Processing

Please wait...

Settings

Settings

1. WO2020005943 - RADAR DATA PROCESSING USING NEURAL NETWORK CLASSIFIER AND CONFIDENCE METRICS

Publication Number WO/2020/005943
Publication Date 02.01.2020
International Application No. PCT/US2019/038990
International Filing Date 25.06.2019
IPC
G PHYSICS
01
MEASURING; TESTING
S
RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
13
Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
02
Systems using reflection of radio waves, e.g. primary radar systems; Analogous systems
50
Systems of measurement based on relative movement of target
52
Discriminating between fixed and moving objects or between objects moving at different speeds
56
for presence detection
G PHYSICS
06
COMPUTING; CALCULATING; COUNTING
F
ELECTRIC DIGITAL DATA PROCESSING
17
Digital computing or data processing equipment or methods, specially adapted for specific functions
10
Complex mathematical operations
14
Fourier, Walsh or analogous domain transformations
G PHYSICS
06
COMPUTING; CALCULATING; COUNTING
N
COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
3
Computer systems based on biological models
02
using neural network models
G01S 13/56 (2006.01)
G06F 17/14 (2006.01)
G06N 3/02 (2006.01)
CPC
G01S 13/343
G01S 13/583
G01S 13/584
G01S 7/021
G01S 7/4056
G01S 7/417
Applicants
  • TEXAS INSTRUMENTS INCORPORATED [US/US]; P. O. Box 655474 Mail Station 3999 Dallas, TX 75265-5474, US
  • TEXAS INSTRUMENTS JAPAN LIMITED [JP/JP]; 24-1, Nishi-Sinjuku 6-chome Shinjuku-ku, 160-8366, JP (JP)
Inventors
  • BHARADWAJ, Sachin; IN
  • RAO, Sandeep; IN
Agents
  • DAVIS, Jr., Michael A.; US
  • ABRAHAM, Ebby; US
Common Representative
  • TEXAS INSTRUMENTS INCORPORATED; P. O. Box 655474 Mail Station 3999 Dallas, TX 75265-5474, US
Priority Data
16/204,45729.11.2018US
62/689,44625.06.2018US
Publication Language English (EN)
Filing Language English (EN)
Designated States
Title
(EN) RADAR DATA PROCESSING USING NEURAL NETWORK CLASSIFIER AND CONFIDENCE METRICS
(FR) TRAITEMENT DE DONNÉES RADAR À L'AIDE D'UN CLASSIFICATEUR DE RÉSEAUX DE NEURONES ARTIFICIELS ET DE MESURES DE CONFIANCE
Abstract
(EN)
A radar data processing device (1000) includes at least one analog-to-digital converter (ADC) (1002A-1002N) configured to digitize input signals (213A-213N). Each input signal (213A-213N) includes radar chirp and radar chirp reflection information received at one of multiple receiver antennas (210A-210N). The radar data processing device (1000) also includes fast Fourier transform (FFT) logic (240A) configured to generate FFT output samples based on each digitized input signal. At least some of the generated FFT output samples are across antenna FFT output samples associated with at least two of the receiver antennas (210A-210N). The radar data processing device (1000) also includes a processor (250A) configured to determine object parameters based on at least some of the generated FFT output samples. The processor (250A) uses a neural network classifier (260A) trained to provide a confidence metric for at least one of the object parameters.
(FR)
L'invention concerne un dispositif de traitement de données radar (1000) comprenant au moins un convertisseur analogique-numérique (CAN) (1002A-1002N) configuré pour numériser des signaux d'entrée (213A-213N). Chaque signal d'entrée (213A-213N) comprend des informations de signal radar de type chirp et de réflexion de signal radar de type chirp reçues au niveau d'une antenne parmi de multiples antennes de réception (210A-210N). Le dispositif de traitement de données radar (1000) comprend également une unité logique de transformée de Fourier rapide (FFT) configurée pour générer des échantillons de sortie FFT sur la base de chaque signal d'entrée numérisé. Au moins certains des échantillons de sortie FFT générés sont des échantillons de sortie FFT entre antennes associés à au moins deux des antennes de réception (210A-210N). Le dispositif de traitement de données radar (1000) comprend également un processeur (250A) configuré pour déterminer des paramètres d'objet sur la base d'au moins certains des échantillons de sortie FFT générés. Le processeur (250A) utilise un classificateur de réseaux de neurones artificiels (260A) entraîné pour fournir une mesure de confiance pour au moins l'un des paramètres d'objet.
Latest bibliographic data on file with the International Bureau