Radar Direction-of-Arrival Super-Resolution via Snapshot Extrapolation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Automotive radar systems face challenges in achieving accurate direction of arrival (DoA) estimation due to inaccuracies in antenna placement and calibration errors, which limit their resolution and the number of detectable objects, especially under real-world conditions with varying temperatures and antenna coupling issues.

Innovation Solution

A radar system employs an autoregressive model to extrapolate signal values, increasing the size of the input snapshot and covariance matrix, enabling enhanced super-resolution algorithms to accurately determine the direction of arrival and detect more objects by constructing a larger full-rank covariance matrix.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If conventional DoA estimation algorithms are used, then the system is simpler to implement, but the resolution and number of detectable objects are limited due to antenna placement inaccuracies and calibration errors

Engineering Contradiction:
Improvedirection of arrival estimation accuracyVSAvoidsignal processing complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent applies preliminary calibration to determine actual antenna positions and coupling coefficients before performing DoA estimation. This preliminary action compensates for manufacturing inaccuracies and establishes correction factors that improve subsequent measurement precision without increasing the complexity of the main estimation algorithm.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent introduces an intermediary calibration process that measures and characterizes antenna coupling effects separately. This intermediary step creates a correction model that mediates between the imperfect physical antenna array and the DoA estimation algorithm, improving accuracy without requiring a complete redesign of the estimation process.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Measurement precision

If the number of antennas is increased to improve resolution, then the measurement precision improves, but the device complexity and cost increase

Engineering Contradiction:
Improvespatial resolutionVSAvoidantenna array complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent changes the parameter space by moving from physical antenna spacing to signal processing domain operations. Through calibration and coupling compensation, the system achieves enhanced spatial resolution through mathematical transformations rather than physical antenna multiplication, avoiding the complexity of larger antenna arrays.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The patent transitions from the physical spatial dimension to the signal processing dimension. By applying calibration corrections and coupling compensation in the signal domain, the system achieves resolution enhancement without adding physical antennas, effectively moving the problem-solving approach to another dimension.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

3Measurement precision

If super-resolution algorithms are applied to compensate for antenna placement errors, then the measurement precision improves, but the algorithms become more sensitive to calibration errors and noise

Engineering Contradiction:
Improvedirection of arrival estimation accuracyVSAvoidalgorithm robustness
Core Design Contradiction:
Measurement precisionVSReliability

Solution Approach 1:

The patent applies beforehand cushioning by performing comprehensive calibration to determine actual antenna positions and coupling coefficients before executing the super-resolution algorithm. This preparatory compensation creates a more robust input signal that reduces the algorithm's sensitivity to noise and residual calibration errors during execution.

Inventive Principle:
Principle #11Beforehand cushioning (Prior cushioning)

Solution Approach 2:

The patent implements feedback through the calibration process that measures actual antenna performance and uses this information to correct subsequent measurements. The calibration data feeds back into the DoA estimation process, creating a closed-loop system that compensates for systematic errors and improves reliability.

Inventive Principle:
Principle #23Feedback

4Adaptability or versatility

If the radar system operates in real-world conditions with temperature variations and antenna coupling, then the system is more practically applicable, but the measurement precision deteriorates due to calibration errors

Engineering Contradiction:
Improvereal-world operational capabilityVSAvoiddirection of arrival estimation accuracy
Core Design Contradiction:
Adaptability or versatilityVSMeasurement precision

Solution Approach 1:

The patent performs preliminary calibration under actual operating conditions to capture temperature effects and antenna coupling characteristics before DoA estimation. This preliminary action establishes condition-specific correction factors that maintain measurement precision despite real-world environmental variations.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent adapts to real-world conditions by changing the calibration parameters to reflect actual operating temperatures and coupling states. Rather than assuming ideal conditions, the system measures and compensates for real-world parameter variations, maintaining precision across different environmental conditions.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP4579280A1Radar signal direction of arrival super-resolution estimation
Publication Date: 2025.07.02 NXP BV
  • EP4579280A1 patent drawingFigure 1
  • EP4579280A1 patent drawingFigure 2
  • EP4579280A1 patent drawingFigure 3

AI summary

A device includes a radar processor that transmits, at a first time, a first radar signal, receives a received signal, and processes the received signal to generate a range-Doppler data frame. The radar processor determines a first snapshot comprising a first plurality of values associated with a first range-Doppler bin of the range-Doppler data frame and processes the first plurality of values in the first snapshot to generate an autoregressive model based upon the first plurality of values. The radar processor uses use the autoregressive model to extrapolate a second snapshot, wherein the second snapshot includes the first plurality of values and a second plurality values generated using the autoregressive model, determines, using the second snapshot, a full rank covariance matrix, and identifies attributes of a plurality of objects using the full rank covariance matrix.