A trust computing system and computing method capable of defining behavior
By combining the trust calculation methods of self-certification and other certificates, using the trust evaluation of the hardware security module and the supervisory server, the problem of the trust evaluation of the trust computing node in the existing technology is solved, and the trust evaluation of the trust computing node is realized. Dynamic trust evaluation and security warning enhance the security of trust computing nodes within the network.
Patent Information
- Application Number
- CN202310152431.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2023-02-23
- Publication Date
- 2024-11-19
- Estimated Expiration
- 2043-02-23
AI Technical Summary
The existing trust computing technology only relies on self-certified data and cannot meet the trust of all parties. It lacks supervision and authentication of computing participants in the network, and cannot make correct predictions on the security and trustworthiness of trusted computing nodes, which violates zero trust. principle.
A trust computing system that can define behavior is proposed, combining self-certification and other certificates, through multi-party participation in consensus computing, using hardware security modules, self-certification modules and other certificates modules, the supervisory server conducts trust evaluation, and continuously verify through trust algorithms and determine the behavior consistency and compliance of trust computing nodes.
Dynamic trust computing for trust computing nodes is realized, and unknown dangerous behaviors can be discovered and warned of in a timely manner, the security and trustworthiness of trust computing nodes in the network is enhanced, and a more reliable trust computing environment is built.
Smart Images

Figure CN116170156B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the fields of trusted computing and network security technology, and in particular to a trust computing system and a computing method capable of defining behaviors. Background Art
[0002] 1. The trusted computing environment proposed by the international organization TCG verifies the integrity of the hardware and software measurement results of the trusted computing nodes. It is based on the trusted root TPM (Trusted Platform Module) to statically measure the hardware and software stack of the trusted computing nodes. The entire link passes through the CPU, OS LOADER, OS and APP, and transfers the corresponding control rights, spreading the trust relationship to the entire system to ensure that the integrity of the hardware devices, systems and applications of the trusted computing nodes is not destroyed, thereby ensuring that the system or software runs in the trusted state expected by the design goal.
[0003] 2. ARM released the TrustZone technology white paper in December 2008. The design based on the TrustZone hardware architecture uses "isolation" technology to divide the system operating environment into TEE (Trusted Execution Environment) and REE (Rich Execution Environment), ensuring the integrity of the system to a large extent and scope.
[0004] 3. The trusted immune computing proposed by domestic scholars is a computer security system that combines cryptographic operations with protection. It is based on the Trusted Platform Control Module (TPCM), which serves as the core module of the trusted source of trusted computing nodes. It provides active measurement, active control, encryption protection and other functions for trusted computing nodes to ensure the integrity and confidentiality of trusted computing nodes.
[0005] Disadvantages of the above three technologies: The above three technologies only solve the self-proven method of trusted computing nodes based on signature and integrity measurement mechanisms. However, the trusted boot chain model based on measurement, or the trust construction based on isolating different operating systems is too simple. Only self-proven data cannot satisfy the trust of all parties, and continuous measurement of the dynamic behavior of entities is not achieved. There is a lack of supervision and authentication of computing participants in the network, and it is impossible to make correct predictions and judgments on the security and credibility of trusted computing nodes in the network, which violates the principle of zero trust. Summary of the invention
[0006] In order to solve the deficiencies of the above three technologies, the present invention proposes a trust computing system and computing method with definable behavior, combining self-certification with external certification, overcoming the trust computing limitations of trust computing nodes or regional isolation, and realizing dynamic computing of trust. The present invention does not assume that the computing participants in the network are trustworthy by default. It requires multiple parties in the network (regulators, authenticators, computing parties, data parties, platform parties, etc.) to participate in consensus computing. The computing party provides data information that has been signed and authenticated, and declares that the dynamic behavior of the commitment is completed through continuous trust consensus computing verification, ensuring that the computing process of data in the network is safe and reliable.
