Existing digital identification systems are often vulnerable to attacks as they are commonly based on authentication methods such as passwords, PIN codes, biometric data, etc., which can be easily forged or compromised. In this letter, we propose a digital identification system based on a unique set of user biometric data processed by Artificial Intelligence (AI) and fuzzy extractors to generate a cryptographically secure password linked to a unique Non-Fungible Token (NFT). Our system provides decentralized identification based on blockchain technology, which eliminates problems associated with centralized identification systems, such as cyber-attacks on central servers and data leaks. Our proposed system offers a higher level of user identification security by linking the user to their data through a unique NFT, generating a cryptographically secure password, and processing large volumes of biometric data using AI and fuzzy extractors. Our system provides a solution to many of these problems, making it important and relevant to many industries, including banking, medical, and financial sectors. The use of decentralized storage of information on the blockchain provides a high level of protection against hacking and reduces the likelihood of data breaches, making our system particularly relevant in the field of financial services and personal data protection.
The need for a global decentralized digital
identity has arisen due to the increasing
demand for identity verification and
authentication in various fields such as
egovernment, e-commerce, and online services
[1–2]. However, the centralized nature of
traditional identification systems creates
several problems such as lack of functional
compatibility, privacy risks, and vulnerability
to cyberattacks [
In this paper, we propose the development
of a global decentralized digital identification
system based on Non-Fungible Tokens (NFTs)
technology [7–8], Artificial Intelligence (AI)
methods [
The proposed system is designed to be
decentralized, relying on a network of nodes
that will be responsible for storing and
verifying identification data [
Overall, the proposed system is tailored for use in embedded systems, where security and efficiency are of paramount importance. By utilizing NFT technology, AI methods, and fuzzy extractors, we aim to develop a decentralized identification system that can be used across various embedded platforms and services, providing an efficient and secure means of identity verification.
The problem of decentralized digital identification has been addressed in many related articles.
The proposed Idenx system in [
The proposed Casper platform appears to
provide a secure and decentralized solution for
digital identity management using blockchain
and a self-sovereign identity-based approach
[
The paper [
While the proposed digital identity
platform “Trust Pass” [
The article [
The papers [
The proposed SmartDID system [
The papers [9, 24] describe systems that use NFTs and smart contracts for document traceability, which aims to prevent fraud, corruption, tampering, and counterfeiting in identity and document verification. These approaches have similarities with our proposed decentralized digital identity system, as both use NFTs and smart contracts to ensure a secure and transparent solution. However, compared to our solution, they lack details on AI-based biometric processing, fuzzy extractors, decentralized architecture, and user control over data, which may result in weaker security measures and limited user privacy.
There are numerous works and studies in the field of decentralized identity, but many unresolved issues remain. Our decentralized digital identity system based on NFT, AI methods, and fuzzy extractors addresses some of these issues, including the ability to process large volumes of biometric data using AI and fuzzy extractors, generating cryptographically secure passwords, and utilizing blockchain technology for decentralized identity management. Additionally, our system offers the convenience of remote access to personal data and eliminates the need for special equipment.
The methodology used in this paper for global digital identification involves the integration of blockchain technology, NFTs, AI methods, and fuzzy extractors.
Blockchain is a decentralized and immutable digital ledger that can be used to securely store and manage identity-related information. By using blockchain, it is possible to create a tamper-proof and transparent record of identity-related transactions.
NFTs are unique digital assets that are used to represent a particular object or item. In the context of digital identification, NFTs can be used to represent a person’s identity and associated attributes. The use of NFTs ensures that each identity is unique, noninterchangeable, and can be easily verified.
AI methods such as deep learning-based face recognition can be used to authenticate a person’s identity based on biometric data. The use of AI can improve the accuracy and efficiency of the authentication process, which is crucial for large-scale digital identification systems.
Finally, fuzzy extractors can be used to extract a secure cryptographic key from biometric data, such as a fingerprint or iris scan. This key can then be used to securely verify a person’s identity without exposing their biometric data. Fuzzy extractors can help address privacy concerns associated with the use of biometric data in digital identification systems. The integration of these technologies provides a robust and secure framework for global digital identification.
The identification blockchain system proposed in this paper is based on distributed ledger technology, which ensures transparency and security in digital identification. The main components of the system are digital wallets, smart contract signatures, data transmission protocols, and NFT creation and verification systems. Digital wallets are used to store digital assets and personal user data, while smart contract signatures automate the identity verification process and the agreement of terms for the use of personal data. Data transmission protocols are used to ensure the secure transfer of information between users and the system, and the NFT creation and verification system ensures the uniqueness and tamper-proof nature of digital identification.
