Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Target Induced Angle Grid Regularized Estimation for Ghost Identification in Automotive Radar

Junho Kweon, Vishal Monga

TL;DR

Ghost targets arising from multipath reflections challenge automotive radar systems; existing approaches like MP-IAA degrade in low SNR. The paper introduces TIGRE, a grid-based estimator that adds a Target Induced Regularizer to enforce physically plausible off-diagonal grid entries and promote sparsity on the DOA/DOD grid, with a formal asymptotic $\ell_0$ sparsity property when $G=Q$. A custom initialization seeds the algorithm using a diagonal-only solution, and a step-by-step iterative procedure updates the grid entries to convergence. Numerical results on co-located $N_t=N_r=8$ radar show that TIGRE, especially with custom initialization, outperforms MP-IAA and other baselines in estimation accuracy and computational efficiency, particularly at moderate to low SNRs, by producing cleaner angle grids and reducing ghost-target false alarms. Overall, the method provides a robust, scalable approach for ghost-target identification in autonomous driving contexts.

Abstract

This study presents a novel algorithm for identifying ghost targets in automotive radar by estimating complex valued signal strength across a two-dimensional angle grid defined by direction-of-arrival (DOA) and direction-of-departure (DOD). In real-world driving environments, radar signals often undergo multipath propagation due to reflections from surfaces such as guardrails. These indirect paths can produce ghost targets - false detections that appear at incorrect locations - posing challenges to autonomous navigation. A recent method, the Multi-Path Iterative Adaptive Approach (MP-IAA), addresses this by jointly estimating the DOA/DOD angle grid, identifying mismatches as indicators of ghost targets. However, its effectiveness declines in low signal-to-noise ratio (SNR) settings. To enhance robustness, we introduce a physics-inspired regularizer that captures structural patterns inherent to multipath propagation. This regularizer is incorporated into the estimation cost, forming a new loss function that guides our proposed algorithm, TIGRE (Target-Induced angle-Grid Regularized Estimation). TIGRE iteratively minimizes this regularized loss and we show that our proposed regularizer asymptotically enforces L0 sparsity on the DOA/DOD grid. Numerical experiments demonstrate that the proposed method substantially enhances the quality of angle-grid estimation across various multipath scenarios, particularly in low SNR environments, providing a more reliable basis for subsequent ghost target identification.

Target Induced Angle Grid Regularized Estimation for Ghost Identification in Automotive Radar

TL;DR

Ghost targets arising from multipath reflections challenge automotive radar systems; existing approaches like MP-IAA degrade in low SNR. The paper introduces TIGRE, a grid-based estimator that adds a Target Induced Regularizer to enforce physically plausible off-diagonal grid entries and promote sparsity on the DOA/DOD grid, with a formal asymptotic sparsity property when . A custom initialization seeds the algorithm using a diagonal-only solution, and a step-by-step iterative procedure updates the grid entries to convergence. Numerical results on co-located radar show that TIGRE, especially with custom initialization, outperforms MP-IAA and other baselines in estimation accuracy and computational efficiency, particularly at moderate to low SNRs, by producing cleaner angle grids and reducing ghost-target false alarms. Overall, the method provides a robust, scalable approach for ghost-target identification in autonomous driving contexts.

Abstract

This study presents a novel algorithm for identifying ghost targets in automotive radar by estimating complex valued signal strength across a two-dimensional angle grid defined by direction-of-arrival (DOA) and direction-of-departure (DOD). In real-world driving environments, radar signals often undergo multipath propagation due to reflections from surfaces such as guardrails. These indirect paths can produce ghost targets - false detections that appear at incorrect locations - posing challenges to autonomous navigation. A recent method, the Multi-Path Iterative Adaptive Approach (MP-IAA), addresses this by jointly estimating the DOA/DOD angle grid, identifying mismatches as indicators of ghost targets. However, its effectiveness declines in low signal-to-noise ratio (SNR) settings. To enhance robustness, we introduce a physics-inspired regularizer that captures structural patterns inherent to multipath propagation. This regularizer is incorporated into the estimation cost, forming a new loss function that guides our proposed algorithm, TIGRE (Target-Induced angle-Grid Regularized Estimation). TIGRE iteratively minimizes this regularized loss and we show that our proposed regularizer asymptotically enforces L0 sparsity on the DOA/DOD grid. Numerical experiments demonstrate that the proposed method substantially enhances the quality of angle-grid estimation across various multipath scenarios, particularly in low SNR environments, providing a more reliable basis for subsequent ghost target identification.
Paper Structure (11 sections, 1 theorem, 22 equations, 3 figures, 4 tables, 1 algorithm)

This paper contains 11 sections, 1 theorem, 22 equations, 3 figures, 4 tables, 1 algorithm.

Table of Contents

  1. Introduction
  2. System Model
  3. Proposed Method
  4. Target Induced Regularizer on DOA$\neq$DOD Grids
  5. Asymptotic $\ell_0$ Sparsity Behavior in DOA$/$DOD Grids
  6. Custom Initialization
  7. Step-By-Step Algorithm of TIGRE
  8. Numerical Results
  9. Simulation Settings
  10. Simulation Results
  11. Conclusion

Key Result

Lemma 1

The proposed regularizer in eq:regularizer encourages $\ell_0$ sparsity asymptotically in the $=$ grids of ${\bf X}$, i.e., the diagonal elements of ${\bf X}$ when $G=Q$.

Figures (3)

  • Figure 1: Geometry of multipath propagation in automotive radar. A is the automotive radar platform, B is real target, and C is reflection point.
  • Figure 2: Estimated / angle grid ${\bf X}$. (a), (b) are the case of two targets; and (c), (d) are the case of three targets. (a), (c) are from ; and (b),(d) are from the proposed algorithm.
  • Figure 3: Angle grids in a vector form ${\bf x}$ estimated by MP-IAA and TIGRE for the scenario of three targets, each having two ghost targets.

Theorems & Definitions (2)

  • Lemma 1
  • proof