[0007] In order to achieve the above object, the present invention adopts the following technical solutions:
[0008] In one aspect, the present invention provides a trust computing system capable of defining behaviors, comprising a plurality of trust computing nodes, interfaces and trust evaluation points;
[0009] The trust computing node includes a hardware security module, a self-certification module and a third-party certification module; the hardware security module has a built-in cryptographic component; the self-certification module is a combination of software and hardware used for signature measurement of key device information of the trust computing node, collection and verification of trusted commitment behavior, remote certification of data information and trust evaluation; the third-party certification module is a combination of software and hardware used to implement third-party trust observation and trust evaluation of the trust computing node;
[0010] The interface includes a trusted report generation engine; the trusted report generation engine is used to aggregate and verify the data collected by the self-certification module and the trust evaluation information of the other certification module, and transmit them to the trust evaluation point through a cryptographic channel;
[0011] The trust evaluation point includes a supervisory server; the supervisory server is used to continuously collect data transmitted by the trusted report generation engine, determine whether it meets the compliance and consistency of the definable behavior, and through the trust algorithm, continuously verify and judge the collected data to evaluate the trust level of the trusted computing node.
[0012] Furthermore, the self-certification module is also used for:
[0013] The hardware security module statically measures the UEFI, GRUB, OS and key process software stack of the trusted computing node to complete the establishment of the trust chain; the system software stack when the trusted computing node is started is measured by measuring the trusted boot chain, and the trust relationship is gradually extended to the entire trusted computing node. At the same time, the measurement values are stored to ensure that the logical combination in the data processing process can be discovered in time after being tampered with or destroyed, thus building a trusted computing environment; the local data of the trusted computing node during operation is collected as a local trusted data storage record.
[0014] Furthermore, the other certification module is also used for:
[0015] Through supervision and certification by computing participants, the network behavior information of trusted computing nodes is collected dynamically in real time as a storage record of trust consensus data.
[0016] Furthermore, the supervisory server is also used for:
[0017] The data of the trusted computing node is collected through the cryptographic channel. The data of the self-certification module-related components are verified by comparing them with the internally stored benchmark values of the trusted computing node. The data of the other-certification module-related components are collected from other trusted computing nodes in the network to analyze and determine whether the data of the other-certification module-related components of the trusted computing node are abnormal.
[0018] Furthermore, the trust algorithm includes:
[0019] The supervisory server obtains the self-certification weight and other-certification weight score ratio of the current category of trusted computing nodes according to the set security policy;
[0020] At the same time, the supervisory server determines the self-certification result based on the current state and the factory-set benchmark value. When the trusted computing node is powered on, its internal trusted boot chain performs SM3 measurement on GRUB through the node BIOS, GRUB measures the GRUB configuration file and OS file, and the OS software stack measures the key APP. The measurement value is recorded and stored and sent to the supervisory server at the same time. If the measurement value received by the supervisory server does not match the internally stored trusted boot chain benchmark value, it is as follows:
[0021] The supervisory server verifies and analyzes the self-certification result. If the verification fails, it proves that the trust node has been tampered with, and the self-certification score is judged to be 0;
[0022] If the BIOS's measurement of GRUB does not match the benchmark value stored on the supervisor's server, the trusted boot chain is considered unsafe and the entire boot chain score is 0;
[0023] When the GRUB measurement values for the GRUB configuration file and OS file do not match the baseline values, GRUB is considered safe, but the subsequent configuration files and OS files are not safe, and the entire boot chain is scored as 10 points;
[0024] When the OS file's measurement value for the critical APP does not match the baseline value, the BIOS and GRUB are considered safe, but the critical APP is not safe, and the entire boot chain is scored as 20 points;
[0025] If all measurement values are consistent with the benchmark values stored in the supervisory server, the self-certification score is 30 points;
[0026] The supervisory server determines the result of the other certification, obtains the process list of the current trusted computing node, and compares it with the process whitelist stored in the supervisory server. If a process that does not exist in the process whitelist appears, the server's behavior is considered inconsistent with the declared behavior, and the current trusted node is deducted based on the number of dangerous processes and their weight.
