Internet of Things access control system and method based on smart contract and trust evaluation

Abstract

The invention discloses an Internet of Things access control system and method based on an intelligent contract and trust evaluation. The system comprises IoT equipment, a resource data memory RDS, an attribute authority AA and a block chain network, the method comprises the steps of initiating a resource access request, searching and responding attributes, evaluating and calculating trust values, deciding access authorization, sending access authorization and tokens, verifying access tokens, judging strategy violation, and sending and accessing resource data. According to the method, trust evaluation is introduced as the basis of authorization decision, and the comprehensive trust value is dynamically evaluated to realize fine-grained and self-adaptive access authorization; and meanwhile, safe and credible access control of the Internet of Things is realized by utilizing the characteristics of traceability and non-tampering of the block chain and adopting an intelligent contract mode. In addition, the sensitive attribute information is separated from other access control data in a main chain and side chain mode, and privacy protection of the sensitive attribute information in the access control process of the Internet of Things is achieved.

Description

Technical Field

[0001] The present invention belongs to the technical field of information security access control, and in particular relates to an Internet of Things access control system and method based on smart contracts and trust evaluation. Background Art

[0002] With the rapid development of smart devices and high-speed networks, the Internet of Things (IoT) has become popular as the main standard for resource-constrained low-power networks and has been widely used in many fields such as smart cities, Internet of Vehicles, smart medical care, smart homes, and smart industries. At the same time, IoT devices often generate massive amounts of data, including a large amount of sensitive data related to personal privacy. Once these private information is leaked, it will cause huge losses to users and even threaten the safety of users' lives. As one of the cornerstone technologies for data protection, access control technology manages user permissions to ensure that data can only be accessed by users with corresponding permissions, preventing unauthorized illegal access. In the era of the Internet of Things, access control technology will continue to be an important means to protect the security and controllable sharing of the Internet of Things.

[0003] Existing IoT access control methods based on traditional RBAC, ABAC, and UCON all use a centralized authorization decision-making entity to make access control decisions during implementation. This centralized approach relies on the credibility of the authorization decision-making entity itself, and often results in single point failure problems. Therefore, some scholars have proposed distributed access control methods such as capability-based access control CapBAC to solve the single point failure problem brought about by centralized authorization entities in the IoT. However, when CapBAC implements distributed access control on lightweight devices in the IoT, due to the weak computing and storage capabilities of IoT devices, it cannot guarantee its own security and is easily attacked by malicious users, threatening the security of the IoT.

[0004] Blockchain is a decentralized distributed technology. The data stored in the blockchain is traceable and cannot be tampered with, providing a trusted computing environment for access control in the Internet of Things. At present, some scholars have combined blockchain technology with access control and applied it to the Internet of Things environment. However, most of the existing methods allocate access rights statically, and cannot adaptively adjust access rights based on the dynamic nature of the Internet of Things, making the problems caused by static authorization more serious in the Internet of Things.

[0005] ABAC usually uses entity attributes to uniformly describe subjects, objects, etc. Through the entity attribute discovery mechanism and the attribute-authority association discovery mechanism, the entity attribute set and the {attribute, authority} association relationship can be quickly mined without manual allocation in advance. At the same time, the attribute is used as an authorization decision factor, and the current attribute value is used to decide whether to grant access rights; therefore, the dynamic problem of the Internet of Things can be solved, and for a large number of Internet of Things devices and users, ABAC management is relatively simple and the system overhead is small.

[0006] However, in the existing blockchain-based ABAC access control method, it only grants the subject access rights through the current attribute values ​​of the subject and object, without establishing a trust relationship between the subject and the object. It lacks a fine-grained access control mechanism that comprehensively considers attributes and trust, and cannot dynamically evaluate the subject's trust value based on the subject's historical access behavior and adaptively adjust the access rights, thus causing security threats such as subject node impersonation and fraud. In addition, in terms of attribute storage and verification during the access control process, the privacy protection of user sensitive information contained in the attributes is less considered, and sensitive attribute information cannot be protected from being leaked during the access control execution process. Summary of the invention

[0007] Purpose of the invention: The purpose of the present invention is to provide an Internet of Things access control system and method based on smart contracts and trust assessment. Trust assessment is introduced on the basis of the ABAC model as the basis for authorization decisions. According to the historical interaction behaviors and trust paths between IoT devices, the direct interaction trust value and the indirect recommendation trust value are calculated respectively, and the comprehensive trust value is dynamically evaluated to achieve fine-grained and adaptive access authorization; at the same time, the traceability and non-tamperability of the blockchain are utilized, and smart contracts are used to achieve secure and reliable access control of the Internet of Things. In addition, sensitive attribute information is separated from other access control data through the main and side chains to protect the privacy of sensitive attribute information during the execution of access control.

[0008] Technical solution: The present invention provides an IoT access control system based on smart contracts and trust evaluation, including: IoT devices, resource data storage RDS, attribute authority AA and blockchain network;

[0009] The IoT device includes a resource owner RO and a resource user RC; RO owns a resource r and is an object node of access control, and RO creates, updates and revokes access control policies for resource r; RC accesses resource data owned by RO and is a subject node of access control;

[0010] The resource data storage RDS is used to store resource information of the RO node; due to the limited storage capacity of the IoT device, only some temporary resource data is stored, and non-temporary resource data is stored in the resource data storage RDS.

[0011] The attribute authority AA verifies the attributes of the IoT device according to its specification and stores the attribute information in the blockchain network;

[0012] The blockchain network adopts a main-side chain layered blockchain network to separate private sensitive attribute information and public access control information; wherein the main chain MC is a public chain that stores access control information other than attribute information; the side chain SC is a private chain that stores attribute information of IoT devices; and attribute registration, trust assessment and access control verification are performed through smart contracts deployed on the blockchain network.

[0013] Furthermore, the resource data storage RDS first performs resource discovery of IoT devices and converts RO i It maps the resource r provided by it, and then stores the mapping result (r, i) and resource data; in the access control process, after obtaining the corresponding access token, RC sends a request to the resource data storage RDS to access the resource data.

