This is an invited paper for the session “Combining geometric propagation models with adaptive waveform technology: a perspective for emerging communications, radar and navigation paradigms” organized by Prof. Danilo Erricolo, Dr. Jon A. Sjogren, and Dr. Michael C. Wicks.
Virtual Performance Evaluation of Automotive Radar Concepts in Realistic Traffic Environments Christian Sturm, Sandra Knörzer, Thomas Zwick, Werner Wiesbeck Institut für Hochfrequenztechnik und Elektronik (IHE), Universität Karlsruhe (TH) Kaiserstr. 12, 76131 Karlsruhe, Germany
[email protected]
Abstract— In this paper a simulation approach is presented for virtually investigating the performance of different system parameters of automotive radar systems. This is achieved by combining dedicated simulation models for the radar transmitter and receiver with a deterministic propagation simulator based on ray tracing. This approach is in particular suitable for investigating the capabilities of beam forming algorithms in radar systems with multiple antenna receivers. The configuration of the developed tool is explained in detail and the performance of classic digital beam forming and the MUSIC algorithm are evaluated. I. INTRODUCTION In recent years the application of radar sensors in upper class cars has become very popular for adaptive cruise control systems that automatically adapt to the speed of the vehicle ahead. In order to allow for a distinction of the vehicles on the different lanes these systems have to provide 2-dimensional radar sensing by detecting the position of other vehicles and obstacles in both, range and azimuth angle. This can be accomplished by multiple antenna processing techniques, which exploit the phase differences between the signals received at different spatial positions. For this numerous algorithms exist, which are based on different mathematical approaches. The aim of the work presented here is to develop a simulation tool that allows for the investigation of the performance of these multiple antenna processing techniques in realistic street scenarios. II. SIMULATION MODEL The new simulation tool combines simulation components for a radar transmitter and receiver with a deterministic propagation simulator based on ray tracing. The radar transmits pseudo random signals with a bandwidth of 100 MHz at a carrier frequency of 24 GHz. The model includes four individual receivers for the multiple antenna signal processing. The ray tracing simulator supports reflection, diffraction and scattering in a dynamic urban traffic scenario. The deterministic simulation results are subsequently processed with the radar algorithms. III. SIMULATION RESULTS In Fig. 1 an example simulation result of the comparison of two multiple antenna techniques in a typical street scenario is shown. The picture on the left shows the investigated scenario in the ray tracing simulation. The two approaches that are compared are the classic digital beam forming approach, based on Fourier transform, and the MUSIC algorithm, which is a popular technique in stochastic signal processing. As it can be seen the MUSIC algorithm outperforms the classic beam forming approach by far.
Fig. 1. Simulation result comparing two different multiple antenna techniques