[0027] The supervisory server analyzes the interaction between other trusted computing nodes in the network and the current trusted computing node. If other trusted computing nodes determine that the network traffic of the current trusted computing node is abnormal or that there is access to a dangerous port, 20 points will be deducted from the other certificate score.
[0028] The initial trust score of each trust calculation node is 100 points, and points are deducted according to the abnormal error status, and finally the trust score of the current trust node is obtained.
[0029] Another aspect of the present invention further provides a trust calculation method capable of defining behavior, comprising:
[0030] Step 1: When the trusted computing node is powered on, the trusted boot chain performs SM3 measurement on GRUB through the trusted computing node BIOS, GRUB measures the GRUB configuration file and OS file, and OS measures the key APP, and the measurement values are recorded and stored;
[0031] Step 2: During the operation of the trusted computing node, the self-certification module obtains the device's system version number, firewall switch status, device permission password security status, and process whitelist local behavior, and stores the above data as local trusted data; the other-certification module continuously and dynamically monitors the device's network behavior as trusted consensus data;
[0032] Step 3: The trusted report generation engine in the trusted computing node aggregates the local trust data and the trust consensus data into a trusted report and sends it to the supervisory server through a cryptographic channel;
[0033] Step 4: The supervisory server receives the trusted report uploaded by the trusted computing node in the network, verifies and compares it with the expected data, analyzes and processes the data uploaded by different trusted computing nodes in the network, determines abnormal data and abnormal trusted computing nodes, and issues an alarm and reports the abnormal trusted computing nodes.
[0034] Compared with the prior art, the present invention has the following beneficial effects:
[0035] Compared with traditional network security protection, the components of the other certification module of the trusted computing node are authenticated and supervised by the supervisor and the authenticator to jointly build a trust relationship among the computing participants in the network. The present invention receives the data of the trusted computing node in the network through the supervisor server, and performs trust evaluation on the data inside the supervisor server. It can discover dangerous behaviors and data that cannot be detected inside the trusted computing node, and timely warn of unknown dangerous behaviors, solving the self-security problems of the trusted computing node (computer equipment). BRIEF DESCRIPTION OF THE DRAWINGS
[0036] Figure 1 This is one of the schematic diagrams of the architecture of a trust computing system capable of defining behaviors according to an embodiment of the present invention;
[0037] Figure 2 This is a second schematic diagram of the architecture of a trust computing system capable of defining behaviors according to an embodiment of the present invention;
[0038] Figure 3 The figure is a flow chart of a trust calculation method capable of defining behavior according to an embodiment of the present invention. DETAILED DESCRIPTION
[0039] The present invention will be further explained below with reference to the accompanying drawings and specific embodiments:
[0040] like Figure 1 , Figure 2 As shown, a trust computing system capable of defining behaviors includes a plurality of trust computing nodes, interfaces and trust evaluation points;
[0041] The trust computing node includes a hardware security module, a self-certification module and a third-party certification module; the hardware security module has a built-in cryptographic component; the self-certification module is a combination of software and hardware used for signature measurement of key device information of the trust computing node, collection and verification of trusted commitment behavior, remote certification of data information and trust evaluation; the third-party certification module is a combination of software and hardware used to implement third-party trust observation and trust evaluation of the trust computing node;
[0042] The interface includes a trusted report generation engine; the trusted report generation engine is used to aggregate and verify the data collected by the self-certification module and the trust evaluation information of the other certification module, and transmit them to the trust evaluation point through a cryptographic channel;
[0043] The trust evaluation point includes a supervisory server; the supervisory server is used to continuously collect data transmitted by the trusted report generation engine, determine whether it meets the compliance and consistency of the definable behavior, and through the trust algorithm, continuously verify and judge the collected data to evaluate the trust level of the trusted computing node.
[0044] Specifically, the indicators of trust computing include four parts: identity consistency, entity integrity, behavior consistency and behavior compliance of trust computing nodes.
[0045] The identity consistency of the trusted computing node is that the trusted computing node within the network signs the information data collected by the self-certification module and verifies the signature through the supervisory server. If the supervisory server fails the signature verification, it proves that the identity information of the trusted computing node has been tampered with.
