Cloud instance privacy security enhancement method based on security channel dynamic measurement

Abstract

The invention discloses a cloud instance privacy security enhancement method based on security channel dynamic measurement, and the method comprises the steps: building a secure and credible communication channel through a secure communication client, and connecting the communication channel to a local user side through the Internet; a local user side deploys a TPCM secure communication module which is responsible for secure communication with the multi-cloud service rental instance. And the trusted software base is responsible for maintaining a measurement strategy and performing measurement judgment and control when a user process runs in the cloud service lease instance. The system specifically comprises a judgment mechanism module, a control mechanism module, a measurement mechanism module and a credible reference library. On the premise that infrastructures of cloud service providers are not trusted, in order to achieve safety and credibility of user remote cloud service lease instance data and application during operation and privacy protection of users, a measurement agent and a safety communication client are deployed in a cloud service lease instance; meanwhile, the tenant can perform deep monitoring on the process in the cloud instance, and it is ensured that sensitive information such as a monitoring strategy and reference data is not exposed to a cloud service provider.

Description

Technical Field

[0001] This invention relates to a method for enhancing the privacy and security of cloud instances based on dynamic measurement of secure channels, belonging to the field of cloud computing security technology. Background Technology

[0002] The development of internet technology has brought great convenience to human society. Cloud computing is another new innovation in the information age after the internet and computers. It represents a giant leap forward in the information age. This technology provides scalable and inexpensive computing services on demand through the network, with strong scalability and convenience, allowing users to access unlimited resources through the network, without being limited by time or space.

[0003] Leveraging these technological advantages, cloud computing has become a core engine for enterprise digital transformation, driving a comprehensive upgrade in business agility, flexibility, and innovation. However, with the widespread application of cloud computing technology, computer security issues in cloud computing scenarios are becoming increasingly prominent. Furthermore, in a cloud computing environment, when users or enterprises entrust their data and computing tasks to cloud service providers, the providers gain priority access to that data or application. Since this data or application may involve core business algorithms, personal privacy data, or sensitive logic, relying solely on the security protection provided by cloud service providers to defend against data breaches, authentication vulnerabilities, malware attacks, virtualization security issues, and other security risks has certain limitations. Therefore, in complex cloud computing scenarios, ensuring both the security of cloud computing resources and the protection of user and enterprise data and application privacy becomes increasingly important.

[0004] Trusted computing is a forward-looking security technology aimed at establishing a proactive information security protection system. Currently, to address the security risks of cloud computing, the national Cybersecurity Classified Protection System 2.0, released in 2019, requires trusted verification for all important information systems, including cloud computing. Faced with increasingly severe security challenges, the security protection of trusted systems should include multiple layers, considering both static and dynamic security. Meanwhile, the domestic trusted computing 3.0 industry ecosystem is developing rapidly; Trusted 3.0 is also known as proactive immune trusted computing. Proactive immune trusted computing technology employs a dual-engine architecture that combines computation and protection. It implements security protection while performing computation to ensure that the computation process and results always meet expectations, achieving measurability, controllability, and interference resistance in the computation process.

[0005] In order to ensure the security of user and enterprise data and applications in complex cloud computing scenarios, while preventing cloud service providers from spying on user privacy, there is an urgent need to develop a method based on trusted computing technology that can improve user privacy and security.

[0006] A cloud platform based on a Trusted Infrastructure Unit (TIPU), publication number CN119442250A, proposes a cloud platform architecture based on a Trusted Infrastructure Unit (TIPU), which offloads trusted measurement functions through a combination of hardware and virtualization technologies. The core design includes: a TIPU card hardware-integrated Trusted Cryptographic Module (TIPU-TCM) supporting hardware-level virtualization, providing a hardware root of trust (hTCM) and multiple virtual machine roots of trust (vTCM). The hTCM is used only for trusted support within the TIPU card itself, while the vTCM is dynamically scheduled to physical hosts and virtual machines to handle trusted measurement and security functions. Each hTCM and vTCM is assigned a unique UUID identifier. Key materials and vTCM status information are encrypted by an encryption / decryption engine and stored on a remote secure medium, decrypted and used after authentication via a USBKey. The TIPU card operating system deploys a physical driver and a vTCM simulator. The simulator is responsible for vTCM lifecycle management (creation, initialization, pause / resume, destruction) and request scheduling, interacting with virtual machines through front-end and back-end drivers, and employing priority queues and time-sharing mapping strategies to optimize instruction processing. The TIPU card can be powered via PCIe or independently, enhancing isolation and reliability.

[0007] This invention achieves trusted measurement functionality by combining hardware and virtualization technologies. Although it can solve the trusted measurement functionality at the underlying hardware level for multiple instances in cloud facilities, it still has certain shortcomings: (1) This solution relies on cloud devices with dedicated TIPU card hardware, which has high deployment costs and compatibility barriers for existing cloud platform infrastructure; (2) The measurement strategy, benchmark value, and measurement judgment are all completed in the hardware and software controlled by the cloud service provider, which has no privacy for user data and applications.

