Biometric radar system and method for identifying persons and positional states of persons
AssigneeRaytheon Company
InventorsWesley H. Dwelly, Vinh N. Adams
Life of this patent
Estimating the term…
Abstract
A biometric radar system and method for identifying a person's positional state are generally described herein. The biometric radar may phase adjust a sequence of radar return signals received through two or more receive antennas to remove at least some phase noise due to the stationary objects. The biometric radar may also segment the phase adjusted radar return signals into a plurality of multi-resolutional Doppler components. Each multi-resolutional Doppler component may be associated with one of a plurality of biometric features. The biometric radar system may also combine and weight the segmented radar returns for each biometric feature to generate weighted classifications for a feature extraction process.
Claims33
A biometric radar comprising:
- phase adjusters to phase adjust a sequence of radar return signals received through two or more receive antennas, the phase-adjusting removing at least some phase noise due to stationary objects;
- a signal processor to segment the phase adjusted radar return signals into a plurality of multi-resolutional Doppler components, each multi-resolutional Doppler component being associated with one of a plurality of biometric features;
- and
- a non-linear convolutional neural network to combine and weight the segmented radar returns for each biometric feature to generate weighted classifications for a feature extraction process.
A method for identifying a person and a person's positional state comprising:
- phase adjusting a sequence of radar return signals received through two or more receive antennas, the phase-adjusting removing at least some phase noise due to stationary objects;
- segmenting the phase adjusted radar return signals into a plurality of multi-resolutional Doppler components, each multi-resolutional Doppler component being associated with one of a plurality of biometric features;
- and
- combining and weighting the segmented radar returns for each biometric feature in a non-linear convolutional neural network to generate weighted classifications for a feature extraction process.
A radar system comprising:
- transmitter circuitry to transmit a series of short pulses stepped in frequency;
- phase adjusters to phase adjust a sequence of radar return signals received through two or more receive antennas, the phase-adjusting removing at least some phase noise due to stationary objects, the transmitted series of short pulses corresponding to the received sequence of radar return signals;
- a signal processor to segment the phase adjusted radar return signals into a plurality of multi-resolutional Doppler components, each multi-resolutional Doppler component being associated with one of a plurality of biometric features;
- and
- a non-linear convolutional neural network to combine and weight the segmented radar returns for each biometric feature to generate weighted classifications for a feature extraction process,
- wherein each multi-resolutional Doppler component has a frequency structure associated with one of the biometric features, and
- wherein the biometric features associated with humans and comprise heartbeat, breathing, postural sway and gait.
Dependencies are inferred from the claim text.