Hamza Ali Imran
Get in touch
PhD Researcher · Sensors · AIoT · Open to postdoctoral and industrial opportunities

Gas sensors, impedance spectroscopy, wearable sensing, and AI for real-world intelligent systems.

I am a Marie Skłodowska-Curie doctoral researcher at Saarland University, working with the Lab for Measurement Technology and collaborating with Bosch Sensortec. My current research focuses on metal-oxide gas sensors, electrical impedance spectroscopy, and machine learning for sensing, with applications in food freshness monitoring and compact deployable sensor systems.

Alongside my current sensor research, I have a strong publication record in wearable sensing, human activity recognition, emotion recognition from gait, and edge AI. I am especially interested in postdoctoral and industrial roles that sit at the intersection of sensing hardware, intelligent signal processing, and applied machine learning.

Email Google Scholar ResearchGate LinkedIn GitHub
PhD
Marie Curie doctoral researcher in Germany
Sensors + AI
From impedance modeling to edge-ready learning systems
Industry + Academia
Experience spanning Bosch collaboration and product-focused engineering
Portrait of Hamza Ali Imran

About

I received my BS in Electrical Engineering from FAST-NUCES and my MS in Computer Science from NUST-SEECS. I am currently pursuing my PhD in Systems Engineering at Saarland University. My doctoral work investigates how temperature- and frequency-tuned electrical impedance spectroscopy, model-based feature extraction, and machine learning can improve the robustness and practicality of MOS gas sensing.

Before starting my PhD, I spent more than five years at Emumba, where I worked across embedded systems, QA automation, and AI-oriented product development. That industry background still shapes how I approach research today: I am interested not only in achieving strong experimental results, but also in building sensing and inference pipelines that are compact, interpretable, and realistic for deployment.

In parallel to gas sensing, I have worked extensively on wearable inertial sensing for human activity recognition, affective computing, sleep monitoring, and Parkinson's symptom analysis. This combination of sensor systems, signal understanding, and applied AI defines the core profile I bring in.

Education

  • PhD
    Systems Engineering
    Saarland University, Saarbrücken, Germany
    Currently enrolled
  • MS
    Computer Science
    NUST-SEECS, Islamabad, Pakistan
  • BS
    Electrical Engineering
    FAST-NUCES, Islamabad, Pakistan

Research Focus

Gas Sensing & Impedance Spectroscopy

Electrical impedance spectroscopy for MOS gas sensors, sparse-frequency excitation, frequency-domain analysis, humidity effects, and deployable sensing strategies for food freshness and gas classification.

Wearable Sensing & Digital Health

Human activity recognition, emotion recognition from gait, actigraphy-based sleep monitoring, Parkinson's symptom analysis, and edge-oriented AI for inertial sensor systems.

AI for Sensing Systems

Signal modeling, feature engineering, transfer learning, compact deep learning, explainability, and practical ML pipelines that connect measurement science with robust downstream inference.

Selected Highlights

  • Marie Skłodowska-Curie doctoral researcher working on advanced EIS for MOS gas sensors.
  • Published across wearable AI, HAR, emotion recognition, and sensing-oriented machine learning.
  • Built expertise spanning sensor hardware, embedded systems, Python-based analysis, and deep learning.
  • Strong industry experience from Emumba and collaborative research exposure with Bosch Sensortec.
  • Interested in industial and postdoctoral roles bridging intelligent sensing, digital health, and AIoT systems.
  • Comfortable working across experimentation, modeling, publication writing, and applied prototyping.

Experience

Lab for Measurement Technology (LMT), Saarland University

Nov 2024 – Feb 2026

EU Researcher (Doktorand) — PhD research on excitation strategies, impedance modeling, and AI-supported analysis for MOS gas sensors targeting robust food freshness monitoring.

Bosch Sensortec GmbH

May 2024 – Oct 2024 / Feb 2026 - Present

Developing and evaluating frequency- temperature-tuned excitation schemes to enhance gas sensor sensitivity and selectivity, focusing on efficient Electrical Impedance Spectroscopy through sparse frequency sampling and information-preserving excitation strategies.

