Payment Card Authentication Using Machine Learning Proximity Detection

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing credit or debit card transaction authorization systems require user authentication through PIN or signature input, which is cumbersome and increases network bandwidth usage, and lack effective fraud detection.

Innovation Solution

Implementing NFC and RFID technology in combination with machine learning to estimate the proximity radius of devices associated with the card, using a machine learning model trained on RSSI data to authorize transactions based on predefined thresholds, and continuously update these thresholds for improved fraud detection.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If manual authentication (PIN or signature) is used, then transaction authorization is achieved, but user convenience deteriorates and network bandwidth consumption increases

Engineering Contradiction:

Improveauthentication convenienceVSAvoidnetwork bandwidth consumption

Core Design Contradiction:

Ease of operationVSLoss of energy

Solution Approach 1:

The patent replaces manual authentication mechanisms (PIN entry, signature capture) with an automated machine learning-based proximity detection system. The system uses RSSI data from NFC/RFID devices to calculate proximity radius and determines authentication automatically, eliminating the need for manual user input and reducing network communication overhead.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Reliability

If traditional authentication systems are used, then transaction processing is completed, but fraud detection capability is insufficient

Engineering Contradiction:

Improvefraud detection accuracyVSAvoidauthentication system complexity

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The system employs a machine learning model that autonomously analyzes proximity radius data and makes authentication decisions without requiring external verification systems. The model continuously learns from transaction patterns and automatically adjusts fraud detection thresholds, enabling the system to self-improve its reliability while managing complexity internally.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The patent implements a feedback mechanism where the machine learning model continuously receives transaction outcomes and proximity data, updates its internal parameters, and refines fraud detection thresholds. This closed-loop system improves fraud detection accuracy over time by learning from actual transaction patterns and anomalies.

Inventive Principle:

Principle #23Feedback

3Measurement precision

If machine learning model training with multiple devices is implemented, then fraud detection accuracy is improved, but computational complexity increases

Engineering Contradiction:

Improveproximity detection accuracyVSAvoidmodel training complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent segments the authentication system into distinct functional modules: device pairing module, RSSI data collection module, proximity radius calculation module, and machine learning model module. Each module handles specific computational tasks independently, making the overall complex system manageable and maintainable while achieving high measurement precision through coordinated operation of specialized components.

Inventive Principle:

Principle #1Segmentation

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

Reduces the need for manual authentication, decreases network bandwidth, and enhances transaction authorization confidence by detecting fraudulent activities in real-time.

Implementation Method 1

pairing a payment card to a mobile electronic device and a wearable device, the payment card including at least one of near-field communication (NFC) or radio frequency identification (RFID) capability

Methodology Applied

Scientific EffectNear-field communication (NFC): Electromagnetic Induction

Implementation Method 2

Methodology Applied

Scientific EffectRadio frequency identification (RFID): Electromagnetic Induction

Implementation Method 3

training a machine learning model by obtaining first received signal strength indicator (RSSI) data from the payment card, the mobile electronic device, and the wearable device at calibrated distances, calculating a first estimated proximity radius

Methodology Applied

Scientific EffectReceived signal strength indicator (RSSI): Electromagnetic Propulsion

Data Source

PatentUS12417447B2Machine learning for authentication based on device proximity

Publication Date: 2025.09.16 MASTERCARD INT INC

AI summary

A device and method of authentication includes pairing a card to a mobile electronic device and a wearable device. A machine learning model is trained by obtaining first received signal strength indicator (RSSI) data from the card, the mobile electronic device, and the wearable device at calibrated distances. A first estimated proximity radius encompassing the card, the mobile electronic device, and the wearable device is calculated. The first estimated proximity radius is classified to be within a threshold. Upon receipt of a request to authorize a request, second RSSI data from the card, the mobile electronic device, and the wearable device is obtained. A second estimated proximity radius encompassing the card, the mobile electronic device, and the wearable device is calculated. Using the trained machine learning model, the second estimated proximity radius is determined to be within the threshold. Authentication is then complete.