Huawei

wangdonghui124@huawei.com

Huawei

liufei19@huawei.com

Huawei

lilun20@huawei.com

Security Secure Patterns for Internet CrEdentials public key service SPICE provides a selective disclosure mechanism of credentials from issuer. However, future network services may be built on the trust between multiple entities. Obtaining the public key of multiple issuers for a verifer from potential multiple sources can be complex. In this contribution, an optional public key service is proposed in SPICE architecture for the issue of obtaining the public keys of the issuers from multiple trusted entities. The basic function of public key service is proposed including public key registration, token verification, and a potential implementation such as the distributed ledger. We hope that the proposed contribution can be used as infomative for SPICE regarding to the token validation procedure.

As Web3.0 evolves, future society and networks will enables greater cross-field collaboration. Consequently, establishing trust between different domains is crucial. Consider this scenario: Device manufacturer A produces device a, which is sold to network operator X. During device a's deployment, it must communicate with network operator X's existing device x. Here, device manufacturer A and network operator X are issuers, while device a and device x are holders that also act as verifiers. In the current PKI system, there are three common approaches to achieve trust between a and x: 1. Manual Configuration: Both device a and device x maintain a trusted list. Administrators add the peer issuer's credentials to these lists. For example, device a adds network operator X's credentials, and device x adds device manufacturer A's. 2. Trusted Third Party: A trusted third-party T acts as the issuer to endorse their public keys for both A and X, creating trust chains T->A->a for x and T->X->x for a. 3. Cross-Credentials: A issues a credential for X's public key, and X does the same for A's, forming trust chains A->X->x for a and X->A->a for x. However, future network is very likely to face a much more dynamic multi-stakeholder environment. Network operator X may want to dynamically engage with multiple device providers, and device provider A may supply devices for multiple network providers. In this situation, the cross-certification approach will be inefficient. Therefore, this document proposes a method to simplify the complexity of cross-domain trust establishment by introducing a distributed trusted platform, referred to as the Public-Key-Service-Provider (PSKP). This new entity is designed to handle the registration, updating, revocation, and querying of public keys of the issuers in a multi-stakeholder environment.

The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in .

Figure 1 illustrates the traditional approach where issuers, i.e. CAs, must establish pairwise mutual trust. This process incurs high workload, and maintaining trust will require even more effort when public key updates and revocations occur.

| | | Issuer A |<------------ | Issuer B | +---------------+ +---------------+ | ^ | ^ | | | | | | | | | | | | v v | +---------------+ +---------------+ | |------------> | | | Issuer C |<------------ | Issuer D | +---------------+ +---------------+ Figure 1 Establish mutual trust between every two issuers. ]]>

When we can establish mutual trust among multiple parties based on PKSP, as shown in Figure 2, the issuer only needs to publish registration and revocation messages of public keys/credentials to the public-Key-Service-Provider(PKSP), which can then be accessed by other issuers or verifiers when they are performing verification, then, during the verification, the verifiers may verify the token such as RFC 9207 and RFC 9449 , or the selective disclosed token such as SD-CWT()

During the registration process, identity verification is usually carried out to ensure that only legitimate entities can add public keys. The service checks on the format and validity of the public key to ensure it meets the system requirements. The registration process between the issuer and the PKSP is as follows: 1. Submission by the Issuer. The issuer initiates registration by sending a request to the PKSP. This request contains the issuer's public keys. Meanwhile, the issuer declares the purposes of these keys, such as for token issuance, revocation, and so on. Along with the public keys, detailed descriptive information about the issuer is provided. This includes the issuer's name, identifier, the expiration time of the public keys, as well as a declaration of the claims of the tokens the issuer can provide. 2. Validation by the PKSP. Upon receiving the registration request, the PKSP first validates issuer's public key. The descriptive information provided by the issuer is then examined for accuracy and completeness. The PKSP may cross-reference this information with existing databases or industry-recognized sources to verify the issuer's legitimacy. 3. Response from the PKSP. If all the information is found to be valid and compliant, the PKSP sends a registration response to the issuer. This response includes a confirmation of successful registration. The PKSP may also provide additional information like the storage location of the issuer's public key within the PKSP. In case the registration request is not approved, the PKSP sends a rejection notice to the issuer, stating the reasons for rejection. 4. Completion of Registration. Once the issuer receives a positive response from the PKSP, the registration process is considered complete. The issuer can then start using the registered public key for its intended purposes and issue tokens according to the specifications provided during registration. The PKSP updates its records to reflect the newly registered issuer and is ready to support the verifiers in matters related to public key querying.

As a public public key service platform, PKSP accepts public key registration from issuers, stores the registered public keys securely and in a non-tamperable manner. When a verifier queries for a key, it provides the corresponding public key.

| Verifier | | | +----------+ Figure 1 System architecture of a PKSP ]]>

When a verifier is performing SD-CWT verification, it can query the public key information of relevant issuers through the PKSP. It supports various query methods, such as precise queries based on the identity identifier of the entity, credential number, etc., to quickly obtain the required public key. The following steps shall be performed between verifiers and PKSP: 1. Token Receipt. The Holder sends the token to the Verifier. 2. Request for public key from PKSP. After receiving the token, the verifier extracts the identification information of the issuer from the token. For example, the "iss" field in the header or payload may be found to identify the issuer. The verifier then uses this identification information to send a request to the PKSP, asking for the public key of this issuer. 3. PKSP Processes the Request and Returns the Public Key. When the PKSP receives the verifier's request, it first authenticates the verifier's identity. After successful authentication, based on the issuer identification provided in the request, it searches for the corresponding public key in its storage system. Then the PKSP returns this public key to the verifier as a response message. The response message may also contain some metadata, such as the expiration date of the public key and the type of the public key. 4. Token Verification. The Verifier uses the public key obtained from the PKSP to verify the token according to the signature algorithm specified in the token. 5. Processing of Verification Results If the verification passes, the verifier can perform subsequent business logic processing based on the information in the token payload, such as allowing the holder to access specific resources. If the verification fails, the verifier usually rejects the operations related to this token and may record relevant log information for subsequent security audits and problem-troubleshooting.

As for the implementation technology of PKSP, distributed ledger technology strongly enables PKSP to meet the following requirements. Decentralized storage spreads data across multiple nodes to prevent single-point failures. The tamper-proof data structure, formed by complex encryption, makes data modification extremely hard. Besides, the consistency maintenance based on multi - party consensus, such as the Practical Byzantine Fault Tolerance (PBFT) algorithm, ensures that multiple nodes conduct verification for public key operations in the PKSP. Only when most nodes approve are operations recorded, reducing the impact of malicious or wrong actions by individual nodes.

TODO

TODO

IANA Considerations This document has no IANA considerations.

TODO

This document describes a mechanism for sender-constraining OAuth 2.0 tokens via a proof-of-possession mechanism on the application level. This mechanism allows for the detection of replay attacks with access and refresh tokens. This document specifies a new parameter called iss. This parameter is used to explicitly include the issuer identifier of the authorization server in the authorization response of an OAuth authorization flow. The iss parameter serves as an effective countermeasure to "mix-up attacks". In many standards track documents several words are used to signify the requirements in the specification. These words are often capitalized. This document defines these words as they should be interpreted in IETF documents. This document specifies an Internet Best Current Practices for the Internet Community, and requests discussion and suggestions for improvements.

Acknowledgments TODO