[0046] During the power-on startup of the trusted computing node, the built-in measurement component adds measurements from UEFI, GRUB, OS to key processes one by one. The measurement value is signed by the cryptographic component built into the hardware security module of the trusted computing node and sent to the supervisory server for verification to complete the remote trust evaluation. If the measurement value does not match the node benchmark value stored on the server, it proves that the entity integrity of the node is destroyed.
[0047] The consistency of definable behavior means that the components of the trusted computing node collect their own process information and the network interaction behavior information with the adjacent nodes in the network, extract the trust standard information of the defined behavior, analyze the audit information of each component through local data verification, and send the verification results to the supervisory server for trust evaluation. The supervisory server verifies and judges based on the received data information and the internally stored benchmark value. At the same time, the trusted computing node deploys a verification module to collect and observe the characteristic behavior of the trusted computing node, and sends the data to the supervisory server. The supervisory server conducts trust evaluation based on the characteristic behavior information of other trusted computing nodes in the network through comparative analysis.
[0048] The compliance of definable behaviors in trust computing is that the trust computing nodes in the network need to declare their own behavioral characteristics to the supervisory server in order to prove that they are safe. Specifically, they include the following behaviors: 1. Declare that the promised behaviors are legal and compliant; 2. Plan behaviors for specific scenarios; 3. The number of declared promised behaviors is limited, and other behaviors are dangerous behaviors in the current scenario. The supervisory server has administrator policies deployed inside: 1. Determine whether the individual behavior of the node is compliant; 2. Determine whether the group behavior rules of all nodes controlled by the entire computing network are compliant.
[0049] Furthermore, self-certification is a trust evaluation based on the computing node's behavior commitment and self-attestation of behavior data. The self-certification module is a combination of hardware and software that combines signature measurement of key device information of trusted computing nodes, collection and verification of trusted commitment behaviors, remote attestation of data information, and trust evaluation.
[0050] Heteronormative authentication is a way to evaluate the trustworthiness of computing nodes based on the computing node behaviors and behavior legitimacy judgment norms (codes of conduct) observed from a third-party perspective; the heteronormative authentication module is a combination of software and hardware that implements third-party trust observation and trust evaluation of computing nodes.
[0051] The self-certification module completes the measurement of the system software stack when the computing platform (trusted computing node) is started by measuring the trusted boot chain, and the trust relationship is extended to the entire trusted computing node step by step. At the same time, the measurement value is stored to ensure that the logical combination in the data processing process can be discovered in time after being tampered with and destroyed, so as to build a trusted computing environment. The self-certification module collects the local data of the trusted computing node during operation as a local trusted data storage record; the other certification module collects the network behavior information of the trusted computing node in real time and dynamically after supervision and certification by the computing participants, as a trusted consensus data storage record. The trusted report generation engine aggregates and verifies the local trust data and the trust consensus data, and sends them to the trusted server certified by the third party, that is, the supervisory server, through the cryptographic channel. The supervisory server continuously collects the local trust data and trust consensus data of the trusted computing nodes in the network, determines whether they meet the compliance and consistency of the definable behavior, and continuously verifies and judges the collected data through the trust algorithm to evaluate the trust level of the computing node.
[0052] The present invention supports flexible deployment in a variety of scenarios, including endpoint security protection, cloud security management, container / virtual machine security certification, industrial network equipment security certification and other sites.
[0053] Based on traditional trusted computing, the present invention adds a definition of trusted commitment behavior of trusted computing nodes, node computing participants build security policies, and the supervisory server trusts the algorithm to jointly ensure the network security of trusted computing nodes.
[0054] Specifically, the components related to the self-certification module of the trusted computing node statically measure the UEFI (Unified Extensible Firmware Interface), GRUB (GRand Unified Bootloader, the GNU project's multi-operating system boot program), OS (Operating System) and key process software stack of the trusted computing node through the hardware security module, complete the establishment of the trust chain, obtain the measurement value and sign and authenticate it, and send the self-certification signature value to the supervisory server through the cryptographic channel to ensure that the integrity of the trusted computing node's operating system and key processes are not tampered with.