The blockchain identification system works as follows: when a user creates an account in the system, their data is hashed and stored on the blockchain. The system then creates an NFT, which is linked to this personal data, and sends it to the user’s digital wallet. Every time the user wants to use their data to authenticate themselves in online services or other systems, the system requests access to the corresponding NFT. If the user grants permission to use the NFT, the smart contract signature is used to verify the authenticity of the user’s data.
To ensure security and protect the confidentiality of user data, the system uses AI methods such as machine learning and deep learning to detect and prevent fraud and cyberattacks. Fuzzy extractors are used to compress and store user biometric data in a secure format [25–26].
Thus, the blockchain identification system developed in this paper represents an innovative and secure approach to global digital identification. 4.1.
Let U be the set of users, D be the set of their data, H : D → H(D) be the hash function for hashing personal data, W be the set of digital wallets, N be the set of NFTs, T be the set of smart contract rules, and P be the set of data transmission protocols.
Then, the identification system can be represented as a tuple (U, D, H,W, N,T, P) , where: each user u U has their account associated with their data d D , which is hashed by the function H and stored in the blockchain; each user also has their digital wallet wW , where their data and NFTs tied to them are stored; each NFT n N is linked to a specific user and is used for authentication and authorization when requesting access to their data; each smart contract t T represents rules for using the user’s data and automatically verifies the authenticity of that data; each data transmission protocol p P ensures secure transmission of information between the user and the system; when using the system, each user u provides access to the corresponding
NFT n , which is automatically verified by the smart contract t 4.2.
When creating an account in the identification system, the user provides their data du , which is hashed using a hash function H , i.e., hu = H(du) . The resulting hash value hu is then stored in the blockchain registry of the system. Mathematical notation of the step: hu = H(du) , hu {0,1}l , du D , u U , where: hu is the hash value of the user’s data u ; H is the hash function; du is the user’s data; D is the set of personal data; U is the set of users; l is the length of the hash value (in bits).
To generate unique digital identifiers for users, biometric data is used. Let B be the set of fuzzy biometric data, F : B → F(B) be an AI function that extracts essential features from the fuzzy data, E : F(B) → E(F(B)) be a fuzzy extractor that processes the features and generates a cryptographically secure password (key), and N be the set of NFTs, where each element n N is associated with a user’s password.
The procedure for creating an NFT for the user can be described as follows: 1. The user u provides their fuzzy biometric data b B . 2. The AI function F is used to extract essential features from the user’s biometric data: f = F(b) . 3. The fuzzy extractor E processes the extracted features f and generates a cryptographically secure password p : p = E( f ) . 4. A unique NFT n is created, associated with the user’s password p : n = createNFT( p) . 5. The NFT n is sent to the user’s digital wallet for further use in the authentication process.
Thus, this process ensures the uniqueness and security of the user’s identifier, using their fuzzy biometric data and a cryptographically secure password.
To implement digital identification in a decentralized system, we will use smart contracts that provide automatic verification of users’ data and coordination of the conditions for their use. Specifically, we will use smart contracts that can verify the authenticity of a user’s identifier based on the NFT associated with their data. These contracts will contain rules for the use of personal data and automatically verify them for compliance with these rules when requesting access to this data.
Also, to ensure the security and protection of data, we can use smart contracts that can perform operations for processing and storing users’ data in a secure and protected manner, using cryptographic methods and data transmission protocols.
and NFT generation is a part of the software code that is executed upon user account creation in the identification system. Its main task is to create a unique NFT that will be used for user authentication in the future.
The structure of the smart contract can be described as follows: Contract : RegisterUser(u,b) → n , where: u is a user who creates an account; b is fuzzily defined biometric data of the user; n is a unique NFT linked to the user’s password.
The smart contract itself consists of three main steps: 1. Generation of a cryptographically secure password based on the user’s biometric data: p = E(F(b)) , where F there is an AI function that extracts essential features from fuzzy biometric data and E is a fuzzy extractor that processes the extracted features and generates a cryptographically secure password. 2. Creation of a unique NFT linked to the generated password: n = createNFT( p) , where createNFT is a function that creates a unique NFT based on the cryptographically secure password. 3. Saving the created NFT in the user’s digital wallet and returning it as the result of the smart contract execution: saveNFT(u,n) , where saveNFT is a function that saves the created NFT in the user’s digital wallet.
Thus, the smart contract for registering an account and generating an NFT ensures the security and uniqueness of the user’s identifier based on their fuzzily defined biometric data and a cryptographically secure password.
