DEV Community

lyc233333
lyc233333

Posted on

Hongmeng Next user authentication advanced application: from biometrics to distributed collaboration🔄

On the basic functions, Hongmeng Next's user authentication system has built a safer and smarter authentication system through biometric enhancement, distributed collaboration and industry customization.This article analyzes three major advancement directions, with practical cases~

1. Biometric authentication: upgrade from "identification" to "anti-counterfeiting" 🚀

1. In-depth integration of live detection technology

  • Fingerprint live detection: Combined with the capacitance sensor to detect the change in the skin capacitance value and determine whether it is a living finger. 
// Enable fingerprint live detection
  biometrics.setLivenessDetection({  
    type: biometrics.BiometricType.FINGERPRINT,  
    enable: true  
  });
Enter fullscreen mode Exit fullscreen mode
  • Face Anti-Counterfeiting Enhancement: Through 3D structured light + infrared camera, identify the differences between real people and photos/videos (such as eye reflection dynamics).

2. Multimodal Fusion Certification

Scenarios Certification Combination Solutions Security Level
Financial Transactions Iris + Dynamic Password + Device Fingerprint ★★★★★
Enterprise Access Control Face + Finger Vein + NFC Industrial Card ★★★★☆
Smart Home Fingerprint + voice command (voiceprint recognition) ★★★☆☆

Code example: voiceprint + fingerprint dual authentication

async function multiModalAuth() {  
const voiceValid = await verifyVoiceprint(); // Voiceprint verification
  const fingerValid = await biometrics.verifyFingerprint();  
  return voiceValid && fingerValid;  
}
Enter fullscreen mode Exit fullscreen mode

2. Distributed authentication: a cross-device "trust network" 🌐

1. Transmission of authentication chain between devices

  • Scenario: After mobile phone certification, the tablet/car machine will automatically trust
  • Technical Implementation:
  • Generate a short-term trust token after successful mobile phone authentication (valid for 5 minutes)
  • Encrypt the transmission of tokens to other devices through NearLink protocol
  • Target device verification token legality (real-time verification with the certification center)

2. Context-aware authentication

  • Dynamic permission adjustment: 
// Adjust the certification strength according to the equipment position
  const location = await getDeviceLocation();  
  if (location.inCompany) {  
allowLowAuth(); // Allows fast fingerprint authentication
  } else {  
requireMultiFactor(); // Forced password + verification code
  }
Enter fullscreen mode Exit fullscreen mode
  • Repeat-free authentication for continuous operation: In the same user session (such as within 30 minutes), there is no need for repeated authentication to operate across devices, only the device trust status is verified.

3. Industry customization: "Scenario-based adaptation" of safety and experience⚙️

1. Financial Certification Program (Bank App)

  • Triple Protection:
  • Login: Fingerprint + Device Binding (IMEI + Chip ID)
  • Transfer: iris + dynamic SMS verification code
  • Exception reminder: Login from another place triggers facial recognition review
  • Audit Log: Record the certification time, IP, device model, and certification method to meet the requirements of warranty 2.0.

2. Multi-role certification in the education industry

  • Student end: Quick registration of mobile phone number + graphic verification code, supporting face login (class sign-in scenario).
  • Teacher end: Work number + fingerprint + teacher qualification certificate OCR recognition, access courseware management and other sensitive functions.
  • Management side: USB Key+iri+second approval process, operation permissions are bound to CA certificate.

3. Performance optimization practice

  • Biometric Algorithm Acceleration: Using NPU hardware to accelerate fingerprint feature comparison, the time taken to reduce from 500ms to 200ms.
  • Certification Cache Policy: High-frequency user authentication information is cached to memory (LRU elimination mechanism), reducing database query pressure.

4. Future trends: Insensitivity certification and active safety 🛡️

1. Invisible authentication technology

  • Behavioral Biometrics: Analyze the user's gesture habits when holding the mobile phone through a gyroscope + accelerometer to achieve "quiet default certificate".
  • Environmental Perception: Automatically determine the trusted environment based on commonly used WiFi and Bluetooth devices (such as smart watches) to reduce authentication pop-ups.

2. Active security defense

  • Risk Prediction: Detect account abnormalities based on user behavior trajectory (such as changes in typing speed), and trigger secondary authentication in advance.
  • Dynamic Certificate Update: The biometric template is automatically updated after every 3 authentications are completed to prevent the risk of template leakage.

Summary: "Three-dimensional evolution" of certification system

  1. Safety Dimension: From single authentication to multimodal fusion, build body protection
  2. Experience dimension: From passive input to active perception, realizing "invisible authentication"
  3. Scenario dimension: From device islands to distributed collaboration, open up cross-end trust chains

Top comments (0)