Emumba Pvt. Ltd.

Dec 2018 – Apr 2024
  • ML / Gen-AI Engineer: Built LLM-based chatbots and RAG pipelines; reproduced and adapted ML models from literature for backend applications.
  • QA Automation Engineer: Automated Python-based GUI testing; trained 5+ engineers in Python, Linux, and Git; supported technical hiring.
  • Embedded / Design Engineer: Developed C/C++ firmware for MCUs and SBCs; built custom embedded Linux (Yocto) images and rapid IoT prototypes.

Selected Publications

Journal Articles

Title Authors Venue / Year Links
From Steps to Sentiments: Cross-Domain Transfer Learning for Activity-Based Emotion Detection in Wearable IoT Systems Hamza Ali Imran, Qaiser Riaz, Kiran Hamza, Shaida Muhammad, Björn Krüger IEEE Internet of Things Journal, 2026 Online PDF
Generisch-Net: A Generic Deep Model for Analyzing Human Motion with Wearable Sensors in the Internet of Health Things Kiran Hamza, Qaiser Riaz, Hamza Ali Imran, Mehdi Hussain, Björn Krüger Sensors, 2024 Online PDF
Smart-Wearable Sensors and CNN-BiGRU Model: A Powerful Combination for Human Activity Recognition Hamza Ali Imran, Qaiser Riaz, Mehdi Hussain, Hasan Tahir, Razi Arshad IEEE Sensors Journal, 2023 Online PDF
Machines Perceive Emotions: Identifying Affective States from Human Gait Using On-Body Smart Devices Hamza Ali Imran, Qaiser Riaz, Muhammad Zeeshan, Mehdi Hussain, Razi Arshad Applied Sciences, 2023 Online PDF
Khail-Net: A Shallow Convolutional Neural Network for Recognizing Sports Activities Using Wearable Inertial Sensors Hamza Ali Imran IEEE Sensors Letters, 2022 Online PDF
UltaNet: An Antithesis Neural Network for Recognizing Human Activity Using Inertial Sensors Signals Hamza Ali Imran IEEE Sensors Letters, 2022 Online

Conference

  • Optimizing AC Excitation Frequency for Linear Complex Impedance Response in MOS Gas Sensing
    Hamza Ali Imran, Oliver Brieger, Christian Bur, Andreas Schütze — Eurosensors 2025 (to appear)
  • Spak-Net: A Lightweight Convolutional Neural Network for Activity Recognition with Wearable Inertial Sensors
    Shaida Muhammad, Hamza Ali Imran, Wasim Zaman, Kiran Hamza — ComTech 2025 · Online · PDF
  • Metal Oxide Semiconductor Gas Sensor Systems for Food Freshness Detection: A Review of Recent Studies
    Hamza Ali Imran, Bora Ersöz, Richard Fix, Christian Bur, Andreas Schütze — 17th Dresdner Sensor‑Symposium, 2024 · Online
  • Mukhtasir-khail-net: An ultra-efficient convolutional neural network for sports activity recognition with wearable inertial sensors
    Hamza Ali Imran, Shaida Muhammad, Saad Wazir, Ataul Aziz Ikram, Obaidullah Arshad — IEEE ICoDT2, 2024 · Online
  • Enhanced human activity recognition using inertial sensor data from smart wearables: A neural network approach with residual connections
    Hamza Ali Imran, Shaida Muhammad, Saad Wazir, Ataul Aziz Ikram, Obaidullah Arshad — IEEE ICECT, 2024 · Online
  • Paying Attention to Human Activities: Recognizing Daily Activities with an Attention-based Neural Network and Inertial Sensors
    Shaida Muhammad, Kiran Hamza, Hamza Ali Imran, Ataul Aziz Ikram, Saad Wazir — IEEE ICIC, 2024 · Online
  • Edgeharnet: An edge-friendly shallow convolutional neural network for recognizing human activities using embedded inertial sensors of smart-wearables
    Hamza Ali Imran, Ataul Aziz Ikram, Saad Wazir, Kiran Hamza — IEEE C‑CODE, 2023 · Online
  • HARResNext: An efficient resnext inspired network for human activity recognition with inertial sensors
    Hamza Ali Imran, Kiran Hamza, Zubair Mehmood — IEEE ICoDT2, 2022 · Online
  • IoT based Smart Fan Dimmer with suppressed Humming Sound and Nonlinear Effect of Inverter
    Ataul Aziz Ikram, Hamza Ali Imran, Ahmed Jamal Ikram, Kiran Hamza, Khawaja Usman Riaz Sehgal — IEEE ICOSST, 2021 · Online
  • HHARNet: Taking inspiration from inception and dense networks for human activity recognition using inertial sensors
    Hamza Ali Imran, Usama Latif — IEEE HONET, 2020 · Online
  • HARDenseNet: A 1D densenet inspired convolutional neural network for human activity recognition with inertial sensors
    Kiran Mehmood, Hamza Ali Imran, Usama Latif — IEEE INMIC, 2020 · Online
  • Multi‑cloud: A Comprehensive Review
    Hamza Ali Imran, Usama Latif, Ataul Aziz Ikram, Maryam Ehsan, Ahmed Jamal Ikram, Waleed Ahmad Khan, Saad Wazir — IEEE INMIC, 2020 · Online
  • HPC as a Service: A naïve model
    Hamza Ali Imran, Usama Latif, Saad Wazir, Ahmed Jamal Ikram, Ataul Aziz Ikram, Hanif Ullah, Maryam Ehsan — IEEE ICICT, 2019 · Online

