Essentials Towards Signal Processing and Data Management for AIoT Applications
·
2 min read
Fundamental skills in machine learning and digital signal processing, with applications to real-world problems and research projects.
Course Overview
This course prepares students with foundational knowledge in machine learning and digital signal processing, then guides them to apply these fundamentals to real-world AIoT applications in HCI and health monitoring domains.
Topics Covered
- Embedded systems and sensing technologies
- Digital signal processing (time/frequency domain, filtering, FFT)
- Audio signal processing
- Machine learning fundamentals (classification, regression)
- Deep learning for edge devices
- TinyML and on-device deployment
- Statistical analysis and experimental design
Course Schedule
| Week | Topic | Content |
|---|---|---|
| 1 | Course Introduction | Course overview, AIoT applications in HCI and health monitoring, end-to-end design methodology (sensors + signal processing + ML) |
| 2 | Embedded Systems & Sensing | System components, resistive/capacitive sensing, motion sensors, communication protocols |
| 3 | Signal Processing Fundamentals | Time/frequency domain, filters, FFT, peak detection, sampling. Case: Heart rate from wearable PPG |
| 4 | Audio Signal Processing | Audio concepts, processing pipeline, touch gesture sound analysis. Project brainstorming |
| 5 | Holiday Break | - |
| 6 | Project Proposal Presentations | Application background, technical feasibility, innovation discussion |
| 7 | Machine Learning I | Train/test split, supervised/unsupervised learning, SVM, decision trees, kNN. Case: Flower classification |
| 8 | Machine Learning II | Classification performance evaluation, information visualization |
| 9 | Regression Analysis | Linear regression, curve fitting, gradient boosting, overfitting. Case: Housing price prediction |
| 10 | Deep Learning I | Neural networks, activation functions, convolution, data augmentation. Case: Digit recognition |
| 11 | Deep Learning II | Lightweight networks, transfer learning |
| 12 | Edge Deployment | TinyML, CMSIS_NN, quantization, deployment acceleration. Case: Wake word detection, gesture recognition |
| 13 | Project Progress Presentations | Technical implementation, results demonstration, innovation highlights |
| 14 | Statistical Analysis I | Parametric/non-parametric tests, t-test, ANOVA, Wilcoxon test, SPSS and Python |
| 15 | Statistical Analysis II | Case studies: Height analysis, user feedback analysis, performance evaluation |
| 16 | Final Project Presentations | Complete project demonstration with user study and statistical analysis |
Assignments
- Assignment 1: Self-introduction and smart earbuds innovation report
- Assignment 2: Touch gesture sound signal processing and recognition
- Assignment 3: Efficient recognition algorithm on acoustic event dataset
- Course Project: High social impact AIoT application (team-based)
Authors
Associate Professor (Research Track)
Yuntao Wang’s research centers on physiobehavioral computing and
intelligent interaction for mobile and wearable systems. His work
focuses on (1) developing robust, efficient sensing that performs
reliably on mainstream devices, (2) extracting spatiotemporal patterns
from multimodal signals to infer interaction intent by leveraging
natural behavioral correlations, and (3) designing edge-efficient interfaces
that deliver high performance on mobile and wearable platforms.
He has published 90+ papers, received 10 international conference awards,
and holds 30+ granted patents. His contributions have been
recognized with honors including the Wu Wenjun AI Outstanding Youth
Award (2024), the CAST Young Elite Scientists Sponsorship Program (2022),
the Qinghai High-Level Innovation & Entrepreneurship Leading Talent (2024),
and the First Prize of the China Electronics Institute Science & Technology Award (2019).