[0014] Furthermore, the smart contract includes a trust evaluation contract CTR_TE and an access control contract CTR_AC deployed on the MC chain, and an attribute registration contract CTR_AR deployed on the SC chain;

[0015] The trust evaluation contract CTR_TE is configured to evaluate and calculate the comprehensive trust value of RC based on its historical access behavior, dynamically adjust and update the trust value, and feed the trust value back to the access control contract CTR_AC;

[0016] The access control contract CTR_AC is configured to store and manage access control policies and make authorization decisions based on access requests, attribute information and trust evaluation values;

[0017] The attribute registration contract CTR_AR is configured to store the attribute information of the IoT device in the SC chain after the AA verifies the attribute registration request, and to feed back the attribute information to the access control contract CTR_AC when making an authorization decision.

[0018] Furthermore, the system construction includes: building a main chain MC for storing public access control policy information and a main-side chain layered blockchain network of a side chain SC for storing private sensitive attribute information; deploying the trust evaluation contract CTR_TE and the access control contract CTR_AC on the main chain MC, deploying the attribute registration contract CTR_AR in the side chain SC, and monitoring the attribute search events of the contract CTR_AC on the MC; when the contracts CTR_TE, CTR_AC and CTR_AR are successfully deployed on the blockchain, they return three different addresses respectively, and the contracts CTR_AC and CTR_TE that need to interact, as well as CTR_AC and CTR_AR, obtain the addresses of the other contracts in advance.

[0019] Furthermore, the attribute registration includes: when the RC node and the RO node in the Internet of Things are connected to the Internet of Things, they need to send an attribute registration request to register their attributes to the AA. The AA verifies each attribute information contained in its attribute set, such as sensor type, according to the specifications of the RC node and the RO node. After the verification is completed, the AA sends an attribute registration transaction to the CTR_AR contract. or Store the attribute information in the side chain SC;

[0020] The property is expressed as α=<attr_name,attr_relation,attr_value> , where attr_name and attr_value represent the attribute name and attribute value respectively. Indicates the relationship between attribute name and attribute value, such as belongs to, does not belong to, is between, is equal to, is not equal to, is greater than, etc.; the attributes of RO nodes and RC nodes are information related to IoT devices, verified by AA and stored on the SC chain; environmental attributes are information related to the context of the access request; operation attributes are the types of operations on access resources;

[0021] The attribute set is denoted as A x ={α1,α2,…,α N},x∈{RO,RC,e,op}, where α1,α2,…,α N Refers to the N attributes, A RO , A RC , A e , A op Respectively represent the attribute set of RO node, RC node, context environment e or operation op;

[0022] The attribute registration transaction or transaction Contains: attribute set A x, x∈{RO,RC}, RC's signature sig(RC) or RO's signature sig(RO), AA's signature sig(AA), and registration time reg_time; transaction and They are defined as:

[0023]

[0024] Furthermore, the storage method of the access control policy includes: the RO node defines the access control policy P of the resource r it owns r , and send the policy storage transaction TX to the CTR_AC contract pol Store the access control policy in the main chain MC;

[0025] The access control policy P r It consists of one or more rules. The definition of rule is:

[0026]

[0027] in, They represent the attributes of the RC node, RO node, context environment e and operation op required to access resource r, CT min represents the minimum comprehensive trust value required to access resource r, Result represents the authorization result of the rule, Result∈{Permit, Deny, not applicable};

[0028] The policy storage transaction TX pol Contains: Access control policy P for resource r r 、Criteria creation time crt_time, RO signature sig(RO), transaction TX pol is defined as:

[0029] TX pol =sendTransaction(P r , crt_time, sig(RO))

[0030] The present invention also provides an access control method for the Internet of Things access control system based on smart contracts and trust evaluation, comprising the following steps:

[0031] Step 1: RC j Send resource access request transaction TX to the CTR_AC contract arq , the transaction includes: the requested resource r, the operation op on the resource, and RC j The signature sig(RC j ), where transaction TX arqis defined as:

[0032] TX arq =sendTransation(r,op,sig(RC j ))

[0033] Step 2: The CTR_AC contract first sends a resource location request to the RDS to determine the RO to which the resource r belongs. i Then, the CTR_AC contract sends an attribute search event to the side chain SC. After SC listens to the event from MC, it calls the CTR_AR contract to query RC in SC. j and RO i The attribute set and return attribute response transaction TX As To CTR_AC contract;

[0034] Step 3: The CTR_AC contract calculates RO by calling the CTR_TE contract i About RC j Direct interaction trust value DIT i，j and indirect recommendation trust value IRT i，j , and evaluate and calculate its comprehensive trust value CT i，j ;

[0035] Step 4: CTR_AC contract returns RO i and RC j The attribute set of the resource r, the attribute set of the operation op on the resource r, the attribute set of the context e where the access control occurs, and the evaluation result CT of the comprehensive trust value i，j , make access authorization decisions and judge RC j Whether the resource access request satisfies the access control policy P r ;

[0036] Step 5: If the access control policy is met, the CTR_AC contract sends the access token i，j (op, r) to RC j Authorized RC j Visit RO i Resource r, token i，j (op, r) contains: the creation time crt_time of the token, the expiration time exp_time and the minimum interval between two consecutive accesses interval; where the access token token i，j (op, r) is defined as:

[0037] token i，j (op, r) = (crt_time, exp_time, interval)

[0038] Step 6: RC j Send the resource access message Access(r) to RDS for verification. The message Access(r) carries RC. j Signed token i，j (op, r);

[0039] Step 7: RDS uses Access(r) to determine whether an access control policy violation has occurred:

[0040] Step 8: If there is no access control policy violation, RDS sends a resource access result message to RC. j In response to the access request, the resource access result message includes: i Resource data signed by the public key and resource access time;

[0041] Step 9: RC j After receiving the response, the resource data is decrypted using its private key to access the resource.

[0042] The direct interaction trust value DIT described in step 3 i，j It is the trust evaluation of node ROi based on its historical interaction behavior with node RCj; the indirect recommendation trust value IRT i，j It is a node RO i Through other nodes RO k About RC j The comprehensive trust value CT is the trust evaluation obtained by direct interactive trust; i，j Based on direct interaction trust DIT i，j and indirect recommendation trust IRT i，j The final trust rating obtained; DIT i,j , IRT i，j and CT i，j The value range is [0, 1]; the larger the value, the more RO i About RC j The more trust, 1 means RO i About RC j Full trust, 0 means RO i About RC j Total distrust.