Fig. 1 depicts a UML sequence diagram for the registration of a user account and the generation of an NFT.
authenticity of a user’s identifier based on NFT consists of the following elements: • A data structure containing information about the user and their NFT: Auth(u) = (nu,du,tu) where: nu is the identifier of the NFT associated with the user’s data u ; du is the personal data of the user u , which must correspond to the data associated with the NFT nu ; tu is the time of the last update of the user's data u . • Method for authenticating a user’s data: authenticate(u,n,d) → bool where: u is the user attempting to access their data; n is the NFT identifier provided by the user u for authentication. • d is the personal data that the user u wants to use for authentication. Algorithm for updating a user’s data: update(u,d) → bool where: u is the user whose personal data needs to be updated; d is the new personal data of the user.
When attempting to access a user’s data u , the smart contract first extracts information about the user u and their NFT nu from the Auth(u) structure. Then it calls the authenticate(u,n,d) method to authenticate the personal data provided by the user u , using their NFT nu and stored personal data du . If the data is authentic, the smart contract returns the value true and updates the time of the last update of the personal data tu in the Auth(u) structure. If the data is invalid, the smart contract returns the value false . If there is a need to update a user’s data, the smart contract calls the update(u,d) method to update the personal data in the Auth(u) structure and the associated NFT. If the update is successful, the method returns the value true , otherwise—the value false . Thus, the structure of the smart contract for verifying the authenticity of a user’s identifier based on NFT consists of three main elements: data structure, authentication method, and personal data update method.
In Fig. 2, a UML sequence diagram for the user identity authenticity verification based on NFT is presented.
The smart contract for processing and storing personal user data includes the following components: 1. A data structure for storing personal data: let Du be the set of personal data of the user u . We will use a data structure, such as an array or hash table, to store the user’s data. Suppose we use the hash table DHTu , where each element stores a key-value pair (k,v) , where k is the identifier of the data and v is the data itself. Thus, DHTu[k] = v . 2. An encryption function for personal data: let E be the encryption function that converts personal data d Du into encrypted form e = E(d) . 3. A data structure for storing encrypted personal data: let Eu be the set of encrypted personal data of the user u . We will use a data structure, such as an array or hash table, to store the encrypted personal data of the user. Suppose we use the hash table EHTu , where each element stores a key-value pair (k,v) , where k is the identifier of the data and v is the encrypted data. Thus, EHTu[k] = v . 4. An authenticity verification function for user data: let C be the function for verifying the authenticity of user data. The smart contract will use the function C to verify the authenticity of the user’s data that will be requested by other network participants.
Thus, the smart contract for processing and storing personal user data can be formalized as follows:
DHTu[k] = d , where d Du ; EHTu[k] = e , where e = E(d) ;
1, if E(key) = e C(e,key) = 0, else ; where: k is the identifier of the data; key is the key for decrypting the data; Du is the set of personal data of user u ; E is the encryption function; Eu is the set of encrypted personal data of user u ; C is the function for verifying the authenticity of user data; DHTu is the hash table for storing personal data.
This diagram depicts the interaction between the user, the personal data processing and storage smart contract (DSC), the digital signature smart contract (DSS), and the Blockchain Registry (BR).
The main scenario starts with the user’s request to store personal data, to which the DSC encrypts the data and stores them in the blockchain registry. The DSC then returns the data identifier to the user. When the user requests access to personal data, the DSC finds the data in the registry, decrypts it, and requests a digital signature from the DSS. The DSS verifies the authenticity of the user and signs the data, after which the DSC returns them to the user.
The UML Component diagram (Fig. 4) provides a comprehensive visualization of the main components and their relationships within the decentralized digital identity system. Each component plays a crucial role in the overall functionality and security of the system.
• User Interface. The User Interface serves as the point of interaction for users and service providers. It receives input, displays relevant information, and facilitates communication with other components in the system. • Blockchain System. The Blockchain System is responsible for managing the decentralized network, ensuring the immutability and security of the digital identity system. It connects the nodes in the network and maintains the overall architecture. • Node. Nodes are responsible for storing and verifying identity data within the blockchain network. They contribute to the decentralized nature of the system and provide redundancy and resilience against attacks. • Smart Contract. Smart Contracts automate and enforce the rules and processes of the digital identity system. They interact with the Identity Data Storage and NFT Management components to create, verify, and manage digital identities. • AI Biometric Processor. The AI Biometric Processor processes biometric data, such as facial recognition, to enable secure and efficient identity verification. It extracts unique biometric features and passes them to the Fuzzy Extractor. • Fuzzy Extractor. Fuzzy Extractors generate cryptographic keys based on the biometric features provided by the AI Biometric Processor. These keys enhance the security of the system by providing an additional layer of protection for identity data. • Identity Data Storage. The Identity Data Storage securely stores user data and is accessed by Smart Contracts. It maintains user information while ensuring data privacy and protection against unauthorized access. • NFT Management. NFT Management creates unique digital identities using NFTs. These digital identities cannot be replicated or duplicated, thus preventing identity theft and fraud. • Biometric Features. Biometric Features represent the unique characteristics of users, such as fingerprints or facial features. They are used as input for the Fuzzy Extractor to generate cryptographic keys. • Cryptographic Keys. Cryptographic Keys are generated from biometric features and are used to encrypt and decrypt identity data. These keys provide a secure mechanism to protect user data and authenticate users within the system. • User Data. User Data contains the personal information of users and is protected by cryptographic keys. It is securely stored in the Identity Data Storage and can be accessed only through proper authentication. • Service Provider. Service Providers interact with the system to authenticate and authorize users for various services. They rely on the decentralized digital identity system to ensure secure and efficient access to their platforms.
