Automotive FMCW Radar System Design Using 3D Framework for Scenario Modeling

Keysight Technologies
[Application Note]

Introduction

According to government reports, thousands of people annually lose their lives and millions more have been injured due to car accidents all over the world [1]. That’s a prime reason why scientists and engineers are working on automotive applications, trying to develop automotive radar systems to reduce human casualties from car collisions. As the result of this market demand, various radar systems, such as adaptive cruise control (ACC), stop-and-go, blind spot detection (BSD), lane change assist (LCA), and rear crash warning (RCW), are now widely used in vehicles.

Automotive radar based on a frequency modulated continuous waveform (FMCW) [2] is one technology that is today widely used. Unlike pulse radar, FMCW Radar using continuous wave modulation can avoid high peak-to-average power ratio (PAPR) in transmission, which simplifies the design process for antennas and RF components like power amplifiers. Consequently, an automotive radar system based on this technology offers more advantages, such as good performance with simplified RF components, small size, light weight, and low cost.

This application note proposes design approaches for advanced FMCW radar systems in which multi-antenna, digital beamforming (DBF), multi-dimensional DSP signal processing, and parameter estimation algorithms are required.

Based on market investigation, automotive radar designers want a tool that can provide the following:

  • Different types of FMCW signal generation and analysis
  • Multi-antenna/array antenna design capability
  • Complex algorithm design for DBF and multi-target detection, unambiguous range and velocity measurements with high resolution and accuracy, as well as for complex DSP algorithms for collision avoidance and automotive driving
  • Ability to address system complexity in cross-domain architectures with high frequency
  • Complex environment scenarios including moving targets with multi-scattering radar Cross Section (RCS) and clutters plus Interference
  • Test considerations for advanced measurements, such as detection rate, false alarm rate, 2D/3D antenna pattern plots, and range Doppler plots, as well as an integrated auto-test system

The W1905 SystemVue radar Library provides these capabilities and can be used for simulation and test of automotive radar systems. To save development time and reduce cost, the SystemVue Radar Library provides highly-parameterized simulation models and higher-level reference design workspaces that allow designers to create radar system operation scenarios, including radar signal generation, processing, environmental effects such as clutter, interference, targets; receiver algorithm, and measurements. In Figure 1, a setup example is illustrated for a FMCW system design. Because of its strong capability, most of the models shown (e.g., signal generator, transmitter, Tx/Rx antennas, RF receiver, measurements, target RCS, clutter, and interference) can be obtained directly from the library. For more detailed architectures with more accurate results, SystemVue allows RF-DSP for the RF transmitter/RF receiver or DSP-EM for the antenna in cross-domain simulation. The W1905 block set and its example workspaces serve as algorithmic and architectural reference designs to verify radar performance under different signal conditions and environment scenarios. These can include target and RCS scenarios, clutter conditions, jammers and environmental interferers, and more. By accounting for a diverse set of environmental effects, while maintaining an open modeling environment (.m, C++, VHDL, test equipment), the radar system designer can explore architectures with high confidence in early R&D, without requiring expensive outdoor range testing or hardware simulators.

Figure 1

Figure 1. Setup of a FMCW system in SystemVue

Conventional FMCW System

Let’s start from a simple conventional FMCW system that is typically used for short-range Radar and only requires range detection. In the FMCW system, linear frequency modulation (LFM) signals are always used. In Figure 2, different LFM signals that form FMCW Radar are shown.

Figure 2

Figure 2. Typical linear FMCW signals generated by SystemVue are shown including: A) waveform for saw-tooth signal, B) group delay (frequency vs. time) for saw-tooth signal, C) waveform for triangle signal, and D) group delay (frequency vs. time) for triangle signal

 

Figure 3. Frequency versus time characteristics for a FMCW signal

 

Assume a FMCW Radar signal is transmitted via an antenna and reflected on a target. The receiver receives the signal after a time delay as shown in Figure 3. In the FMCW receiver input, the delay can be calculated by…

Start typing and press Enter to search