[0043] The comprehensive trust value CT described in step 3 i，j The evaluation calculation includes the following steps:

[0044] Step 3.1: According to the node RO i With NodeRC j Historical interaction behaviors between them, calculate the direct interaction trust value DIT i，j ;

[0045] Step 3.2: According to the node RO i With NodeRC j The trust path formed by direct interactive trust between them is used to calculate the indirect recommendation trust value IRT i，j ;

[0046] Step 3.3: According to DIT i，j and IRT i，j And their weight distribution, evaluate and calculate the comprehensive trust value CT i，j .

[0047] Furthermore, the direct interaction trust value DIT described in step 3.1 i，j The calculation of includes the following steps:

[0048] Step 3.1.1. Calculate the time decay factor β(k)

[0049] The time decay factor β(k) represents the impact of interactions occurring at different times on the direct interaction trust value, and its calculation formula is as follows:

[0050]

[0051] Among them, t k is the time interval between the kth interaction and the current moment, k∈{1,...,L}; r is the decay rate, which increases with t k As , the rate at which β(k) decreases is greater than 1;

[0052] Step 3.1.2. Calculate the direct interaction weight α(k)

[0053] The direct interaction weight α(k) represents the proportion of the kth interaction in the total L interactions in the direct interaction trust evaluation. The calculation formula is as follows:

[0054]

[0055] Step 3.1.3. Calculate the positive interaction coefficient δ pos

[0056] Positive interaction coefficient δ pos For nodes ROi and RC j During the interaction process, the node RC j There is no access control policy violation, and the positive interaction coefficient is calculated as follows:

[0057]

[0058] Among them, pos(k) is 1, which means the kth interaction is a positive interaction, otherwise it is a negative interaction; μ pos is the weight of positive interaction;

[0059] Step 3.1.4. Calculate the negative interaction coefficient δ neg

[0060] Negative interaction coefficient δ neg In the interaction between node ROi and node RCj, node RC j When an access control policy violation occurs, the negative interaction coefficient is calculated as follows:

[0061]

[0062] Where neg(k) is 1, indicating that the kth interaction is a negative interaction, otherwise it is a positive interaction; μ neg is the weight of negative interaction;

[0063] Step 3.1.5: Calculate the direct interaction trust value DIT i,j

[0064] Direct Interaction Trust Value DIT i,j is the proportion of the positive interaction coefficient in the total direct interaction coefficient. After L interactions, the calculation formula of the direct interaction trust value is as follows:

[0065]

[0066] Furthermore, the indirect recommendation trust value IRT described in step 3.2 i,，j The calculation of includes the following steps:

[0067] Step 3.2.1: Calculate the indirect recommendation trust value of the k-th trust path

[0068] The indirect recommendation trust value of the k-th trust path The calculation formula is as follows:

[0069]

[0070] Among them, the kth trust path is direct interactive trust The formed slave node RO i To RC j The path, x1,x2,…,x n is the serial number of all nodes passed in the k-th trust path;

[0071] Step 3.2.2: Calculate the indirect recommendation trust value IRT i,j

[0072] Node RO i RC for nodes j Indirect recommendation trust value IRT i,j For slave node RO iTo Node RC j The mean of the indirect recommendation trust values ​​of each trust path between them is calculated as follows:

[0073]

[0074] Where m represents the slave node RO i To Node RC j The total number of trust paths.

[0075] Furthermore, the comprehensive trust value CT described in step 3.3 i,j The calculation is evaluated using the following formula:

[0076] CT i,j =γ×DIT i,j +ε×IRT i,，j (8)

[0077] Among them, γ and ε are the weights of direct interaction trust and indirect recommendation trust respectively, and γ, ε∈[0,1] and γ+ε=1.

[0078] The authorization decision process described in step 4 is as follows: i,j Reach access control policy P r When the required trust value threshold is reached, the RO i Node, RC j If the attribute values ​​of the node, operation op, and context environment e meet the attribute requirements of the access control policy, it is RC j Send access token i,j (op,r).

[0079] The policy violation determination in step 7 specifically includes the following steps:

[0080] Step 7.1, RDS Verification RC j Whether to use expired or forged tokens i,j (op,r) access data;

[0081] Step 7.2, RDS determines RC j Whether to send consecutive requests with a minimum interval less than the access token interval;

[0082] Step 7.3: If the access control policy violation of step 7.1 or step 7.2 occurs, the access is recorded as a negative interaction, otherwise it is a positive interaction;

[0083] Step 7.4: RDS notifies the CTR_TE contract to evaluate and update RO based on this interaction. i About RC j The comprehensive trust value.

[0084] Beneficial effects: Compared with the prior art, the present invention has the following significant advantages:

[0085] 1. The system and method described in the present invention make full use of the decentralization, immutability and traceability of blockchain, and realize reliable and secure access control for the Internet of Things environment by setting attribute registration contracts, access control contracts and trust assessment contracts.

[0086] 2. The system and method described in the present invention introduce a trust evaluation mechanism in the access control execution process, propose a trust evaluation method based on direct interactive trust and indirect recommended trust, and dynamically evaluate and update the trust value according to the violation of the access control policy, thereby realizing dynamic and fine-grained access control for the Internet of Things environment.

[0087] 3. The system and method described in the present invention adopt a main-chain and side-chain layered blockchain network, in which the main chain stores access control information other than attribute information; the side chain only stores the user's attribute information, separates private sensitive attribute information from public access control information, and realizes the privacy protection of attribute information in the access control process of the Internet of Things. BRIEF DESCRIPTION OF THE DRAWINGS

[0088] Figure 1 This is an architecture diagram of an Internet of Things access control system based on smart contracts and trust assessment provided in Example 1 of the present invention.

[0089] Figure 2 It is a flow chart of the Internet of Things access control method based on smart contracts and trust assessment provided in Example 2 of the present invention.

[0090] Figure 3 It is a flow chart of the comprehensive trust value evaluation calculation method provided in Example 2 of the present invention.

