Enhancing Automotive Safety with Low-Cost, High-Resolution MIMO Tomographic SAR: A Multi-Frequency Approach

This paper presents a comprehensive study of Multiple-Input Multiple-Output (MIMO) Tomographic Synthetic Aperture Radar (TSAR) and its application in the automotive industry for enhancing active safety features. We explore the utilization of MIMO TSAR in capturing 4D environmental data (range, distance, speed, and material properties), essential for object detection, recognition, and high-resolution imaging. The cornerstone of our approach lies in exploiting the geometric diversity inherent in radar imaging. By deploying transmitters and receivers in multiple states, we enhance the information richness in the measurement domain. This configuration allows for a more detailed and accurate reconstruction of the environment, crucial for automotive applications such as parking assistance and collision avoidance. Our MIMO TSAR system reconstructs images by solving the inverse problem of the linear relationship between the target’s contrast function and the measured scattering field under the first-order Born approximation. This method is resolution independent from bandwidth frequency, allowing us to operate imaging through a single frequency. We propose a novel technique that starts with creating low-cost, low-resolution tomographic images using a single frequency, followed by synthesizing these into a high-resolution multiband image. This approach significantly speeds up the TSAR reconstruction process, making it more suitable for real-time automotive applications. Furthermore, we introduce an adaptive super-resolution method to reconstruct 3D Radio Frequency tomographic images in the X-band. This method is especially beneficial in the automotive field, where higher resolution images are desirable but often restricted by bandwidth and power regulations. Our computational experiments, demonstrated, use 19 transmitters and 51 receivers arranged in a circular geometry, covering a frequency range from 1 GHz to 18 GHz. Our approach’s viability is demonstrated through computational experiments using FEKO and MATLAB, showcasing its potential in enhancing automotive safety. This setup showcases our system’s capability to capture high-resolution 3D tomographic images, providing detailed information about object shapes and materials in the measurement domain. In summary, our paper proposes a groundbreaking approach to MIMO TSAR, offering the automotive industry a cost-effective, high-resolution imaging solution. This technique has the potential to revolutionize safety features in vehicles by providing more accurate and faster environmental assessment and decision-making capabilities.