Dynamic MFA randomly selects biometrics, improving security and efficiency
The claim
The proposed dynamic multi-factor authentication system strengthens cybersecurity for smart organizations
Published as: “Developing a Dynamic Multi-factor Authentication System for Securing Smart Organizations”
So what
The system could be adopted in e-government services where occasional authentication latency is acceptable, but its practical impact is unclear without security effectiveness data.
The five-minute version
The authors report that their dynamic MFA system authenticates users in 1246 ms with fingerprint, 2867 ms with iris, and 2943 ms with face, while the random selection of the factor takes only 36 ms. They claim this balances security and efficiency for smart organizations, especially e-government. The evidence consists solely of average execution times measured on 200 samples; no security-effectiveness metrics or comparisons to other MFA approaches are provided.
What they found 3
- Fingerprint authentication takes 1246 ms on average
“Fingerprint 1246 ms”
- Iris authentication takes 2867 ms on average
“Iris 2867 ms”
- Random selection of the authentication factor takes 36 ms
“Random Selection 36 ms”
What undercuts it the paper’s own results
- Face authentication takes 2943 ms, which may be too slow for real-time use and could hinder usability
- The evaluation is performed manually on only 200 samples, providing limited evidence of security improvement
Stated limits 2
- performance evaluation in this case is done manually.
- It is not easy to gauge how well user-interactive procedures are performing, though.
What’s new here 4
- Randomly selects a biometric factor (fingerprint, iris, or face) using a Markov chain model
- Integrates the dynamic MFA into a Zero Trust Architecture framework
- Uses public CASIA biometric datasets encrypted with AES-128 for storage
- Provides manual timing evaluation of each authentication factor and the selection process
Artifacts & methods 5
- Markov chain model for factor selectionprimitive
- Dynamic multi-factor authentication systemsystem
- CASIA biometric datasetsdataset
- AES-128 encryption of stored featuresprimitive
- BFMatcher (OpenCV) with Hamming distanceprimitive