[0091] Figure 4 It is a trust path diagram in the indirect recommendation trust calculation provided by Example 2 of the present invention. DETAILED DESCRIPTION

[0092] The technical solution of the present invention is further described below in conjunction with the accompanying drawings.

[0093] Embodiment 1:

[0094] Embodiment 1 of the present invention provides an Internet of Things access control system based on smart contracts and trust assessment, such as Figure 1 As shown, the system includes: resource ownership node RO, resource use node RC, resource data storage RDS, attribute authority AA, and blockchain network. Among them, RO nodes and RC nodes are both IoT devices of the Internet of Things.

[0095] RO node: RO owns resource r and is the object node of access control. RO creates, updates and revokes access control policies for resource r. Each RO node has its own attribute value.

[0096] RC node: RC accesses the resource data owned by RO and is the main node of access control. Each RC node has its own attribute value and trust value, which serves as the basis for authorization decision of access control.

[0097] In the Internet of Things, each IoT device can be either RO or RC, which means it can provide resources to other IoT devices while also accessing resources provided by other IoT devices. Since IoT devices (RO nodes and RC nodes) such as sensors and cameras usually have weak performance, they need to be connected to the blockchain network via an IoT gateway and identified by a public key.

[0098] Resource Data Storage RDS: Responsible for storing resource-related information of RO nodes. Due to the limited storage capacity of IoT devices, only some temporary resource data is stored, and non-temporary resource data is stored in the resource data storage RDS. First, RDS performs resource discovery of IoT devices and stores RO i Mapping is performed with the resource r provided by it, and then the mapping result (r, i) and resource data are stored. In the access control process, after obtaining the corresponding access token, RC sends a request to the resource data storage RDS to access the resource data.

[0099] Attribute Authority AA: Responsible for verifying the attributes of IoT devices, such as sensor type, according to their specifications and storing the attribute information in the blockchain to ensure the authenticity of the attribute information.

[0100] Blockchain network: a layered blockchain network with main and side chains that is responsible for separating sensitive attribute information and public access control information. The main chain MC is a public chain that stores access control policies and other access control related information other than attribute information; the side chain SC is a private chain that only stores the attribute information of IoT devices and can only be accessed by AA.

[0101] This embodiment uses the Ethereum blockchain network, and the Ethereum virtual machine (EVM) provides an operating environment for smart contracts. Smart contracts are deployed and called in the EVM, and attribute registration, trust assessment, and access control verification are automatically performed. The platform provides two types of accounts: external control accounts EOA and contract accounts CA. Among them, EOA is protected by a private key, and the contract account is protected by the contract code. Both the main chain MC and the side chain SC use the elliptic cryptographic algorithm (ECC) as a public key cryptographic mechanism to generate public and private key pairs, as well as digital signatures and verifications.

[0102] The multi-smart contract architecture designed in this embodiment includes the trust evaluation contract CTR_TE and the access control contract CTR_AC deployed on the MC, and the attribute registration contract CTR_AR deployed on the SC. The attribute registration, trust evaluation and access control verification are automatically performed through the above three smart contracts. The three smart contracts define their own functions to provide a calling interface for the various access control functions of the system described in this embodiment.

[0103] Trust evaluation contract CTR_TE: responsible for calculating the comprehensive trust value of RC based on its historical access behavior, dynamically adjusting and updating the trust value according to the access control policy violation when RC accesses resources, and feeding back the trust value to the access control contract CTR_AC. Contract CTR_TE defines the function getDITrustValue() for calculating the direct interaction trust value, the function getIRTrustValue() for calculating the indirect recommendation trust value, and the function getTrustValue() for evaluating and calculating the comprehensive trust value.

[0104] Access control contract CTR_AC: responsible for storing and managing access control policies, and making authorization decisions based on access requests, attribute information, and trust evaluation values. Contract CTR_AC defines the function authDecision() for access authorization decisions and the function adminPolicy() for managing access control policies.

[0105] Attribute registration contract CTR_AR: After AA verifies the attribute registration request, it is responsible for storing the attribute information of the IoT device in the SC chain and feeding the attribute information back to the access control contract CTR_AC when making authorization decisions. The CTR_AR contract defines functions such as searchROAttrSet() and searchRCAttrSet() for querying RO and RC attribute sets in the SC chain.

[0106] According to the system described in this embodiment, its initialization is mainly divided into three stages: system construction, attribute registration, and policy storage.

[0107] Phase 1: System Construction

[0108] Build a main chain MC for storing public access control policy and other information, and a main and side chain layered Ethereum blockchain network for storing private sensitive attribute information. Deploy the trust evaluation contract CTR_TE and the access control contract CTR_AC on the main chain MC, deploy the attribute registration contract CTR_AR on the side chain SC, and listen to the attribute search event from the contract CTR_AC on MC. When the contracts CTR_TE, CTR_AC and CTR_AR are successfully deployed on the Ethereum blockchain, they return three different addresses respectively. The contracts CTR_AC and CTR_TE that need to interact, as well as CTR_AC and CTR_AR, need to obtain the address of the other contract in advance.

[0109] Phase 2: Attribute Registration

[0110] When the RC nodes and RO nodes in the IoT are connected to the IoT, they need to send an attribute registration request to register their attributes to the AA. The AA verifies the attribute information such as sensor type according to the specifications of the RC nodes and RO nodes. After the verification is completed, the AA sends an attribute registration transaction to the CTR_AR contract. (or )Store the attributes in the side chain SC.

[0111] The attribute is represented by α=<attr_name,attr_relation,attr_value> , where attr_name and attr_value represent the attribute name and attribute value respectively. Indicates the relationship between the attribute name and the attribute value, such as belongs to, does not belong to, is between, is equal to, is not equal to, is greater than Indicates that the value of the attribute access_time is between 8:00 and 10:00, and sensor_type∈{pressure,speed,position} indicates that the value of the attribute sensor_type is a value among pressure,speed,position. The attributes of RO nodes and RC nodes are information related to IoT devices, such as sensor type, etc., which are verified by AA and stored on the SC chain; the operation attribute is the type of operation on accessing resources, such as reading, writing, creating, deleting, etc. of resource data. The environment attribute is information related to the context of the access request, such as access time, etc.

