Learning in Compressed Domain Development of Light Learning Engine in Fog Computing Environments
PI: 吳安宇教授
計畫簡介:
本計畫以壓縮感知(Compressive Sensing, CS)技術為基礎,進一步開發新世代之壓縮辨識技術(Compressive Analysis, CA)。此CA技術的主要特點是能夠直接在壓縮信號上進行分析與判斷，從而節省了無線傳輸資料量與接收端所需的運算時間和能耗，同時可以對資料提供了一定程度的隱私保密能力(privacy)。最後，本團隊落實該CA技術於心房顫動(Atrial Fibrillation)監測，以及壓縮心電信號(Compressed Electrocardiogram, ECG)之用戶辨識。 壓縮感知(CS)是一個非對稱性的壓縮技術，它能夠簡易的在感測器端對信號進行隨機取樣和即時壓縮，大幅降低感測器的能耗；但付出的代價是其解壓縮的過程需要經過一系列複雜的運算，造成接收端運算資源無法負荷。透過我們新提出之CA技術，我們可以省略解壓縮的步驟直接對壓縮信號進行分析。因此我們可以在大量使用輕量化CS感測器，但同時大幅減少接收端的運算資源。極適合未來5G物聯網之感測運用。
Members
Publications
C. Chou, A. A. Wu, "Low-Complexity Compressive Analysis in Sub-Eigenspace for ECG Telemonitoring System", in ICASSP 2019 - 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp. 7575-7579.
C. Liao, T. Chen and A. Wu, "Real-Time Multi-User Detection Engine Design for IoT Applications via Modified Sparsity Adaptive Matching Pursuit", IEEE Transactions on Circuits and Systems I: Regular Papers, vol. PP, 04 2019, pp. 1-14.
C. Chou et al., "Low-Complexity Privacy-Preserving Compressive Analysis Using Subspace-Based Dictionary for ECG Telemonitoring System", IEEE Transactions on Biomedical Circuits and Systems, vol. 12, no. 4, 2018, pp. 801-811.
K. Hou et al., "Low-Complexity Secure Watermark Encryption for Compressed Sensing-Based Privacy Preserving", in 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp. 6408-6412.
M. Tsai, C. Chou and A. A. Wu, "Robust compressed analysis using subspace-based dictionary for ECG telemonitoring systems", in 2017 IEEE International Workshop on Signal Processing Systems (SiPS), pp. 1-5.
國立臺灣大學