In summary, the UML Component diagram offers a clear representation of the decentralized digital identity system’s components, their relationships, and their roles. The diagram highlights the system’s decentralized architecture, which leverages blockchain technology, AI-based biometric processing, fuzzy extractors, and NFTs to create a secure and efficient digital identity management solution.
Solution B
Yes Partially
No Yes Yes
No Partially
Solution C Partially
No
Yes Partially
No No No
In this research, we have addressed the pressing need for a secure, efficient, and usercentric digital identity management system. Our proposed solution leverages the advantages of blockchain technology, AI techniques for biometric data processing, and fuzzy extractors for cryptographic key generation. This unique combination of technologies offers a novel approach to digital identity management that sets it apart from existing solutions in the market.
The key innovations of our solution include: • Decentralization. Our digital identity system is built on a decentralized architecture, eliminating reliance on a central authority or organization. This ensures better resilience against attacks and minimizes the risk of data breaches and identity theft. • Enhanced Security. The use of AI techniques for processing biometric data, fuzzy extractors for generating cryptographic keys, and NFTs for creating unique digital identities significantly enhances the overall security of the system. These features prevent unauthorized access, duplication, and forgery of identity data. • Improved Privacy. By allowing users to maintain full control over their digital identities and sharing only the necessary information with service providers, our system ensures better privacy for users. It also reduces the risk of unauthorized data access and misuse. • Cross-platform Compatibility. Our solution is designed to be used across various platforms and services, enabling seamless integration with embedded systems in different sectors such as finance, healthcare, and e-commerce.
To illustrate the advantages of our proposed digital identity system compared to other known solutions, we present the comparison in Table 1.
As seen in the comparison table, our proposed digital identity system stands out in terms of decentralization, enhanced security, improved privacy, cross-platform compatibility, AI-based biometric processing, fuzzy extractor integration, and the use of NFTs for creating unique digital identities. While other solutions may offer some of these features, our approach provides a more comprehensive and robust solution to digital identity management.
In the comparison table, we referred to three hypothetical digital identity management solutions: • Solution A: Centralized Digital Identity Management Systems (e.g., traditional single sign-on systems). • Solution B: SSI platforms (e.g., Sovrin, uPort). • Solution C: Federated identity management systems (e.g., OAuth, OpenID Connect).
These examples represent well-known digital identity management solutions currently in use. While each of these solutions has its merits, our proposed decentralized digital identity system offers a more comprehensive approach, combining the advantages of blockchain technology, AI-based biometric processing, fuzzy extractors, and NFTs for creating unique digital identities.
Existing identification systems often rely on authentication methods such as passwords, PIN codes, biometric data, and so on [27–28]. However, these methods are often easily counterfeited or compromised [29–30].
The proposed NFT-based digital identification system, utilizing AI and fuzzy extractor methods, provides a higher level of user identification security. It uses a unique set of user biometric data, processed with AI and fuzzy extractors, to generate a cryptographically secure password linked to a unique NFT.
Firstly, our system utilizes a unique NFT to link the user with their data. This eliminates the possibility of data forgery or alteration and ensures the security of data storage and transmission.
Secondly, by utilizing AI and fuzzy extractors, our system can process large volumes of biometric data and extract the most significant features from them. This improves identification accuracy and reduces the probability of errors, which is especially This project has received funding from the European Union’s Horizon 2020 research and innovation program under the Marie Skłodowska-Curie grant agreement No. 101007820—TRUST. This publication reflects only the author’s view and the REA is not responsible for any use that may be made of the information it contains. important for systems used in the banking and medical sectors.
Thirdly, our system generates a cryptographically secure password linked to the user’s unique NFT, ensuring the security of data transmission and storage. This is critical for systems used in the financial services and personal data protection fields.
Finally, our system provides a high degree of convenience and accessibility for users, as it does not require specialized equipment and provides remote access to personal data.
Thus, our NFT-based digital identification system utilizing AI and fuzzy extractors provides reliability, security, and user convenience, making it an attractive option for a wide range of users and organizations.