[0112] The attribute set is represented by A x ={α1,α2,…,α N},x∈{RO,RC,e,op}, where α1,α2,…,α N Refers to the N attributes, A RO , A RC , A e , Aop They respectively represent the attribute sets of an RO node, an RC node, a context environment e, or an operation op on a resource.

[0113] Property Registration Transactions (or business ) contains: attribute set A x ,x∈{RO,RC}, RC's signature sig(RC) (or RO's signature sig(RO)), AA's signature sig(AA), and registration time reg_time. Transaction and They are defined as:

[0114]

[0115] Phase 3: Policy Storage

[0116] The RO node defines the access control policy P for the resource r it owns. r , and send the policy storage transaction TX to the CTR_AC contract pol The access control policy is stored in the main chain MC.

[0117] Access Control Policy P r It consists of one or more rules. The definition of rule is:

[0118]

[0119] in, They represent the attributes of the RC node, RO node, context environment e and operation op required to access resource r, CT min It represents the minimum comprehensive trust value required to access resource r. Result represents the judgment result of the rule, Result∈{Permit, Deny, not applicable}.

[0120] Policy storage transaction TX pol Contains: Access control policy P for resource r r , the creation time of the policy crt_time, the signature of RO sig(RO). Transaction TX pol is defined as:

[0121] TX pol =sendTransaction(P r , crt_time, sig(RO))

[0122] Embodiment 2:

[0123] Embodiment 2 of the present invention provides an IoT access control method based on smart contracts and trust evaluation, and the method flow is as follows: Figure 2 The specific steps include:

[0124] Step 1: Initiate resource access request

[0125] RCj sends a resource access request transaction TX to the CTR_AC contract arq , the transaction includes: the requested resource r, the operation op on the resource, and RC j The signature sig(RC j ). Among them, transaction TX arq is defined as:

[0126] TX arq =sendTransation(r,op,sig(RC j ))

[0127] Step 2: Attribute search and response

[0128] The CTR_AC contract first sends a resource location request to the RDS to determine the RO to which the resource r belongs. i Then, the CTR_AC contract sends an attribute search event to the side chain SC. After SC listens to the event from MC, it calls the CTR_AR contract to query RC in SC. j and RO i The attribute set and return attribute response transaction TX AS To CTR_AC contract.

[0129] Step 3: Trust value evaluation calculation

[0130] The CTR_AC contract first calculates RO by calling the CTR_TE contract i About RC j Direct interaction trust value DIT i，j and indirect recommendation trust value IRT i，j , and then evaluate and calculate its comprehensive trust value CT i，j The process of comprehensive trust value evaluation calculation is as follows: Figure 3 As shown, the following steps are included:

[0131] Step 3.1: According to the node RO i With NodeRC j The most recent L historical interactions between them are used to calculate the direct interaction trust value DIT i，j ;

[0132] Step 3.2: According to the node RO i With NodeRC jThe trust path formed by direct interactive trust between them is used to calculate the indirect recommendation trust value IRT i，j ;

[0133] Step 3.3: According to DIT i，j and IRT i，j And their weight distribution, evaluate and calculate the comprehensive trust value CT i，j .

[0134] ROi’s direct interactive trust DIT to RCj i，j The node ROi is based on its own relationship with the node RC j The trust evaluation is obtained by evaluating the most recent L interactions. The value of L is adjusted by the contract CTR_TE according to the specific IoT application scenario. The larger the value of L, the more accurate the calculation of the direct interaction trust value, but at the same time the greater the calculation cost.

[0135] Direct Interaction Trust Value DIT i，j The calculation specifically includes steps 3.1.1 to 3.1.5.

[0136] Step 3.1.1. Calculate the time decay factor β(k)

[0137] The time decay factor β(k) represents the impact of interactions occurring at different times on the trust value of direct interactions. Its calculation formula is as follows:

[0138]

[0139] Among them, t k is the time interval between the kth interaction and the current moment, k∈{1,…,L}; r is the decay rate, which increases with t k As increases, the rate at which β(k) decreases is a number greater than 1.

[0140] Step 3.1.2. Calculate the direct interaction weight α(k)

[0141] The direct interaction weight α(k) represents the proportion of the kth interaction in the total L interactions in the direct interaction trust evaluation. The calculation formula is as follows:

[0142]

[0143] Step 3.1.3. Calculate the positive interaction coefficient δpos

[0144] Positive interaction means that the node RO i With NodeRC j During the interaction process, the node RC j There is no access control policy violation, and the positive interaction coefficient is calculated as follows:

[0145]

[0146] Among them, pos(k) is 1, which means the kth interaction is a positive interaction, otherwise it is a negative interaction; μ pos is the weight of positive interaction;

[0147] Step 3.1.4. Calculate the negative interaction coefficient δ neg

[0148] Negative interaction means that the node RO i With node RCj 的 During the interaction, node RC j When an access control policy violation occurs, the negative interaction coefficient is calculated as follows:

[0149]

[0150] Where neg(k) is 1, indicating that the kth interaction is a negative interaction, otherwise it is a positive interaction; μ neg is the weight of negative interaction. In social trust relationships, it is usually more difficult to establish trust than to lose trust. Considering the malicious attacks that may occur when a node obtains a high trust value, μ neg The value should be greater than μ pos .

[0151] Step 3.1.5: Calculate the direct interaction trust value DIT i，j

[0152] Direct Interaction Trust Value DIT i，j is the proportion of the positive interaction coefficient in the total direct interaction coefficient. After L interactions, the calculation formula of the direct interaction trust value is as follows:

[0153]

[0154] RO i About RC j Indirect Recommendation Trust IRT i,j It is a node RO i By integrating other RO k Node to Node RC j Before evaluating the indirect recommendation trust, it is necessary to obtain the node RO i and node RC j A set of trust paths formed by direct interactive trust between parties.