Selected Projects

Temperature- and Frequency-Tuned EIS for MOS Gas Sensors

Research on extracting meaningful and compact information from frequency-domain impedance measurements of MOS gas sensors, with emphasis on sparse excitation, model-based analysis, and deployable sensing strategies for food freshness monitoring.

Keywords: EIS · MOS Sensors · Sparse Sensing · Food Freshness · Signal Modeling

AIoT-Enabled Sleep Monitoring with Actigraphy

Developed lightweight time-series models for sleep-wake detection using wrist-worn inertial sensing and large-scale actigraphy data, targeting practical wearable health monitoring systems.

Keywords: Actigraphy · Edge AI · Transformers · Sleep Monitoring · Wearables

AIoT-Based Parkinson's Symptom Intelligence

Designed transfer-learning and explainability-driven pipelines for wearable analysis of Parkinsonian motion patterns, with a focus on robust representation learning from inertial sensor data.

Keywords: Inertial Sensors · Transfer Learning · XAI · Digital Health

Cross-Domain Emotion Recognition from Gait

Proposed cross-domain transfer learning for emotion recognition using wearable motion sensors, showing that movement priors learned from human activity datasets can improve data-scarce affective computing tasks.

View Publication

Keywords: Wearable AI · Transfer Learning · Emotion Recognition · Edge Intelligence

Skills

AI, ML & Data Analysis

  • Python, NumPy, Pandas, Matplotlib
  • TensorFlow/Keras, scikit-learn
  • Feature engineering, PCA, model evaluation
  • Time-series learning and transfer learning

Sensors, Embedded & IoT

  • Gas sensors, impedance spectroscopy, sensor characterization
  • ESP32, Arduino, FreeRTOS, firmware in C/C++
  • Embedded prototyping and instrumentation workflows
  • Edge AI and AIoT-oriented deployment thinking

Research & Engineering Tools

  • Linux, Git, VS Code
  • LaTeX, scientific writing, visualization
  • Experiment design and publication preparation
  • Interdisciplinary academic-industry collaboration

Awards & Grants

Contact

I am open to collaborations, postdoctoral opportunities, and industry roles related to sensors, intelligent systems, wearable health technologies, and applied AI.

Email Google Scholar ResearchGate LinkedIn GitHub