# US 2026/0037607 A1 · System and Method for Grip Gesture Authentication based on Pressure Gradient Mapping

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2026/0037607 A1
- Document ID: US20260037607A1
- Status: Published application
- Kind: A1
- Application number: 18/793,094
- Filed: 2024-08-02
- Published: 2026-02-05
- Earliest claimed: 2024-08-02
- Assignees: Bank of America Corporation
- Applicants: Bank of America Corporation
- Inventors: Sudarshan Veeramreddy; Banupriya B; Lopamudra Dalai; Priya Gadde; Honey Gautam
- Web page: https://patentgrep.com/patent/US20260037607A1

## CPC

- G06F 21/316: {by observing the pattern of computer usage, e.g. typical user behaviour}
- G06F 21/32: using biometric data, e.g. fingerprints, iris scans or voiceprints
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

A system for performing user authentication receives a request to perform an interaction, accesses sensor data of a first grip gesture performed by a user, wherein the first grip gesture is associated with the request, wherein the sensor data includes at least pressure data captured by pressure sensors, generates a first pressure gradient mapping for the first grip gesture based on the sensor data, determines whether the first pressure gradient mapping matches at least one of the registered pressure gradient mappings corresponding to registered grip gestures associated with the user based on a comparison between the first pressure gradient mapping and each of the registered pressure gradient mappings, approves the interaction if the first pressure gradient mapping matches at least one of the registered pressure gradient mappings, and declines the interaction if the first pressure gradient mapping does not match at least one of the registered pressure gradient mappings.

## Claims

### Claim 1

A system for performing user authentication, comprising:
a memory operable to store one or more registered pressure gradient mappings corresponding to one or more registered grip gestures associated with a user; and
a processor, operably coupled to the memory, and configured to:
receive a request to perform an interaction;
access sensor data associated with a first grip gesture performed by the user, wherein the first grip gesture is associated with the request to perform the interaction, and wherein the sensor data comprises at least pressure data captured by one or more pressure sensors;
generate, based on the sensor data, a first pressure gradient mapping for the first grip gesture;
determine whether the first pressure gradient mapping matches at least one of the registered pressure gradient mappings based on a comparison between the first pressure gradient mapping and each of the one or more registered pressure gradient mappings;
if the first pressure gradient mapping matches at least one of the registered pressure gradient mappings, approve the interaction; and
if the first pressure gradient mapping does not match at least one of the registered pressure gradient mappings, decline the interaction.

### Claim 2 (depends on 1)

The system of claim 1, wherein the processor is further configured to:
extract one or more features from the first pressure gradient mapping, wherein the one or more features are based one or more of an intensity, a location, or a spread of pressure points, and wherein the comparison between the first pressure gradient mapping and each of the one or more registered pressure gradient mappings is based on the extracted features.

### Claim 3 (depends on 1)

The system of claim 1, wherein the processor is further configured to:
calculate a respective score for the comparison between the first pressure gradient mapping and each of the one or more registered pressure gradient mappings, wherein each score indicates a distribution difference between the first pressure gradient mapping and each of the one or more registered pressure gradient mappings; and
determine whether each of the scores is within a threshold value;
wherein determining whether the first pressure gradient mapping matches at least one of the registered pressure gradient mappings is further based on whether each of the scores is within the threshold value.

### Claim 4 (depends on 1)

The system of claim 1, wherein generating the first pressure gradient mapping comprises:
calculating, based on a Gaussian function, a plurality of gradients corresponding to pressure changes across a sensor surface associated with a user device on which the user performed the first grip gesture.

### Claim 5 (depends on 4)

The system of claim 4, wherein the Gaussian function is based on a plurality of parameters, wherein the processor is further configured to:
transmit instructions for prompting the user to register grip gestures for authentication;
access sensor data associated with a plurality of grip gestures performed by the user; and
determine the plurality of parameters based on a calibration of the sensor data.

### Claim 6 (depends on 1)

The system of claim 1, wherein the memory is further operable to store a machine-learning model configured to authenticate users based on features associated with grip gestures, wherein the processor is further configured to:
extract one or more features from the sensor data, wherein the one or more features comprise one or more of pressure distribution, timing, or spatial coordinates;
determine whether the user performed the first grip gesture by inputting the extracted features to the machine-learning model, the machine-learning model outputting a confidence score indicating a likelihood that the user performed the first grip gesture;
if the user performed the first grip gesture, approve the interaction; and
if the user did not perform the first grip gesture, decline the interaction.

### Claim 7 (depends on 6)

The system of claim 6, wherein the processor is further configured to:
normalize the extracted features based on one or more of scaling pressure values associated with pressure distribution, standardizing timing, or standardizing spatial coordinates.

### Claim 8 (depends on 6)

The system of claim 6, wherein the processor is further configured to:
access sensor data associated with a plurality of grip gestures performed by a plurality of users;
train, based on the accessed sensor data, a baseline machine-learning model for authenticating users based on grip gestures; and
generate the machine-learning model by finetuning the baseline machine-learning model based on sensor data associated with grip gestures performed by the user.

