62 6.3.1 Integer CNN Model 64 6.3.2 Post-Training CNN Model Fine-tuning 68 6.3.3 Detection Result 71 6.3.3.1 Computing-in-Memory-Friendly ResNet-18 71 6.3.3.2 Accuracy for Turnover Detection 72 6.3.3.3 Power Consumption and Computation Latency 74 Chapter 7 Conclusion 79 References 81 | - |
| dc.language.iso | en | - |
| dc.subject | 嵌入式系統 | - |
| dc.subject | 機器學習 | - |
| dc.subject | 記憶體內計算 | - |
| dc.subject | 照護資訊系統 | - |
| dc.subject | Embedded systems | - |
| dc.subject | Machine learning | - |
| dc.subject | Computing in memory | - |
| dc.subject | Health care information systems | - |
| dc.title | 設計可偵測時序性姿態變化之低耗能智慧感測系統 | zh_TW |
| dc.title | Design of a Low Power Consumption Smart Sensing System for Detecting Sequential Posture Changes | en |
| dc.type | Thesis | - |
| dc.date.schoolyear | 114-1 | - |
| dc.description.degree | 博士 | - |
| dc.contributor.oralexamcommittee | 梁文耀;曹孝櫟;劉宗德;洪士灝 | zh_TW |
| dc.contributor.oralexamcommittee | Wen-Yew Liang;Shiao-Li Tsao;Tsung-Te Liu;Shih-Hao Hung | en |
| dc.subject.keyword | 嵌入式系統,機器學習記憶體內計算照護資訊系統 | zh_TW |
| dc.subject.keyword | Embedded systems,Machine learningComputing in memoryHealth care information systems | en |
| dc.relation.page | 83 | - |
| dc.identifier.doi | 10.6342/NTU202504512 | - |
| dc.rights.note | 同意授權(全球公開) | - |
| dc.date.accepted | 2026-02-02 | - |
| dc.contributor.author-college | 電機資訊學院 | - |
| dc.contributor.author-dept | 資訊工程學系 | - |
| dc.date.embargo-lift | 2031-01-29 | - |
| 顯示於系所單位： | 資訊工程學系 | |
文件中的檔案：
| 檔案 | 大小 | 格式 | |
|---|---|---|---|
| ntu-114-1.pdf 此日期後於網路公開 2031-01-29 | 5.09 MB | Adobe PDF |
系統中的文件，除了特別指名其著作權條款之外，均受到著作權保護，並且保留所有的權利。