Measurement-Based Analysis of a Non-Coherent MIMO Radar Network for Automotive Applications

In this work, we perform a measurement-based analysis of a four-module cooperative radar network for automotive applications. The radar network is mounted on the front bumper of a car and operates as a multiple-input multiple-output system with 192 channels that exploits spatial diversity. We present an analysis of four different real-world scenarios in which the car is moving and encounters both static and moving targets. The results focus on the behaviour of the network through a non-coherent signal processing that is capable of including the bistatic paths while reducing the complexity of the setup. We include quasi-monostatic and bistatic responses as well as a combined network response that fuses the information from all the modules. This approach offers robustness against miss detections and ghost targets as well as improvements in target identification.