[0055] The components related to the other certification module of the trusted computing node need to be authenticated by the authenticator and supervised by the supervisor during the computing process to confirm that the corresponding software of the other certification module has been installed and has not been tampered with. The components related to the other certification module of the trusted computing node continuously and dynamically collect the hardware security module information, special cryptographic component information, local computing behavior and network interaction behavior of the computing environment.
[0056] The components related to the trusted computing node certification module continuously collect network interaction behaviors during the operation of the trusted computing node, including but not limited to related businesses, related subjects (nodes, users, service interfaces), and the processing and storage behavior information of the data provider, and send them to the supervisory server for verification. The supervisory server is supervised by the supervisor of the computing participants, and determines whether the current behavior information of the trusted computing node meets the expected results and whether there is any dangerous behavior based on the certification data reported by other trusted computing nodes in the network and the internally deployed behavior database. If it is consistent with the behavior rules, it can be determined that the current behavior of the trusted computing node is safe, and the construction of multi-party trust relationships in the computing process is completed in a remote proof manner.
[0057] The supervisory server collects data from the trusted computing node through a cryptographic channel. For the data of components related to the self-certification module, it is verified by comparing it with the benchmark value of the trusted computing node stored inside the supervisory server. For the data of components related to the other-certification module, the supervisory server collects data of components related to the other-certification module of other trusted computing nodes in the network, and analyzes and determines whether the data of components related to the other-certification module of the trusted computing node is abnormal.
[0058] The supervisory server uses the built-in trust algorithm to evaluate the results of self-certification and other-certification, and obtains the trust score of the current computing node. The trust algorithm is as follows:
[0059] Step A: The supervisory server obtains the score ratio of the self-certification weight (30%) and the other-certification weight (70%) of the current category of trusted computing nodes according to the set security policy;
[0060] Step B: The trust algorithm inside the supervisory server determines the self-certification result, which needs to be determined based on both the current state and the factory-set benchmark value. When the trusted computing node is powered on, its internal trusted boot chain performs SM3 measurement on GRUB through the node BIOS, GRUB measures the GRUB configuration file and OS file, and the OS software stack measures the key APP. The measurement value is recorded and stored and sent to the supervisory server at the same time. If the measurement value received by the supervisory server does not match the internally stored trusted boot chain benchmark value, it will be as follows:
[0061] The supervisory server verifies and analyzes the self-certification result. If the verification fails, it proves that the trust node has been tampered with, and the self-certification score is judged to be 0;
[0062] When the BIOS's measurement of GRUB does not match the benchmark value stored on the supervisor's server, the trusted boot chain is considered unsafe and the entire boot chain is scored as 0;
[0063] When the GRUB measurement values for the GRUB configuration file and OS file do not match the baseline values, GRUB is considered safe, but the subsequent configuration files and OS files are not safe, and the entire boot chain is scored as 10 points;
[0064] When the OS file's measurement value for the critical APP does not match the baseline value, the BIOS and GRUB are considered safe, but the critical APP is unsafe, and the entire boot chain is scored 20 points;
[0065] If all measurement values are consistent with the benchmark values stored internally in the supervisory server, the self-certification score is 30 points.
[0066] Step C: The trust algorithm inside the supervisory server determines the result of the other certification. It needs to obtain the process list of the current trusted computing node and compare it with the process whitelist stored inside the supervisory server. If a process that does not exist in the process whitelist appears, it is considered that the server's behavior is inconsistent with the declared behavior. The current trust node is deducted from the trust score, and the score is deducted according to the number of dangerous processes and the weight of the proportion. If there is one process not in the whitelist, 10 points will be deducted. When there are 5 or more processes not in the process whitelist inside the server, 50 points will be deducted.
[0067] The supervisory server analyzes the interaction between other trusted computing nodes and the current trusted computing node in the network. If other trusted computing nodes determine that the network traffic of the current trusted computing node is abnormal or that there is access to a dangerous port, 20 points will be deducted from the other certificate score.