[0008] This patent, CN119337361A, discloses a method and system for blocking illegal programs detected in a dual-system trusted measurement architecture (host management platform + virtual machine agent) in a cloud environment. Its core innovation lies in the storage and access method of the whitelist. Specifically, each virtual machine registers its whitelist with the trusted management platform on the host. The platform stores all whitelists on a high-speed storage device and divides it into multiple read-only sub-devices mapped back to the corresponding virtual machines. From the virtual machine's perspective, a dedicated storage device, such as / dev / mem, is seen, containing its required whitelist. This whitelist is loaded into memory when the virtual machine starts. When a program within the virtual machine requests execution or a file is manipulated, a hook function is triggered. This function queries the whitelist in memory for real-time comparison (checking file existence, hash value matching, and attribute permissions). If a discrepancy is found, execution or operation is immediately blocked, allowing for immediate intervention in case of measurement failure while avoiding the risk of the whitelist being tampered with while stored within the virtual machine.

[0009] The core of this invention lies in enhanced whitelist access control. In a cloud environment, it employs a dual-architecture approach, deploying a trusted management platform on the host side and an agent on the virtual machine. The whitelist is generated by the virtual machine but stored on the host's high-speed storage device. This technology falls under the category of static trust and cannot cover dynamic runtime risks. It has limitations when facing memory attacks (such as cache overflows and code injection) and the replacement of runtime dynamic link libraries (DLLs). Furthermore, in a cloud environment, relying on the host side to store important information such as the whitelist, and having the virtual machine read the whitelist from the host through a mapping channel, even though the virtual machine side is read-only, this mapping channel itself could become a new potential attack surface. It is necessary to ensure that it cannot be used to steal or tamper with whitelist data. Additionally, from a user privacy perspective, completely entrusting the security of the user's virtual machine to the cloud service provider's host side offers no privacy whatsoever for the user's data and applications. Summary of the Invention

[0010] This invention proposes an Infrastructure as a Service (IaaS) leasing scenario in cloud computing, aiming to solve the core problem of security trust dilemma and data privacy leakage risk faced by cloud tenants (users) because their business processes run on the infrastructure provided by cloud service providers.

[0011] In the current technological context, cloud tenants' critical applications and sensitive data run on third-party cloud instances, making their security highly dependent on the security measures and business reputation of cloud service providers. This leads to two major pain points: First, cloud service providers' administrator privileges theoretically allow access to instance memory and disk, potentially exposing tenants' process behavior, core algorithms, and private data to leakage; second, existing remote authentication technologies mostly focus on static startup verification, lacking continuous monitoring and measurement of runtime process behavior, and thus failing to effectively defend against runtime injection, tampering, and other attacks.

[0012] Therefore, the purpose of this invention is to provide a solution that returns the initiative and control over security to cloud tenants, resolving the conflict between privacy protection and security monitoring without trusting the cloud service provider's infrastructure. This allows tenants to perform in-depth monitoring of processes within cloud instances while ensuring that sensitive information such as monitoring policies and baseline data is not exposed to the cloud service provider. Furthermore, it provides a fine-grained, continuous dynamic process measurement mechanism to compensate for the lack of runtime security capabilities and effectively detect runtime threats to processes.

[0013] To ensure secure and reliable data and application runtime for users' remote cloud service rental instances, and to protect user privacy, a measurement agent and a secure communication client are deployed within the cloud service rental instances, assuming distrust of the cloud service provider's infrastructure. A secure and reliable communication channel is established through the secure communication client, connecting to the local user side via the internet. On the local user side, a TPCM secure communication module is deployed, responsible for secure communication with multiple cloud service rental instances. A trusted software base is responsible for maintaining measurement policies, determining and controlling measurement during user process runtime within the cloud service rental instances. This includes a determination mechanism module, a control mechanism module, a measurement mechanism module, and a trusted benchmark library. A vTPCM cryptographic resource management module is responsible for storing cryptographic resources from multiple cloud service rental instances. At the hardware layer, a TCM trusted cryptographic module serves as the starting point of the trust chain, providing hardware cryptographic services to the aforementioned modules.

[0014] Secure Communication Client: This module is deployed in the cloud service rental instance and adopts a C / S architecture. After the cloud service instance starts, it actively establishes a connection with the local user-side TPCM secure communication module. The underlying transport layer protocol uses TCP, and a challenge-response mechanism is added on top of TCP to achieve robust two-way authentication, i.e., the client authenticates the server and the server authenticates the client. The specific process involves a six-way handshake operation: connection initiation and challenge (cloud service rental instance side -> local user side), server challenge request, client response to challenge and challenge the server (cloud service rental instance side -> local user side), server response to challenge and generation of temporary key pair, certificate issuance (local user side -> cloud service rental instance side), client verification and confirmation (cloud service rental instance side -> local user side), session key derivation and secure communication (two-way communication phase).

