Metrics
1 Record Views
Thesis
Contactless Material Characterization Through Occlusion Utilizing Consumer-Grade mmWave Radar
University of West Florida Libraries
Master of Science (MS), University of West Florida
2026
Abstract
As robotic systems and autonomous environments require increasingly robust perception capabilities, traditional optical sensors often fail in non-line-of-sight or occluded conditions. While millimeter-wave (mmWave) radar offers a penetrative alternative, the boundary limits of low-cost, consumer-grade mmWave systems for classifying object materials through secondary physical masks remain largely undefined.This thesis investigates the operational capabilities and physical limits of a Texas Instruments IWR6843ISK Frequency Modulated Continuous Wave (FMCW) radar in detecting and categorizing concealed objects. To isolate physical limitations, the sensor's radio frequency (RF) transmission parameters were held constant while environmental variables including masking material density, mask thickness, object distance, and dataset size were systematically manipulated across six testing phases. Captured multi-antenna radar reflections were processed as standardized time-domain signals and evaluated using neural network classifiers. The results demonstrate the system's high baseline accuracy in classifying metallic and non-metallic targets through varied occlusions, including fabric, styrofoam, and wood. However, systematic limit testing revealed distinct operational thresholds, showing significant signal attenuation and subsequent classification degradation as the masking material's density and thickness increased. Additionally, a comparative evaluation of deep learning architectures demonstrated that while a Convolutional Neural Network (CNN) achieved the highest overall accuracy, a Deep Multi-Layer Perceptron (MLP) offered highly competitive performance with vastly superior computational efficiency and ultra-low inference latency. Ultimately, this research validates the feasibility of utilizing consumer-grade mmWave radar as a non-invasive, contactless alternative to traditional induction-based metal detectors, while successfully defining both the physical and computational constraints necessary for its practical implementation.
Details
- Title
- Contactless Material Characterization Through Occlusion Utilizing Consumer-Grade mmWave Radar
- Resource Type
- Thesis
- Contributors
- Tarek Youssef (Committee Chair)Yazan Alqudah (Committee Member)Thomas Gilbar (Committee Member)
- Publisher
- University of West Florida Libraries
- Format
- pdf
- Number of pages
- 62
- Copyright
- Permission granted to the University of West Florida Libraries by the author to digitize and/or display this information for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires the permission of the copyright holder.
- Identifiers
- 99381757108706600
- Academic Unit
- Dr. Muhammad Harunur Rashid Department of Electrical and Computer Engineering; Hal Marcus College of Science and Engineering
- Language
- English
- Awarding Institution
- University of West Florida; Master of Science (MS)
- Theses and Dissertations
- Master of Science (MS), University of West Florida