[0068] Step D: The initial trust score of each trust calculation node is 100 points, and points are deducted according to the abnormal error status, and finally the trust score of the current trust node is obtained.
[0069] Based on the above embodiments, Figure 3 As shown, the present invention also proposes a trust calculation method capable of defining behavior, including:
[0070] Step 1: When the trusted computing node (computer device) is powered on, the trusted boot chain performs SM3 measurement on GRUB through the device BIOS, GRUB measures the GRUB configuration file and OS file, and OS measures the key APP, and records and stores the measurement values;
[0071] Step 2: During the operation of the device, the self-certification module obtains the device's system version number, firewall switch status, device permission password security status, and process whitelist local behavior, and stores the above data as local trust data; the other-certification module continuously and dynamically monitors the device's network behavior as trust consensus data;
[0072] Step 3: The trusted report generation engine in the device aggregates the local trust data and the trust consensus data into a trusted report and sends it to the supervisory server through a cryptographic channel;
[0073] Step 4: The supervisory server receives the trusted reports uploaded by the devices in the network, verifies and compares them with the expected data, analyzes and processes the data uploaded by different devices in the network, determines abnormal data and abnormal devices, and issues alarms and reports on abnormal devices.
[0074] This trust computing method is mainly used for deployment in complex network environments. It analyzes the network environment of the trusted computing node through the consistency authentication of the identity of the trusted computing node, the integrity verification of the entity, and the consistency and compliance of the definable behavior. The related components of the self-certification module of the trusted computing node complete the static verification by signing the identity information of the trusted computing node and measuring the integrity of the trusted boot chain, collect and process the defined behavior of the node, and send it to the regulatory server for compliance judgment. The trust algorithm deployed in the regulatory server conducts trust evaluation on the trusted commitments and consensus computing data of the trusted computing nodes in the network, and completes the timely prediction and prevention of unknown dangers.
[0075] In summary, compared with traditional network security protection, the components related to the other certification module of the trusted computing node are authenticated and supervised by the supervisor and the authenticator, and jointly build a trust relationship among the computing participants in the network. The present invention receives the data of the trusted computing node in the network through the supervisor server, and performs trust evaluation on the data inside the supervisor server. It can discover dangerous behaviors and data that cannot be detected inside the trusted computing node, and promptly conduct early warning processing on unknown dangerous behaviors, thereby solving the self-security problems of the trusted computing node (computer equipment).
[0076] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.
Claims
1. A trust computing system capable of defining behavior, characterized in that: Includes multiple trust computing nodes, interfaces and trust evaluation points; The trusted computing node includes a hardware security module, a self-certification module and a third-party certification module; The hardware security module has a built-in cryptographic component; the self-certification module is a combination of software and hardware used for signature measurement of key equipment information of trusted computing nodes, collection and verification of trusted commitment behaviors, remote certification of data information, and trust evaluation; the other-certification module is a combination of software and hardware used to implement third-party trust observation and trust evaluation of trusted computing nodes; The interface includes a trusted report generation engine; the trusted report generation engine is used to aggregate and verify the data collected by the self-certification module and the trust evaluation information of the other certification module, and transmit them to the trust evaluation point through a cryptographic channel; The trust evaluation point includes a supervisory server; the supervisory server is used to continuously collect data transmitted by the trusted report generation engine, determine whether it meets the compliance and consistency of the definable behavior, and continuously verify and judge the collected data through the trust algorithm to evaluate the trust level of the trusted computing node; The trust algorithm includes: The supervisory server obtains the self-certification weight and other-certification weight score ratio of the current category of trusted computing nodes according to the set security policy; At the same time, the supervisory server determines the self-certification result based on the current state and the factory-set benchmark value. When the trusted computing node is powered on, its internal trusted boot chain performs SM3 measurement on GRUB through the node BIOS, GRUB measures the GRUB configuration file and OS file, and the OS software stack measures the key APP. The measurement value is recorded and stored and sent to the supervisory server at the same time. If