Misspecified Cramér-Rao Bound for Terahertz Automotive Radar Range Estimation

Moshe Levy-Israel, Joseph Tabrikian, Igal Bilik

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review


High-resolution information about a host vehicle's surroundings is essential for autonomous driving and advanced driver assistance systems (ADAS). This requires radar systems with high-resolution range, Doppler, azimuth, and elevation capabilities. Terahertz (THz) frequency band radar systems could provide nearly-optical resolution while compact and lightweight. However, frequency-selective range-dependent attenuation of the THz propagation channel challenges the implementation of automotive radars at this frequency band. Estimating the parameters of the THz channel can be computationally complex. Therefore, a flat channel model is conventionally considered for the THz radar range estimator. This work investigates the influence of this assumption on the THz radar range performance estimation via the misspecified Cramér-Rao bound (MCRB). The ability of the proposed approach to evaluate the influence of the frequency-selective misspecification on the radar range estimation accuracy and detection range is evaluated via simulations.

Original languageAmerican English
Title of host publication31st European Signal Processing Conference, EUSIPCO 2023 - Proceedings
Number of pages5
ISBN (Electronic)9789464593600
StatePublished - 1 Jan 2023
Event31st European Signal Processing Conference, EUSIPCO 2023 - Helsinki, Finland
Duration: 4 Sep 20238 Sep 2023

Publication series

NameEuropean Signal Processing Conference


Conference31st European Signal Processing Conference, EUSIPCO 2023


  • Automotive radar
  • frequency-selective propagation
  • MCRB
  • range estimation
  • THz radar

All Science Journal Classification (ASJC) codes

  • Signal Processing
  • Electrical and Electronic Engineering


Dive into the research topics of 'Misspecified Cramér-Rao Bound for Terahertz Automotive Radar Range Estimation'. Together they form a unique fingerprint.

Cite this