[0155] The direct interaction between all nodes RO and RC in the Internet of Things forms a directed graph. Node RO and node RC are abstracted as nodes of the directed graph. i and node RC jIf there is direct interaction, there is a slave node RO i To Node RC j The directed edges of DIT i,j is the weight of the edge between these two nodes. The path search algorithm of the graph is used to obtain the weight of the edge from node RO. i To Node RC j All trust paths of Figure 4 The figure shows the trust path formed by the direct interactive trust from node s to node t. Each directed trust edge on the path, such as a→b, indicates that RO node b has a direct interactive trust evaluation on RC node a. The indirect recommended trust of node t to node s is obtained by calculating the indirect recommended trust of the four trust paths s→a→b→t, s→c→d→t, s→e→f→g→t, and s→h→t.

[0156] Indirect Interaction Trust IRT i,j The calculation specifically includes: step 3.2.1 and step 3.2.2.

[0157] Step 3.2.1: Calculate the indirect recommendation trust value of the k-th trust path

[0158] The indirect recommendation trust value of the k-th trust path The calculation formula is as follows:

[0159]

[0160] Among them, the kth trust path is direct interactive trust The formed slave node RO i To RC j The path, x1,x2,…,x n is the sequence number of all nodes passed through in the k-th trust path.

[0161] Step 3.2.2: Calculate the indirect recommendation trust value IRT i,j

[0162] Node RO i RC for nodes j Indirect recommendation trust value IRT i,，j For slave node RO i To Node RC j The mean of the indirect recommendation trust values ​​of each trust path between them is calculated as follows:

[0163]

[0164] Where m represents the slave node RO i To Node RC j The total number of trust paths;

[0165] RO i About RC j The comprehensive trust value of the computing node RO is calculated by evaluating i RC for nodes j The final trust evaluation is obtained by combining the direct interaction trust and indirect recommendation trust. When there is no direct interaction between nodes, the comprehensive trust value is the indirect recommendation trust value.

[0166] Comprehensive Trust Value CT i,，j The evaluation formula is as follows:

[0167] CT i,j =γ×DIT i,j +ε×IRT i,j (8)

[0168] Among them, γ and ε are the weights of direct interaction trust and indirect recommendation trust respectively, and γ, ε∈[0,1] and γ+ε=1.

[0169] Step 4: Access authorization decision

[0170] CTR_AC contract based on the returned RO i and RC j The attribute set of the resource r, the attribute set of the operation op on the resource r, the attribute set of the context e, and the evaluation result CT of the comprehensive trust value i,j , using the defined access authorization decision function authDecision() to judge RC j Whether the resource access request satisfies the access control policy P r The process of access authorization decision is shown in Algorithm 1.

[0171] Algorithm 1: Access authorization decision algorithm

[0172] Input:P r , TX arq .

[0173] Output: true or false.

[0174]

[0175]

[0176] The specific interpretation of Algorithm 1 is as follows:

[0177] First, lines 1 and 2 call getEnvAttrSet() to obtain the context attribute set A. e and calling getOpAttrSet(TX arq ) from access request TX arqGet the operation attribute set A op , lines 3-4 call getRC(TX arq ) and getMapRes((getRes(TX arq ) to obtain the ID information of RO and RC. Lines 5-6 call searchROAttrSet(RO_ID) and searchRCAttrSet(RC_ID) of the CTR_AR contract to query the attribute set A of RO and RC in the SC chain. RO and A RC Line 7 calls the getTrustValue() of the CTR_TE contract to evaluate and calculate the comprehensive trust value CT i，j .

[0178] Lines 8-24 are the authorization decision process of the CTR_AC contract. First, in the access control policy P r Find the rules in Is the attribute name set contained in A? x The attribute name set, where x∈{RO, RC, e, op}. If it is included, it means that the access control rule rule[i] corresponding to the access request has been found. Then, judge CT i，j Whether the comprehensive trust value threshold requirement in the rule is met. If so, then judge A x Whether the attribute name and attribute value of the request satisfy the relationship between the corresponding attribute name and attribute value in the rule. If so, the access request satisfies the rule, that is, A x Neutral Rules The attribute with the same attribute name in the rule has an attribute value that meets the constraints of the rule, and the authorization decision result is given according to the result of the rule. If the result of the rule is Permit, the access token token is sent i，j (op, r), and returns the final authorization decision result true. If the result of the rule is Deny or not applicable, or the access request does not meet the rule constraints, then Algorithm 1 takes false as the decision result.

[0179] Step 5: Access authorization and token sending

[0180] If the access control policy is met, the CTR_AC contract sends an access token i，j (op, r) to RC j Authorized RC j Visit RO i Resource r, token i，j(op, r) contains: the creation time crt_time of the token, the expiration time exp_time and the minimum interval between two consecutive accesses interval. i，j (op, r) is defined as:

[0181] token i，j (op, r) = (crt_time, exp_time, interval).

[0182] In the access token i，j Within the validity period interval of (op, r), RC j The token can be used multiple times to access resource r while meeting the minimum interval time, reducing the impact on access control performance.

[0183] Step 6: Verify the access token

[0184] RC j Send the resource access message Access(r) to RDS for verification. The message Access(r) carries RC. j Signed token i，j (op, r).

[0185] Step 7: Determine whether the access control policy is violated

[0186] RDS uses Access(r) to determine whether an access control policy violation has occurred: 1) Verify RC j Whether to use expired or forged tokens i，j (op, r) access data; 2) determine RC j Whether to send continuous requests at a minimum interval interval lower than the access token to prevent DoS attacks. If there is a violation of the access control policy 1) or 2), this access will be recorded as a negative interaction, otherwise it will be a positive interaction. Then notify the CTR_TE contract to evaluate and update the RO based on this interaction. i About RC j The comprehensive trust value.

[0187] Step 8: Sending resource data

[0188] If there is no access control policy violation, RDS sends a resource access result message to RC. j In response to the access request, the resource access result message includes: i Resource data signed by the public key and the time when the resource was accessed.

[0189] Step 9: Access resource data

[0190] RC j After receiving the response, the resource data is decrypted using its private key to access the resource.

[0191] Any details not mentioned in the specific embodiments of the present invention belong to the known technologies in the art and can be implemented by referring to the known technologies.

[0192] The embodiments involved in the above specific implementation methods are specific support for an Internet of Things access control system and method based on smart contracts and trust assessment proposed in the present invention, but the protection scope of the present invention is not limited to this. All modifications, equivalent substitutions or improvements made on the basis of this technical solution in accordance with the technical ideas proposed in the present invention belong to the protection scope of the present invention.