the measurement value received by the supervisory server does not match the internally stored trusted boot chain benchmark value, it is as follows: The supervisory server verifies and analyzes the self-certification result. If the verification fails, it proves that the trust node has been tampered with, and the self-certification score is judged to be 0; If the BIOS's measurement of GRUB does not match the benchmark value stored on the supervisor's server, the trusted boot chain is considered unsafe and the entire boot chain score is 0; When the GRUB measurement values for the GRUB configuration file and OS file do not match the baseline values, GRUB is considered safe, but the subsequent configuration files and OS files are not safe, and the entire boot chain is scored as 10 points; When the OS file's measurement value for the critical APP does not match the baseline value, the BIOS and GRUB are considered safe, but the critical APP is not safe, and the entire boot chain is scored as 20 points; If all measurement values are consistent with the benchmark values stored in the supervisory server, the self-certification score is 30 points; The supervisory server determines the result of the other certification, obtains the process list of the current trusted computing node, and compares it with the process whitelist stored in the supervisory server. If a process that does not exist in the process whitelist appears, the server's behavior is considered inconsistent with the declared behavior, and the current trusted node is deducted based on the number of dangerous processes and their weight. The supervisory server analyzes the interaction between other trusted computing nodes in the network and the current trusted computing node. If other trusted computing nodes determine that the network traffic of the current trusted computing node is abnormal or that there is access to a dangerous port, 20 points will be deducted from the other certificate score. The initial trust score of each trust calculation node is 100 points, and points are deducted according to the abnormal error status, and finally the trust score of the current trust node is obtained.
2. A trust computing system capable of defining behavior according to claim 1, characterized in that: The self-certification module is also used to: The hardware security module statically measures the unified extensible firmware interface UEFI of the trusted computing node, the multi-operating system boot program GRUB of the GNU project, the operating system OS and the key process software stack to complete the establishment of the trust chain; the system software stack when the trusted computing node is started is measured by measuring the trusted boot chain, and the trust relationship is gradually extended to the entire trusted computing node. At the same time, the measurement value is stored to ensure that the logical combination in the data processing process can be discovered in time after being tampered with and destroyed, thus building a trusted computing environment; Collect local data of the trusted computing node during operation and store it as local trusted data.
3. A trust computing system capable of defining behavior according to claim 1, characterized in that: The other certification module is also used for: Through supervision and certification by computing participants, the network behavior information of trusted computing nodes is collected dynamically in real time as a storage record of trust consensus data.
4. A trust computing system capable of defining behavior according to claim 1, characterized in that: The supervisory server is also used for: The data of the trusted computing node is collected through the cryptographic channel. The data of the self-certification module-related components are verified by comparing them with the internally stored benchmark values of the trusted computing node. The data of the other-certification module-related components are collected from other trusted computing nodes in the network to analyze and determine whether the data of the other-certification module-related components of the trusted computing node are abnormal.
5. A method for trust computing with definable behavior based on a trust computing system with definable behavior according to any one of claims 1 to 4, characterized in that: include: Step 1: When the trusted computing node is powered on, the trusted boot chain performs SM3 measurement on GRUB through the trusted computing node BIOS, GRUB measures the GRUB configuration file and OS file, and OS measures the key APP, and the measurement values are recorded and stored; Step 2: During the operation of the trusted computing node, the self-certification module obtains the device's system version number, firewall switch status, device permission password security status, and process whitelist local behavior, and stores the above data as local trusted data; the other-certification module continuously and dynamically monitors the device's network behavior as trusted consensus data; Step 3: The trusted report generation engine in the trusted computing node aggregates the local trust data and the trust consensus data into a trusted report and sends it to the supervisory server through a cryptographic channel; Step 4: The supervisory server receives the trusted report uploaded by the trusted computing node in the network, verifies and compares it with the expected data, analyzes and processes the data uploaded by different trusted computing nodes in the network, determines abnormal data and abnormal trusted computing nodes, and issues an alarm and reports the abnormal trusted computing nodes.
