Patent · US Active

Radar signal based machine learning

US12313773B2 · kind B2 · utility

0Cited by
1References
18Claims
0Family size

Assignee

Inventors

Key dates

Filing dateDec 31, 2019
Grant dateMay 27, 2025
Priority date
Expiry dateMay 11, 2041

Classification

  • Technology area (CPC G)Physics
  • CPC primaryG01S13/931
  • WIPO fieldMeasurement
  • WIPO sectorInstruments

Abstract

Techniques of object classification are described. In operation, a radar signal reflected by an object within range of the radar detection system is received and at least one set of frequency components corresponding to the object are extracted. At least one set of unit energy samples corresponding to the object is then generated based on the at least one set of frequency components followed by computation of a range compensated energy (RCE) corresponding to the object based on the at least one set of frequency components, where the RCE is indicative of range normalized energy corresponding to the object. A feature vector corresponding to the object is then generated based at least on the at least one set of unit energy samples and the RCE, wherein the feature vector is representative of characteristics of the object.

Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.