Claims

1. An Internet of Things access control system based on smart contracts and trust assessment, characterized in that: include: IoT devices, resource data storage RDS, attribute authority AA and blockchain network; The IoT device includes a resource owner RO and a resource user RC; RO owns resource r and is the object node of access control. RO creates, updates and revokes access control policies for resource r. RC accesses resource data owned by RO and is the subject node of access control. The resource data storage RDS is used to store resource information of the RO node; The attribute authority AA verifies the attributes of the IoT device according to its specification and stores the attribute information in the blockchain network; The blockchain network adopts a main-side chain layered blockchain network to separate private sensitive attribute information and public access control information; wherein the main chain MC is a public chain that stores access control information other than attribute information; the side chain SC is a private chain that stores attribute information of IoT devices; and attribute registration, trust assessment and access control verification are performed through smart contracts deployed on the blockchain network.

2. According to claim 1, an Internet of Things access control system based on smart contracts and trust assessment is characterized in that: The resource data storage RDS first performs resource discovery of IoT devices and converts the RO i It maps the resource r provided by it, and then stores the mapping result (r, i) and resource data; in the access control process, after obtaining the corresponding access token, RC sends a request to the resource data storage RDS to access the resource data.

3. The IoT access control system based on smart contracts and trust assessment according to claim 1, characterized in that: The smart contracts include the trust evaluation contract CTR_TE and the access control contract CTR_AC deployed on the MC chain, and the attribute registration contract CTR_AR deployed on the SC chain; The trust evaluation contract CTR_TE is configured to evaluate and calculate the comprehensive trust value of RC based on its historical access behavior, dynamically adjust and update the trust value, and feed the trust value back to the access control contract CTR_AC; The access control contract CTR_AC is configured to store and manage access control policies and make authorization decisions based on access requests, attribute information and trust evaluation values; The attribute registration contract CTR_AR is configured to store the attribute information of the IoT device in the SC chain after the AA verifies the attribute registration request, and to feed back the attribute information to the access control contract CTR_AC when making an authorization decision.

4. The IoT access control system based on smart contracts and trust assessment according to claim 1, characterized in that: The system construction includes: building a main chain MC for storing public access control policy information and a main-side chain layered blockchain network of the side chain SC for storing private sensitive attribute information; deploying the trust evaluation contract CTR_TE and the access control contract CTR_AC on the main chain MC, deploying the attribute registration contract CTR_AR in the side chain SC, and monitoring the attribute search events of the contract CTR_AC on MC; when the contracts CTR_TE, CTR_AC and CTR_AR are successfully deployed on the blockchain, they return three different addresses respectively, and the contracts CTR_AC and CTR_TE that need to interact, as well as CTR_AC and CTR_AR, obtain the addresses of each other's contracts in advance.

5. The IoT access control system based on smart contracts and trust assessment according to claim 1, characterized in that: Attribute registration includes: RC nodes and RO nodes in the Internet of Things need to send attribute registration requests to register their attributes to AA when they are connected to the Internet of Things. AA verifies each attribute information contained in its attribute set according to the specifications of the RC nodes and RO nodes. After the verification is completed, AA sends an attribute registration transaction to the CTR_AR contract. or Store the attribute information in the side chain SC; The property is expressed as α=<attr_name,attr_relation,attr_value> , where attr_name and attr_value represent the attribute name and attribute value respectively. Indicates the relationship between attribute name and attribute value, such as belongs to, does not belong to, is between, is equal to, is not equal to, is greater than, etc.; the attributes of RO nodes and RC nodes are information related to IoT devices, verified by AA and stored on the SC chain; environmental attributes are information related to the context of the access request; operation attributes are the types of operations on access resources; The attribute set is denoted as A x ={α1,α2,…,α N },x∈{RO,RC,e,op}, where α1,α2,…,α N Refers to the N attributes, A RO , A RC , A e , A op Respectively represent the attribute set of RO node, RC node, context environment e or operation op; The attribute registration transaction or transaction Contains: attribute set A x ,x∈{RO,RC}, RC's signature sig(RC) or RO's signature sig(RO), AA's signature sig(AA), and registration time reg_time; transaction and They are defined as:

6. The IoT access control system based on smart contracts and trust assessment according to claim 1, characterized in that: The storage of access control policies includes: the RO node defines the access control policy P of the resource r it owns r , and send the policy storage transaction TX to the CTR_AC contract pol Store the access control policy in the main chain MC; The access control policy P r It consists of one or more rules. The definition of rule is: in, They represent the attributes of the RC node, RO node, context environment e and operation op required to access resource r, CT min represents the minimum comprehensive trust value required to access resource r, Result represents the authorization result of the rule, Result∈{Permit, Deny, not applicable}; The policy storage transaction TX pol Contains: Access control policy P for resource r r 、Criteria creation time crt_time, RO signature sig(RO), transaction TX pol is defined as: TX pol =sendTransaction(P r ,crt_time,sig(RO)).

7. An Internet of Things access control method based on smart contracts and trust evaluation, characterized in that: The steps include: Step 1: RC j Send resource access request transaction TX to the CTR_AC contract arq , the transaction includes: the requested resource r, the operation op on the resource, and RC j The signature sig(RC j ), where transaction TX arq is defined as: TX arq =sendTransaction(r,op,sig(RC j )) Step 2: The CTR_AC contract first sends a resource location request to the RDS to determine the RO to which the resource r belongs. i Then, the CTR_AC contract sends an attribute search event to the side chain SC. After SC listens to the event from MC, it calls the CTR_AR contract to query RC in SC. j and RO i The attribute set and return attribute response transaction TX AS To CTR_AC contract; Step 3: The CTR_AC contract calculates RO by calling the CTR_TE contract i About RC j Direct interaction trust value DIT i,j and indirect recommendation trust value IRT i,j , and evaluate and calculate its comprehensive trust value CT i,j ; Step 4: CTR_AC contract returns RO i and RC j The attribute set of the resource r, the attribute set of the operation op on the resource r, the attribute set of the context e where the access control occurs, and the evaluation result CT of the comprehensive trust value i,j , make access authorization decisions and judge RC j Whether the resource access request satisfies the access control policy P r ; Step 5: If the access control policy is met, the CTR_AC contract sends the access token i,j (op,r) to RC j Authorized RC j Visit RO i Resource r, token i,j (op,r) contains: the creation time of the token ctr_time, the expiration time exp_time and the minimum interval between two consecutive accesses interval; among them, the access token token i,j The definition of (op,r) is: token i,j (op,r)=(crt_time,exp_time,interval) Step 6: RC j Send the resource access message Access(r) to RDS for verification. The message Access(r) carries RC. j Signed token i,j (op,r); Step 7: RDS determines whether access control policy violation occurs through Access(r); Step 8: If there is no access control policy violation, RDS sends a resource access result message to RC. j In response to the access request, the resource access result message includes: i Resource data signed by the public key and resource access time; Step 9: RC j After receiving the response, the resource data is decrypted using its private key to access the resource.

