Tag: Academia – Student Project

Very well done everyone!

From Sensors to Edge Intelligence at UMPSA STEM Lab

Senior Design Project 2 marks an important milestone for final-year students — a transition from theory to real-world engineering practice. In UMPSA STEM Lab, this cohort demonstrated not only technical competence, but also maturity in system integration, optimization, and data-driven decision making.

This year’s projects shared a common theme: intelligent sensing at the edge — combining microcontrollers – LilEx 3, FPGA – LilEx 4, wireless communication, and machine learning to solve practical monitoring problems in health and environment domains.

1. FPGA-Based Edge Computing for Environmental Monitoring

Jian Kun’s project focused on bandwidth and energy optimization in wireless sensor networks using FPGA-based edge processing. By integrating a Kalman filter directly on the FPGA, the system was able to smooth sensor signals before transmission, reducing redundant data updates.

Key Results & Discussion (Bandwidth Optimization)

1. Compression Efficiency

   • Standard JSON packet: ~98 bytes per transmission

   • Delta packet: ~59 bytes per transmission

2. Reduction Rate

   • Achieved approximately 39.8% reduction in data payload size during stable environmental conditions

3. Energy Implications

   • Reduced transmission time on the HC-12 module

   • Directly contributes to extended battery life for the wireless sensor node

This project highlights how edge intelligence and signal processing can significantly improve the efficiency of IoT systems, especially in resource-constrained deployments.

2. Real-Time Health Monitoring System with LoRa

Nur Alysa’s project explored machine learning for activity-based heart rate classification, implemented on an embedded platform. She systematically compared CNN, k-NN, and Random Forest models before selecting Random Forest as the best-performing and most suitable for embedded deployment.

Highlights

1. Classified heart rate patterns across:

   • Running

   • Jogging

   • Brisk walking

2. Successfully deployed the trained Random Forest model onto the STEM Cube (Raspberry Pi Pico)

3. Integrated sensing, inference, and wireless transmission using LoRa

This work demonstrates a complete end-to-end edge ML pipeline, from data collection and model evaluation to real-time inference on a microcontroller.

3. Wireless Sensor Network for Environmental Monitoring Using LoRa

Sasrizwan’s project focused on air quality monitoring using LoRa-based wireless sensor networks, with an added layer of intelligence through machine learning classification.

Key Contributions

1. Designed a LoRa-based WSN for environmental data collection

2. Applied ML techniques to classify three different air quality levels

3. Demonstrated how low-power communication and intelligent data processing can coexist in practical deployments

This project emphasized scalable environmental sensing, suitable for smart city and community monitoring applications.

4. FPGA-Based Health Monitoring System with LoRa

Misharienna’s project explored the use of FPGA as the main processing unit for health monitoring applications. Unlike microcontroller-centric designs, this work focused on hardware-level data acquisition and processing.

Key Highlights

1. FPGA used to collect and process physiological sensor data

2. Integrated LoRa for wireless transmission

3. Demonstrated parallel processing advantages of FPGA for real-time health monitoring

This project showcases how reconfigurable computing can play a role in future high-performance wearable or medical monitoring systems.

5. Air Quality Monitoring and Prediction System for Urban Areas

Nur Munirah’s project utilized the Raspberry Pi Pico to develop an air quality monitoring and prediction system tailored for urban environments.

Key Contributions

1. Used sensor data to monitor air quality parameters

2. Applied predictive techniques to estimate air quality trends

3. Demonstrated how low-cost microcontrollers can support data-driven environmental insights

 

Beyond Individual Projects: Shared Learning Outcomes

Across all projects, several common strengths emerged:

1. Students successfully controlled multiple sensors via microcontrollers and FPGA

2. Built real-time dashboards for visualization and monitoring

3. Applied machine learning models on constrained hardware

4. Implemented Kalman filtering and signal smoothing for improved data quality

5. Understood trade-offs between accuracy, bandwidth, power, and computation

More importantly, students learned that engineering is not just about making systems work — it is about making them efficient, reliable, and scalable.

The Senior Design Project 2 experience under UMPSA STEM Lab demonstrates the commitment to hands-on, future-ready engineering education. By engaging with edge AI, FPGA, IoT, and wireless systems, students are not only preparing for industry — they are contributing ideas that align with current research and real-world challenges.

We are proud of this cohort and look forward to seeing these projects evolve into research publications, prototypes, and impactful deployments.

Congratulations to Nur Alysa Nabilla Binti Muhamad Arif for emerging as the winner of the Senior Design Project 2 (SDP2) Competition, a well-deserved recognition of her strong technical execution and successful deployment of machine learning on an embedded platform.

Congratulations to Ooi Jian Kun and Sasrizwan bin Saman, who went beyond the targeted proof of concept, demonstrating deeper system optimization, edge intelligence, and engineering maturity in their respective projects.

Congratulations to Nur Munirah Binti Mohd Yusof and Misharienna Andrea Kuek for their successful implementation of complete monitoring systems, showcasing solid integration of sensing, processing, and wireless communication technologies.

Overall, this cohort has set a high benchmark for future SDP projects under UMPSA STEM Lab, and we look forward to seeing these works grow into impactful research and real-world applications.

Well done to all guys =) !

The UMPSA STEM Lab conducted a Structured Literature Review (SLR) Workshop today aimed at helping final-year students strengthen their understanding and writing of Chapter 2 for their senior design projects. Recognizing that the literature review is the foundation of any research, this session focused on introducing students to a systematic way of collecting, analyzing, and presenting existing research, ensuring their work is both comprehensive and credible.

In research writing, there are generally four main types of reviews: narrative review, systematic review, scoping review, and structured literature review (SLR). Among these, the SLR approach is particularly valuable for engineering and technology-based projects, as it enables researchers to identify research gaps and establish a clear direction for their study based on evidence.

During the workshop, students were guided through several key stages of the SLR process, starting from identifying the right keywords, searching for relevant articles across reputable databases, to clustering and categorizing the collected literature according to research themes. Through hands-on practice, participants learned how to use digital tools to organize their sources efficiently, while maintaining critical analysis throughout their writing.

The session not only strengthened students’ academic writing skills but also encouraged research discipline and analytical thinking, both essential for producing high-quality theses.

The UMPSA STEM Lab remains committed to supporting students through capacity-building workshops like this, ensuring that each research project reflects both academic rigor and innovative spirit.