Wearable Sensor Authentication Using Meal-Related Heartbeat Patterns

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current methods for specifying individuals using wearable sensors, such as those based on electrocardiographic waveforms and behavior patterns, face challenges like incorrect wear detection and impersonation due to deviations in wearing position and similarity in vital information, leading to erroneous authentication.

Innovation Solution

The use of characteristic heartbeats associated with meals, which exhibit distinct patterns regardless of meal quantity or digestive ease, allows for accurate identification by analyzing the area size, amplitude, and response times of heartbeats peaks, enabling precise specification of individuals without relying on physical characteristics or circadian rhythms.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of time

If short-time biological features (electrocardiographic waveforms) are used for wearer specification, then authentication speed is improved, but measurement precision deteriorates due to deviation of wearing position and contact state

Engineering Contradiction:
Improveauthentication speedVSAvoidwearer specification accuracy
Core Design Contradiction:
Loss of timeVSMeasurement precision

Solution Approach 1:

The patent changes the parameter basis from short-time electrocardiographic waveforms to long-time biological features (resting heart rate, maximum blood pressure, minimum blood pressure) that are less sensitive to wearing position deviations. This parameter change maintains authentication functionality while improving measurement precision under varying wear conditions.

Inventive Principle:
Principle #35Parameter changes

2Measurement precision

If vital information (resting heart rate, blood pressure) is used for wearer specification, then measurement precision is improved, but reliability deteriorates due to similar vital information among different persons

Engineering Contradiction:
Improvewearer specification accuracyVSAvoidauthentication reliability
Core Design Contradiction:
Measurement precisionVSReliability

Solution Approach 1:

The patent merges multiple long-time biological features (resting heart rate, maximum blood pressure, minimum blood pressure) into a comprehensive authentication criterion. By combining these features, the system achieves both measurement precision and reliability, as the combination creates a more unique physiological signature that is less likely to be shared between different individuals.

Inventive Principle:
Principle #5Merging (Combining)

3Reliability

If behavior pattern (motion characteristics, positional history) is used for wearer specification, then authentication reliability is improved, but device complexity increases due to need for acceleration sensor and processing

Engineering Contradiction:
Improveauthentication reliabilityVSAvoidsensor and processing complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent extracts and utilizes only the essential biological features (resting heart rate, blood pressure) that can be obtained from existing wearable sensor data, without requiring additional acceleration sensors or complex positional history processing. This extraction approach maintains authentication reliability while reducing device complexity by focusing on the most discriminative physiological parameters.

Inventive Principle:
Principle #2Taking out (Extraction)

Data Source

PatentEP2992816B1Subject person specifying apparatus, subject person specifying program, and subject person specifying method
Publication Date: 2021.08.11 FUJITSU LTD
  • EP2992816B1 patent drawingFigure 1~2
  • EP2992816B1 patent drawingFigure 3~4
  • EP2992816B1 patent drawingFigure 5~6

AI summary

A subject person specifying apparatus includes a unit configured to acquire characteristic quantities of variations of heart rate related to a meal of a subject person from heart rate data of the subject person as a target of an individual specifying process, a unit configured to calculate a distribution range of history values of the characteristic quantities associated with a plurality of candidates and a unit configured to extract the candidate corresponding to the subject person from within the plurality of candidates, based on a relation between the characteristic quantities acquired from the subject person and the distribution range of the history values of the characteristic quantities.