Granted Patent B2

US 12,474,782 · App. 18/508,988 · Granted Nov 18, 2025

Inventors: Priyabrata Parida (Dallas, TX); Vutha Va (Plano, TX); Saifeng Ni (Santa Clara, CA); Anum Ali (Frisco, TX); Boon Loong Ng (Plano, TX)

Assignee: Samsung Electronics Co., Ltd.

G06F3/017G01S7/415G01S13/581G01S13/88

Abstract

Claims (98)

1 . An electronic device comprising:

a transceiver configured to transmit and receive radar signals; and

a processor operatively coupled to the transceiver, the processor configured to:

identify, based on the received radar signals, a plurality of radar frames related to an activity of a target;

extract a plurality of features from the plurality of radar frames;

compute burst attributes for the extracted features;

predict a gesture based on the burst attributes;

determine whether the predicted gesture is a valid gesture; and

if the predicted gesture is a valid gesture, perform an action corresponding to the predicted gesture.

2 . The electronic device of claim 1 , wherein to extract the plurality of features from the plurality of radar frames, the processor is further configured to, for each radar frame:

obtain a Range-Doppler map;

select a column corresponding to a peak of the target in a range profile;

calculate an average radial velocity of the target;

obtain an azimuth and elevation angle spectrum corresponding to a target peak;

calculate an average angle of the target; and

obtain average azimuth/elevation tangential velocities based on a target angle variation over two frames.

3 . The electronic device of claim 1 , wherein the processor is further configured to:

determine, based on the burst attributes, whether the activity of the target is a valid activity,

wherein the gesture is predicted based on the activity of the target being the valid activity.

4 . The electronic device of claim 1 , wherein to predict the gesture, the processor is further configured to:

determine whether symmetry and pattern criteria for the burst attributes are met;

based on whether the symmetry and pattern criteria for the burst attributes are met, determine whether the activity is a valid activity;

perform a gesture classification on the activity if the activity is a valid activity, wherein the gesture classification is based on a previously obtained gesture set; and

predict the gesture based on the gesture classification.

5 . The electronic device of claim 4 , wherein the previously obtained gesture set comprises:

a swipe center-left-center (CLC) gesture;

a swipe center-right-center (CRC) gesture;

a swipe center-down-center (CDC) gesture; and

a swipe center-up-center (CUC) gesture.

6 . The electronic device of claim 4 , wherein to obtain the previously obtained gesture set, the processor is further configured to:

define a set of gestures with common burst attributes;

collect a plurality of samples for each gesture;

compute burst attributes for each sample;

derive statistical summaries for each burst attribute; and

obtain a set of thresholds to separate gestures from non-gestures.

7 . The electronic device of claim 6 , wherein the processor is further configured to:

collect a plurality of samples for non-gestures;

determine whether a probability of false positive and false negative gesture classifications is below an acceptable threshold; and

if the probability of false positive and false negative gesture classifications is below the acceptable threshold, accept the set of thresholds to separate gestures from non-gestures.

8 . The electronic device of claim 6 , wherein each gesture from the gesture set is defined in relation to a number of major bursts, minor bursts, and a pattern of bursts in a sequence.

9 . The electronic device of claim 1 , wherein to determine whether the predicted gesture is a valid gesture, the processor is further configured to:

determine a correlation between the extracted features; and

determine whether the predicted gesture is a valid gesture based on the correlation between the extracted features.

10 . The electronic device of claim 1 , wherein the plurality of features comprises:

radial velocity;

azimuth velocity; and

elevation velocity.

11 . The electronic device of claim 1 , wherein the burst attributes comprise at least one of:

burst length;

burst height;

burst area;

burst sign; and

whether a burst is part of a burst chain.

12 . A method of operating an electronic device, the method comprising:

identifying, based on received radar signals, a plurality of radar frames related to an activity of a target;

extracting a plurality of features from the plurality of radar frames;

computing burst attributes for the extracted features;

predicting a gesture based on the burst attributes;

determining whether the predicted gesture is a valid gesture; and

if the predicted gesture is a valid gesture, performing an action corresponding to the predicted gesture.

13 . The method of claim 12 , wherein extracting the plurality of features from the plurality of radar frames comprises, for each radar frame:

obtaining a Range-Doppler map;

selecting a column corresponding to a peak of the target in a range profile;

calculating an average radial velocity of the target;

obtaining an azimuth and elevation angle spectrum corresponding to a target peak;

calculating an average angle of the target; and

obtaining average azimuth/elevation tangential velocities based on a target angle variation over two frames.

14 . The method of claim 12 , further comprising:

determining, based on the burst attributes, whether the activity of the target is a valid activity,

wherein the gesture is predicted based on the activity of the target being the valid activity.

15 . The method of claim 12 , wherein to predicting the gesture comprises:

determining whether symmetry and pattern criteria for the burst attributes are met;

based on whether the symmetry and pattern criteria for the burst attributes are met, determining whether the activity is a valid activity;

performing a gesture classification on the activity if the activity is a valid activity, wherein the gesture classification is based on a previously obtained gesture set; and

predicting the gesture based on the gesture classification.

16 . The electronic device of claim 15 , wherein obtaining the previously obtained gesture set comprises:

defining a set of gestures with common burst attributes;

collecting a plurality of samples for each gesture;

computing burst attributes for each sample;

deriving statistical summaries for each burst attribute;

obtaining a set of thresholds to separate gestures from non-gestures;

collecting a plurality of samples for non-gestures;

determining whether a probability of false positive and false negative gesture classifications is below an acceptable threshold; and

if the probability of false positive and false negative gesture classifications is below the acceptable threshold, accepting the set of thresholds to separate gestures from non-gestures.

17 . The method of claim 15 , wherein each gesture from the gesture set is defined in relation to a number of major bursts, minor bursts, and a pattern of bursts in a sequence.

18 . The method of claim 12 , wherein determining whether the predicted gesture is a valid gesture comprises:

determining a correlation between the extracted features; and

determining whether the predicted gesture is a valid gesture based on the correlation between the extracted features.

19 . The method of claim 12 , wherein the plurality of features comprises:

radial velocity;

azimuth velocity; and

elevation velocity.

20 . The method of claim 12 , wherein the burst attributes comprise at least one of:

burst length;

burst height;

burst area;

burst sign; and

whether a burst is part of a burst chain.

Assignments (1)

Continuity (2)

References Cited (11)