Source-linked AI summary
Developing a Dynamic Multi-factor Authentication System for Securing Smart Organizations
Mohsen Saffar, Hamid Reza Naji
TL;DR
Smart organizations’ growing use of digital technologies makes securing their systems important. The paper proposes a Zero Trust MFA system that randomly selects biometric factors using a Markov chain; its analysis reports a balance between security and efficiency and suitability for e-government.
Problem
The paper addresses securing smart organizations as reliance on digital technologies grows and cyber threats evolve.
Source: §I · p. 1Method
The system uses a Markov chain to randomly select fingerprint, iris, or face authentication within a Zero Trust MFA framework.
Source: §I · p. 1, §III · p. 3Results
The authors report that the system balances security and efficiency and is particularly suitable for e-government applications.
Source: §V · p. 4Takeaways & Limitations
The paper presents the system as a flexible, real-time defense against evolving threats in smart organizations.
Source: §V · p. 4
Abstract
from arXiv · showhide
As smart organizations increasingly rely on technologies like AI, big data, and IoT, securing these systems is crucial. This research introduces a dynamic multi-factor authentication (MFA) system within the Zero Trust Architecture (ZTA) framework, which enforces continuous verification. The system randomly selects biometric authentication factors, such as fingerprint, iris, and face recognition using a Markov chain model to ensure adaptability and reduce predictability in security measures. Using CASIA biometric datasets and AES-128 encryption, the system securely manages user data and authentication. Performance analysis shows that the dynamic selection process balances security and efficiency, making it suitable for smart organizations, particularly in e-government contexts. This MFA system strengthens cybersecurity by providing a flexible, real-time defense against evolving threats.
I. INTRODUCTION
Smart organizations use digital technologies to support e-government, making security essential as reliance on these tools and data grows. The introduction frames Zero Trust and dynamic defense as responses to evolving cyber threats.
I. INTRODUCTION
- Smart organizations use AI, big data, and IoT to support government operations and e-government services.These technologies are described as helping process data and handle public resources and citizen requests.Source: §I · p. 1
- Greater reliance on digital tools exposes smart organizations to risks including hacking, data leaks, and unauthorized access.The passage notes that security failures can cause financial loss and reputational damage.Source: §I · p. 1
- Zero Trust requires continuous verification and restricts access to what each user needs, rather than treating internal networks as inherently safe.The introduction presents multifactor authentication and real-time monitoring as measures within this framework.Source: §I · p. 1
- The paper presents dynamic defense within Zero Trust as an approach to securing smart organizations in support of e-government.Dynamic defense is described as adapting to changing threats through real-time data and adaptive security measures.Source: §I · p. 1
II. REVIEW
The review positions MFA within Zero Trust’s continuous verification approach and describes dynamic defense as a response to changing cyber threats. It motivates randomized authentication using Markov chains.
II. REVIEW
- Zero Trust Architecture applies “never trust, always verify” through continuous authentication and authorization for access requests.The review also lists dynamic policies, strict authentication, granular access control, and continuous monitoring among its principles.Source: §II · pp. 1–2
- MFA combines multiple credential types and is presented as a fit for Zero Trust’s emphasis on stringent verification.The review groups authentication factors as knowledge, possession, and inherent factors, and lists biometrics such as face, iris, and fingerprint recognition.Source: §II · p. 2
- Dynamic defense continuously evaluates and adjusts system components to adapt to evolving threats rather than relying on fixed measures.The review identifies Moving Target Defense as altering system parameters or authentication dynamics, and describes dynamic authentication as part of this strategy.Source: §II · p. 2
- The review motivates randomness in MFA and describes Markov chains as a way to model system behavior and transitions in authentication.It also notes their use in analyzing system dynamics during impersonation attacks.Source: §II · p. 2
- The paper uses Markov chains to randomly allocate MFA authentications as a dynamic Zero Trust method for smart organizations.The passage presents this modeling choice as a way to address evolving threats.Source: §II · p. 3
III. MATERIAL AND METHODS
The system randomly selects among fingerprint, iris, and face authentication rather than asking users to choose a factor. It uses CASIA datasets, AES-128-encrypted biometric features, and a Markov-chain selection process.
