Multiple Beam Radar Motion Estimation via Height Dimension

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Low cost radar sensors lack the necessary parameter matching and accuracy for precise human motion estimation due to unsynchronized frequency and phase, leading to difficulties in distinguishing individual human body parts and accurately determining motion characteristics, especially in scenarios with multi-path reflections and scintillation.

Innovation Solution

A method utilizing a multiple beam radar system to estimate height and cross-range information from radar data, allowing for the fitting of an object model to determine specific object motion characteristics, such as human body motion, by receiving data from multiple radar stations and applying techniques like trilateration and non-linear least-squares estimation to improve range and speed resolution.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of manufacture

If low cost radar sensors are used to build a non-coherent radar sensor network, then the system cost is reduced, but the measurement precision deteriorates due to unsynchronized frequency and phase

Engineering Contradiction:
Improvesystem costVSAvoidrange and speed resolution
Core Design Contradiction:
Ease of manufactureVSMeasurement precision

Solution Approach 1:

The patent introduces height as a new dimension by deploying radar sensors at different vertical positions. This spatial arrangement enables the system to resolve ambiguities in range and speed measurements that plague conventional planar sensor networks, thereby improving measurement precision without requiring expensive synchronized hardware

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

Solution Approach 2:

The patent divides the observation space into multiple height layers, with each radar sensor responsible for a specific vertical sector. This segmentation allows independent processing of signals from different height levels, enabling accurate motion estimation despite the non-coherent nature of individual sensors

Inventive Principle:
Principle #1Segmentation

2Loss of information

If multiple low cost radar sensors are combined to provide additional object information, then the quantity of object information increases, but the difficulty of detecting and measuring increases due to multi-path reflections and scintillation

Engineering Contradiction:
Improveobject information completenessVSAvoiddifficulty in distinguishing human body parts
Core Design Contradiction:
Loss of informationVSDifficulty of detecting and measuring

Solution Approach 1:

By adding the height dimension through vertically distributed sensors, the system creates unique measurement geometries for different body parts. This dimensional diversity enables the algorithm to distinguish between multi-path reflections from different locations and actual micro-Doppler signals from body part motions

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

Solution Approach 2:

The patent introduces an object model fitting algorithm that acts as an intermediary between raw radar measurements and motion characteristic extraction. This model-based approach filters out spurious signals from multi-path and scintillation effects while preserving genuine micro-Doppler information

Inventive Principle:
Principle #24Intermediary (Mediator)

3Device complexity

If fixed parameter radar sensors are used, then the device complexity is reduced, but the measurement precision deteriorates due to lack of parameter matching for specific applications

Engineering Contradiction:
Improveparameter adjustabilityVSAvoidmotion characteristic estimation accuracy
Core Design Contradiction:
Device complexityVSMeasurement precision

Solution Approach 1:

The patent employs post-processing algorithms that dynamically adjust effective measurement parameters based on the specific application scenario. By adapting the interpretation of fixed sensor data through software-based parameter optimization, the system achieves application-specific measurement precision without requiring hardware reconfiguration

Inventive Principle:
Principle #35Parameter changes

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

Enables detailed motion characteristic estimation, effectively distinguishing human body parts and improving classification accuracy, with correct classification rates of 90% for swinging and non-swinging arms and 93% for legs and arms or body, even in scenarios with overlapping antenna beams.

Implementation Method 1

If the object or a part of the object has rotation or translation in addition to the body motion, it might induce a frequency modulation and amplitude modulation on the radar return signal. These modulations are the so-called micro-Doppler or micro-motion effects

Methodology Applied
Scientific EffectDoppler effect: Doppler Effect

Data Source

PatentEP2382486B1A method for estimating an object motion characteristic from a radar signal, a computer system and a computer program product
Publication Date: 2016.10.26 NEDERLANDSE ORG VOOR TOEGEPAST NATUURWETENSCHAPPELIJK ONDERZOEK TNO
  • EP2382486B1 patent drawingFigure 1
  • EP2382486B1 patent drawingFigure 2a~2f
  • EP2382486B1 patent drawingFigure 3a~3d

AI summary

The invention relates to a method for estimating an object motion characteristic from a radar signal. The method comprises the step of receiving radar data of an object from a multiple beam radar system. Further, the method comprises the steps of associating radar data with estimated height and/or cross-range information of object parts causing the corresponding radar data and fitting an object model with radar data being associated with a selected estimated height and/or cross-range information interval. The method also comprises the step of determining an object motion characteristic from the fitted object model.