8. The IoT access control method based on smart contracts and trust assessment according to claim 7 is characterized in that: In step 3, the direct interaction trust value DIT i,j It is a node RO i Based on itself and node RC j Trust evaluation based on historical interaction behaviors; The indirect recommendation trust value IRT i,j It is a node RO i Through other nodes RO k About RC j The comprehensive trust value CT is the trust evaluation obtained by direct interactive trust; i,j Based on direct interaction trust DIT i,j and indirect recommendation trust IRT i,j The final trust rating obtained; DIT i,j , IRT i,j and CT i,j The value range is [0, 1]; the larger the value, the more RO i About RC j The more trust, 1 means RO i About RC j Full trust, 0 means RO i About RC j Total distrust.

9. The IoT access control method based on smart contracts and trust assessment according to claim 7 is characterized in that: In step 3, the comprehensive trust value CT i,j The evaluation calculation includes the following steps: Step 3.1: According to the node RO i With NodeRC j Historical interaction behaviors between them, calculate the direct interaction trust value DIT i,j ; Step 3.2: According to the node RO i With NodeRC j The trust path formed by direct interactive trust between them is used to calculate the indirect recommendation trust value IRT i,j ; Step 3.3: According to DIT i,j and IRT i,j And their weight distribution, evaluate and calculate the comprehensive trust value CT i,j .

10. The IoT access control method based on smart contracts and trust assessment according to claim 9 is characterized in that: Direct interaction trust value DIT described in step 3.1 i,j The calculation of includes the following steps: Step 3.1.

1. Calculate the time decay factor β(k) The time decay factor β(k) represents the impact of interactions occurring at different times on the direct interaction trust value, and its calculation formula is as follows: Among them, t k is the time interval between the kth interaction and the current time, k∈{1,…,L}; r is the decay rate, which increases with t k As , the rate at which β(k) decreases is greater than 1; Step 3.1.

2. Calculate the direct interaction weight α(k) The direct interaction weight α(k) represents the proportion of the kth interaction in the total L interactions in the direct interaction trust evaluation. The calculation formula is as follows: Step 3.1.

3. Calculate the positive interaction coefficient δ pos Positive interaction coefficient δ pos RO i With NodeRC j During the interaction process, the node RC j There is no access control policy violation, and the positive interaction coefficient is calculated as follows: Among them, pos(k) is 1, which means the kth interaction is a positive interaction, otherwise it is a negative interaction; μ pos is the weight of positive interaction; Step 3.1.

4. Calculate the negative interaction coefficient δ neg Negative interaction coefficient δ neg RO i With NodeRC j During the interaction process, the node RC j When an access control policy violation occurs, the negative interaction coefficient is calculated as follows: Where neg(k) is 1, indicating that the kth interaction is a negative interaction, otherwise it is a positive interaction; μ neg is the weight of negative interaction; Step 3.1.5: Calculate the direct interaction trust value DIT i，j Direct Interaction Trust Value DIT i,j is the proportion of the positive interaction coefficient in the total direct interaction coefficient. After L interactions, the calculation formula of the direct interaction trust value is as follows:

11. The IoT access control method based on smart contracts and trust assessment according to claim 9, characterized in that: Indirect recommendation trust value IRT described in step 3.2 i，j The calculation of includes the following steps: Step 3.2.1: Calculate the indirect recommendation trust value of the k-th trust path The indirect recommendation trust value of the k-th trust path The calculation formula is as follows: Among them, the kth trust path is direct interactive trust The formed slave node RO i To RC j The path, x1,x2,…,x n is the serial number of all nodes passed in the k-th trust path; Step 3.2.2: Calculate the indirect recommendation trust value IRT i,j Node RO i RC for nodes j Indirect recommendation trust value IRT i,j For slave node RO i To Node RC j The mean of the indirect recommendation trust values ​​of each trust path between them is calculated as follows: Where m represents the slave node RO i To Node RC j The total number of trust paths.

12. The IoT access control method based on smart contracts and trust assessment according to claim 9, characterized in that: The comprehensive trust value CT described in step 3.3 i,j The calculation is evaluated using the following formula: CT i,j =γ×DIT i,j +ε×IRT i,j (8) Among them, γ and ε are the weights of direct interaction trust and indirect recommendation trust respectively, and γ, ε∈[0,1] and γ+ε=1.

13. The IoT access control method based on smart contracts and trust assessment according to claim 7, characterized in that: In step 4, the authorization decision process is to use the comprehensive trust value CT i,j Reach access control policy P r When the required trust value threshold is reached, the RO i Node, RC j If the attribute values ​​of the node, operation op, and context environment e meet the attribute requirements of the access control policy, it is RC j Send access token i,j (op,r).

14. The IoT access control method based on smart contracts and trust assessment according to claim 7, characterized in that: The policy violation determination in step 7 specifically includes the following steps: Step 7.1, RDS Verification RC j Whether to use expired or forged tokens i,j (op,r) access data; Step 7.2, RDS determines RC j Whether to send consecutive requests with a minimum interval less than the access token interval; Step 7.3: If the access control policy violation of step 7.1 or step 7.2 occurs, the access is recorded as a negative interaction, otherwise it is a positive interaction; Step 7.4: RDS notifies the CTR_TE contract to evaluate and update RO based on this interaction i About RC j The comprehensive trust value.

Images