Using cloud models of heartbeats as the entity identifier to secure mobile devices

Area of Science:

• Biometrics
• Mobile Security
• Cognitive Computing

Background:

• Mobile devices store sensitive data, necessitating robust security measures.
• Traditional authentication methods like passwords are vulnerable to theft and unauthorized access.
• Existing biometric systems may have limitations in accuracy or implementation.

Purpose of the Study:

• To propose and evaluate a new biometric authentication system for mobile devices.
• To utilize cognitive cloud models of heartbeats for unique user identification.
• To assess the system's performance in terms of accuracy, speed, and power consumption.

Main Methods:

• Heartbeats are captured using two ECG electrodes connected to a mobile device.
• A backward normal cloud generator creates standard cloud models of heartbeats.
• User authentication involves comparing newly generated heartbeat cloud models with stored templates.

Main Results:

• The proposed method achieved an 86.04% true acceptance rate.
• A low false acceptance rate of 2.73% was recorded.
• Authentication is completed within 6 seconds, consuming approximately 2000 mW of power.

Conclusions:

• Cognitive cloud models of heartbeats offer a viable biometric authentication solution for mobile devices.
• The system demonstrates promising accuracy and efficiency for secure mobile access.
• This approach provides an alternative to conventional, less secure authentication techniques.