### Claim 9

A method for performing user authentication comprising, by one or more computing systems:
receiving a request to perform an interaction;
accessing sensor data associated with a first grip gesture performed by a user, wherein the first grip gesture is associated with the request to perform the interaction, and wherein the sensor data comprises at least pressure data captured by one or more pressure sensors;
generating, based on the sensor data, a first pressure gradient mapping for the first grip gesture;
determining whether the first pressure gradient mapping matches at least one of one or more registered pressure gradient mappings corresponding to one or more registered grip gestures associated with the user based on a comparison between the first pressure gradient mapping and each of the one or more registered pressure gradient mappings;
if the first pressure gradient mapping matches at least one of the registered pressure gradient mappings, approving the interaction; and
if the first pressure gradient mapping does not match at least one of the registered pressure gradient mappings, declining the interaction.

### Claim 10 (depends on 9)

The method of claim 9, further comprising:
extracting one or more features from the first pressure gradient mapping, wherein the one or more features are based one or more of an intensity, a location, or a spread of pressure points, and wherein the comparison between the first pressure gradient mapping and each of the one or more registered pressure gradient mappings is based on the extracted features.

### Claim 11 (depends on 9)

The method of claim 9, further comprising:
calculating a respective score for the comparison between the first pressure gradient mapping and each of the one or more registered pressure gradient mappings, wherein each score indicates a distribution difference between the first pressure gradient mapping and each of the one or more registered pressure gradient mappings; and
determining whether each of the scores is within a threshold value;
wherein determining whether the first pressure gradient mapping matches at least one of the registered pressure gradient mappings is further based on whether each of the scores is within the threshold value.

### Claim 12 (depends on 9)

The method of claim 9, wherein generating the first pressure gradient mapping comprises:
calculating, based on a Gaussian function, a plurality of gradients corresponding to pressure changes across a sensor surface associated with a user device on which the user performed the first grip gesture.

### Claim 13 (depends on 12)

The method of claim 12, wherein the Gaussian function is based on a plurality of parameters, wherein the method further comprises:
transmitting instructions for prompting the user to register grip gestures for authentication;
accessing sensor data associated with a plurality of grip gestures performed by the user; and
determining the plurality of parameters based on a calibration of the sensor data.

### Claim 14 (depends on 9)

The method of claim 9, further comprising:
extracting one or more features from the sensor data, wherein the one or more features comprise one or more of pressure distribution, timing, or spatial coordinates;
determining whether the user performed the first grip gesture by inputting the extracted features to a machine-learning model configured to authenticate users based on features associated with grip gestures, the machine-learning model outputting a confidence score indicating a likelihood that the user performed the first grip gesture;
if the user performed the first grip gesture, approving the interaction; and
if the user did not perform the first grip gesture, declining the interaction.

### Claim 15 (depends on 14)

The method of claim 14, further comprising:
accessing sensor data associated with a plurality of grip gestures performed by a plurality of users;
training, based on the accessed sensor data, a baseline machine-learning model for authenticating users based on grip gestures; and
generating the machine-learning model by finetuning the baseline machine-learning model based on sensor data associated with grip gestures performed by the user.

### Claim 16

A non-transitory computer-readable medium storing instructions that when executed by a processor cause the processor to:
receive a request to perform an interaction;
access sensor data associated with a first grip gesture performed by a user, wherein the first grip gesture is associated with the request to perform the interaction, and wherein the sensor data comprises at least pressure data captured by one or more pressure sensors;
generate, based on the sensor data, a first pressure gradient mapping for the first grip gesture;
determine whether the first pressure gradient mapping matches at least one of one or more registered pressure gradient mappings corresponding to one or more registered grip gestures associated with the user based on a comparison between the first pressure gradient mapping and each of the one or more registered pressure gradient mappings;
if the first pressure gradient mapping matches at least one of the registered pressure gradient mappings, approve the interaction; and
if the first pressure gradient mapping does not match at least one of the registered pressure gradient mappings, decline the interaction.

### Claim 17 (depends on 16)

The non-transitory computer-readable medium of claim 16, wherein the instructions further cause the processor to:
extract one or more features from the first pressure gradient mapping, wherein the one or more features are based one or more of an intensity, a location, or a spread of pressure points, and wherein the comparison between the first pressure gradient mapping and each of the one or more registered pressure gradient mappings is based on the extracted features.

### Claim 18 (depends on 16)

The non-transitory computer-readable medium of claim 16, wherein the instructions further cause the processor to:
calculate a respective score for the comparison between the first pressure gradient mapping and each of the one or more registered pressure gradient mappings, wherein each score indicates a distribution difference between the first pressure gradient mapping and each of the one or more registered pressure gradient mappings; and
determine whether each of the scores is within a threshold value;
wherein determining whether the first pressure gradient mapping matches at least one of the registered pressure gradient mappings is further based on whether each of the scores is within the threshold value.

### Claim 19 (depends on 16)

The non-transitory computer-readable medium of claim 16, wherein generating the first pressure gradient mapping comprises:
calculating, based on a Gaussian function, a plurality of gradients corresponding to pressure changes across a sensor surface associated with a user device on which the user performed the first grip gesture.

### Claim 19 (depends on 19)

on-transitory computer-readable medium of claim 19, wherein the Gaussian function is based on a plurality of parameters, wherein the instructions further cause the processor to:
transmit instructions for prompting the user to register grip gestures for authentication;
access sensor data associated with a plurality of grip gestures performed by the user; and
determine the plurality of parameters based on a calibration of the sensor data.