III. MATERIAL AND METHODS
- The system randomly selects fingerprint, iris, or face authentication, so users do not choose the factor themselves.The flowchart depicts the implemented MFA system, while the method text identifies the three biometric factors.Source: §III · p. 3, Figure 3 · p. 4
- The evaluation uses CASIA iris images from 700 individuals, fingerprint images from 500 individuals, and face images from 500 individuals.The collections contain 54,601 iris images, 20,000 fingerprint images, and 2,500 face images, respectively.Source: §III · p. 3
- Usernames and extracted fingerprint, iris, and face features are stored in four SQL database columns, with biometric data encrypted using AES-128.After username verification, a Markov chain selects an initial factor using a 4x4 transition matrix and a sequence of ten random numbers.Source: §III · p. 3
- Fingerprint authentication extracts image features and compares them with the decrypted fingerprint features stored in the database.The iris pipeline identifies the eye, converts the image to grayscale, filters noise, detects edges, segments the iris, and extracts features for comparison with decrypted stored features.Source: §III · p. 3, Figure 1 · p. 3
- Face authentication first detects the face using a Haar Cascade, then extracts distinguishing features for comparison with decrypted stored face data.Figure 2 is captioned as showing stages of face identification.Source: §III · p. 3, Figure 2 · p. 3
- The system compares extracted image features using OpenCV’s Brute Force Matcher and Hamming distance before granting access.Figure 3 is captioned as a flowchart of the implemented MFA system.Source: §III · p. 3, Figure 3 · p. 4
IV. ANALYSIS OF THE MODEL
The analysis reports manual authentication-time evaluation across 200 samples and describes the system’s time-complexity components. Overall complexity depends on random selection and the combination of factors executed in sequence.
IV. ANALYSIS OF THE MODEL
- Authentication-time evaluation used 200 samples and reports individual-agent averages, authentication-method selection time, and overall average time after four tries.The evaluation was performed manually because user-interactive procedures are difficult to gauge.Source: §IV · p. 4, Table I · p. 4
- The iris process includes grayscale conversion, edge detection, segmentation, feature extraction, and descriptor matching.The passage gives its complexity as O(k·n), where n is image size and k is the number of keypoints; the supplied expression is corrupted, so no exact formula can be recovered.Source: §IV · p. 4
- Fingerprint processing involves feature extraction, data decryption, and matching, while face processing involves image processing and descriptor matching.The factor-selection stage generates a transition matrix and moves between states based on probability distributions; the overall expression combines the respective section complexities.Source: §IV · p. 4
- Overall complexity depends on the random selection process and the combination of different factors executed in sequence.The analysis also presents the system’s time complexity in Table II.Source: §IV · p. 4, Table II · p. 4
V. CONCLUSION
The conclusion presents a dynamic MFA system within ZTA that randomly selects biometric factors using a Markov chain. It reports a balance between security and efficiency and positions the system for smart-organization and e-government use.
V. CONCLUSION
- The paper presents a dynamic MFA system integrated into ZTA to enhance security for intelligent organizations.The conclusion describes it as adaptable and reports that performance analysis validates versatile, immediate protection against developing threats.Source: §V · p. 4
- A Markov chain randomly selects fingerprint, iris, or face recognition to promote adaptability and reduce predictability.The conclusion also names CASIA biometric data and AES-128 encryption for handling user data and authentication.Source: §V · p. 4
- The conclusion reports a balance between security and efficiency and identifies e-government as a particularly suitable application context.It characterizes the approach as aligned with ZTA’s priorities of ongoing verification and low reliance on trust.Source: §V · p. 4