Detection of Multiple Stationary Humans Using UWB MIMO Radar

Area of Science:

• Radar Systems Engineering
• Biomedical Engineering
• Signal Processing

Background:

• Detecting single stationary humans has advanced, but challenges remain in identifying multiple individuals due to interference like sidelobes and shadows.
• Existing radar methods struggle with the complex interactions and signal masking inherent in multi-human scenarios.

Purpose of the Study:

• To develop and validate an ultra-wideband (UWB) multiple-input multiple-output (MIMO) radar system for enhanced detection and localization of multiple stationary humans.
• To overcome limitations of mutual interference and environmental clutter in multi-human radar detection.

Main Methods:

• A novel signal model incorporating bi-static and attitude angles for human vital signs was developed.
• Preprocessing enhanced signal-to-noise ratio (SNR), followed by a vital-sign-enhanced imaging algorithm to suppress clutter and interference.
• An automatic detection algorithm using constant false alarm rate (CFAR), morphological filtering, and clustering was implemented.

Main Results:

• The proposed UWB MIMO radar method produced high-quality 2D images of multiple humans.
• The system successfully discriminated and localized adjacent human targets, even behind brick walls.
• Significant improvement in detecting weak human targets amidst heavy clutter and shadow effects was demonstrated.

Conclusions:

• UWB MIMO radar offers a promising solution for robust multi-human detection and localization.
• The developed signal processing and imaging techniques effectively mitigate interference and environmental challenges.
• This technology enables reliable identification of multiple stationary individuals in complex, obstructed environments.