A Novel Method For Secure Finger Vein Biometric during User Authentication Process Based On Blockchain-PSO-AES Techniques
This research aims to develop a new method to secure finger-vein (FV) biometric information during user authentication process against different types of network security attacks. Hence, a novel method was proposed to solve the existing problem in biometric authentication systems, which is the leaka...
Saved in:
| Main Author: | |
|---|---|
| Format: | thesis |
| Language: | eng |
| Published: |
2019
|
| Subjects: | |
| Online Access: | https://ir.upsi.edu.my/detailsg.php?det=6799 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | This research aims to develop a new method to secure finger-vein (FV) biometric information during user authentication process against different types of network security attacks. Hence, a novel method was proposed to solve the existing problem in biometric authentication systems, which is the leakage of biometric information. Our proposed method meets information security definition standard requirements (CIA). The research design was carried out in two stages. In the first stage, the researcher developed a new merge algorithm in order to produce a new hybrid biometric pattern by merging Radio Frequency Identification (RFID) features with FV biometric features to increase the randomization, and enhance the security and structure of the new pattern. While, in the second stage, a new secure user verification method is developed based on blockchain technique, AES algorithm and a new steganography method that is based on Particle Swarm Optimization (PSO) algorithm. A dataset was used comprising of 6000 samples of FV images. The experimental results showed the effectiveness of the proposed authentication method. Where, this method achieved a high level of performance accuracy of 97.9% during the implementation of this method using FV biometric for 106 users. Furthermore, the proposed method has an advantage 55.56% higher than the benchmark method as a result of the comparison between the proposed method and benchmark method, depending on some security issues where, the proposed method covered 100% of these issues. Whereas, the benchmark method covered only 44.44% as indicated in Chapter 5. Moreover, the results showed that the structure of hybrid pattern is robust and immune towards detection by the attacker, and is flexible to being cancelable and reconstructed again in case of loss of this pattern. More so, the proposed method showed high resistance against spoofing and brute-force attacks. Clearly, such empirical results suggest that the FV information are confidential, integrated and available only for specific authorized persons only in the entire steps of the authentication process.The implication of this study is that, the proposed method can be applied in decentralised network architectures by eliminating the central point, and addressing the network failure problem and security at the same time. |
|---|