Abstract
Through-the-wall radar imaging (TWRI) is an essential technology for military and rescue applications; however, its performance in detecting and visualizing high-quality images of targets behind walls is significantly degraded by multipath reflections and signal attenuation. This paper reviews the current state of TWRI and its challenges, and explores the transformative potential of deep learning, particularly convolutional neural networks (CNNs), in addressing these challenges. Peer-reviewed articles published from 2018 to 2024 were analysed to examine CNN applications in addressing TWRI challenges. The analysis reveals that using CNNs, TWRI systems can be more effective by filtering wall distortions, reducing noise, lowering computational demands, and improving overall target detection and image quality. Empirical evidence from previous studies highlights this potential. For example, a CNN-based multi-hop residual model has been shown to iteratively refine TWRI images and enhance target detection, with 99% accuracy while using only 10% of the training data, demonstrating exceptional robustness in data-scarce environments. However, challenges persist, including large data requirements and high computational demands. Addressing these challenges could further improve the accuracy and reliability of TWRI systems. This review emphasizes the transformative potential of CNNs while identifying critical gaps and future research directions for optimizing CNN performance in TWRI. In particular, there is a need to develop real-time processing algorithms, enhance object classification, advance transfer learning and object adaptation techniques, improve robustness of image reconstruction models, and improve simulation and synthetic data generation. The review further highlights the importance of utilizing advancements in edge computing technologies and addressing the ethical implications associated with TWRI applications. Other opportunities lie in human activity recognition, collaborative and swarm radars, and Internet of Things integration. Overall, integrating CNNs into TWRI presents a significant opportunity to revolutionize its capabilities, particularly in challenging and sensitive environments.
Recommended Citation
Edgar, T., & Abdalla, A. T. (2026). Leveraging Convolutional Neural Networks for Through-the-Wall Radar Imaging: Challenges, Impacts, and Future Directions. Tanzania Journal of Engineering and Technology, 45(1), 1-11. https://doi.org/https://doi.org/10.65085/2619-8789.1104
Publisher Name
University of Dar es Salaam
Included in
Artificial Intelligence and Robotics Commons, Computer-Aided Engineering and Design Commons, Cybersecurity Commons, Databases and Information Systems Commons, Systems and Communications Commons