Biometric Recognition Using Inertial Sensor Tremor Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Biometric recognition systems face challenges in balancing security and user acceptability, particularly in real-life scenarios, as they often require intrusive methods or complex equipment, and existing tremor-based systems struggle to distinguish between physiological and pathological tremors effectively.
Innovation Solution
A biometric recognition system utilizing inertial sensors, such as accelerometers and gyroscopes, to detect and classify involuntary hand tremor, providing a fast, less intrusive authentication method by analyzing tremor patterns and features extracted through Weighted Frequency Fourier Linear Combiner (WFLC) filtering and spectral density estimation, allowing for secure device unlocking.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional biometric recognition methods (e.g., facial recognition, fingerprint scanning) are used to ensure security, then security is improved, but user acceptability deteriorates due to intrusive procedures and complex equipment requirements
Solution Approach 1:
The system uses the user's own involuntary physiological tremor as the biometric identifier, eliminating the need for active user participation. The tremor occurs naturally without user control, making authentication automatic and non-intrusive while maintaining security through unique physiological patterns
Solution Approach 2:
The patent replaces complex mechanical or optical biometric systems with an inertial sensor-based detection system. Instead of using cameras, fingerprints, or iris scanners, the system uses simple accelerometers to detect tremor patterns, significantly reducing equipment complexity while maintaining authentication capability
2Extent of automation
If existing tremor-based biometric systems are used, then authentication capability is provided, but the system fails to effectively distinguish between physiological and pathological tremors, reducing reliability
Solution Approach 1:
The system changes the parameters used for tremor analysis by focusing on specific frequency ranges (4-12 Hz for physiological tremor) and using multiple inertial measurement parameters (acceleration, velocity, displacement) to differentiate between physiological and pathological tremors based on their distinct spectral characteristics
Solution Approach 2:
The system incorporates a feedback mechanism where the classifier continuously analyzes tremor patterns and adjusts authentication decisions based on the distinction between physiological and pathological tremor characteristics, improving reliability through iterative pattern recognition
3Reliability
If complex biometric recognition equipment is deployed to improve security, then security is improved, but device complexity increases
Solution Approach 1:
The patent replaces complex mechanical or optical biometric systems with an inertial sensor-based detection system. Instead of using cameras, fingerprints, or iris scanners, the system uses simple accelerometers to detect tremor patterns, significantly reducing equipment complexity while maintaining authentication capability
Solution Approach 2:
The inertial sensors used in the system serve multiple functions: they detect tremor patterns for authentication, track device orientation, and monitor user interaction patterns. This multi-functionality reduces the need for dedicated biometric hardware, simplifying the overall device architecture
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
Enables reliable and efficient biometric recognition with minimal user intervention, capable of authenticating a device owner within one second, offering a balance between security and acceptability by differentiating between authorized and unauthorized users based on tremor patterns, even in dynamic real-life conditions.
Implementation Method 1
a first inertial sensor, such as an accelerometer, to detect linear acceleration of the hand
Implementation Method 2
a second inertial sensor, such as a gyroscope, to detect angular velocity of the hand
Implementation Method 3
Filtering and feature extraction are performed using a Weighted Frequency Fourier Linear Combiner (WFLC)
Data Source
AI summary
A biometric recognition system for a hand held computing device incorporating an inertial measurement unit (IMU) comprising a plurality of accelerometers and at least one gyroscope is disclosed. A tremor analysis component is arranged to: obtain from the IMU, accelerometer signals indicating device translational acceleration along each of X, Y and Z axes as well as a gyroscope signal indicating rotational velocity about the Y axis during a measurement window. Each of the IMU signals is filtered to provide filtered frequency components for the signals during the measurement window. The accelerometer signals are combined to provide a combined filtered accelerometer magnitude signal for the measurement window. A spectral density estimation is provided for each of the combined filtered accelerometer magnitude signal and the filtered gyroscope signal. An irregularity is determined for each spectral density estimation; and based on the determined irregularities, the tremor analysis component attempts to authenticate a user of the device.