[0015] Metrics Proxy: This module is deployed within the cloud service rental instance. After a secure communication channel is established, it provides security protection for user processes within the cloud service rental instance during its operation, based on the metrics policy issued by the local user side. Specifically, it collects information about the process through the `task_struct` pointer within a certain time period, calculates a hash value as the metric, and then sends it to the local user side for real-time trusted metric measurement of the process. The local user-side decision-making mechanism module determines whether to stop the process based on the metric result. This module works in conjunction with the local user-side trusted software base to achieve a privacy architecture that separates metric and decision-making processes in the cloud service rental scenario, enabling collaboration between the cloud and local environments.

[0016] TPCM Secure Communication Module: This module is deployed on the local user side. As the server for both cloud service instances, it provides secure data communication between the local user and the client. The communication protocol design adds a challenge-response mechanism to the TCP protocol. Specifically, it uses the SM2 asymmetric encryption algorithm for key exchange and session key generation, the SM3 algorithm to ensure data integrity during communication, and the SM4 symmetric encryption algorithm to ensure data encryption. In particular, since it cannot be guaranteed that cloud service providers will not spy on user privacy or be secure and trustworthy, a temporary session key is generated for each client-server communication session to ensure forward security. This means that even if the long-term private key of the cloud instance is leaked in the future, it cannot be used to decrypt any historical communication sessions.

[0017] Trusted Software Base: This module is deployed on the local user side and includes a judgment mechanism module, a control mechanism module, a measurement mechanism module, and a trusted benchmark library. The trusted software module is used to judge the measurement values ​​of user processes from remote cloud service rental instances, generate judgment results, and manage measurement policies for different cloud service rental instances. The trusted benchmark library is responsible for storing the trusted benchmark values ​​of user processes for each cloud service instance.

[0018] Cryptographic Resource Management Module: This module is deployed on the local user side, providing cryptographic algorithm interfaces to the Trusted Software Base and TPCM secure communication module and managing secure communication session keys for multiple cloud service leased instances. Specifically, it involves session key generation, session key deletion, certificate signing, certificate challenge, etc.

[0019] TCM Trusted Cryptography Module: This module is installed on the local user side via hardware. It serves as the source of the overall system's trusted chain, providing hardware-based trusted cryptography services to upper-layer applications.

[0020] A method for enhancing cloud instance privacy and security based on dynamic measurement via a secure channel includes establishing a secure communication channel between the cloud service instance and the local user, distributing measurement policies, dynamically measuring user processes on the cloud service instance, determining measurement results on the local user side, and distributing the measurement determination results. Specifically, it includes the following steps: S1. A unique device identifier is generated for the cloud service rental instance. The local user side registers the unique device identifier of the cloud service rental instance into the trusted whitelist of the local trusted software base.

[0021] S2. The user configures a trusted policy for the cloud service rental instance. The cloud service rental instance-side measurement agent completes the first full measurement of the specified business process according to the trusted policy, generates an initial measurement value as a baseline value, sends it to the local user side, and stores it in the trusted baseline library.

[0022] S3, the cloud service instance metrics agent locally generates a pair of long-term SM2 asymmetric keys, and the user issues a digital certificate. The certificate and public key are synchronized to the local user-side password resource management module and stored in association with the instance's device identifier.

[0023] S4. The cloud service instance metrics agent and the local user side execute a six-way handshake protocol, including Hello and challenge exchange: Both parties exchange random number challenges Nonce_C, Nonce_S, and long-term certificate Cert_client. Two-way authentication: Both parties use their respective long-term private keys to sign and verify each other's challenges, completing robust two-way authentication. Temporary key distribution: After successful authentication, the local user side generates a temporary SM2 key pair (S_temp, P_temp), encrypts S_temp with the public key corresponding to the previously stored cloud service instance device identifier, and sends it to the cloud instance. Session key derivation: Both parties use the temporary key pair and the SM2 key negotiation protocol to derive a symmetric master key MasterSecret specific to this session.

[0024] S5, the cloud service instance measurement agent periodically traverses the `task_struct` structure pointers of user business processes with kernel privileges to collect their information, calculates hash values ​​as measurement values ​​for real-time reliable measurement, and then assembles the measurement results, process IDs, timestamps, and other information into a measurement message.

[0025] S6. The cloud service instance measurement agent uses the session key derived in step 4 to encrypt the measurement message and sends it to the local user via the Internet.

[0026] S7. The local user-side TPCM secure communication module receives the measurement message and decrypts it in a secure environment. The trusted software base determination mechanism module queries the trusted benchmark library to obtain the benchmark value of the corresponding process of the corresponding cloud service instance.

[0027] S8. Based on the trusted policy in the Trusted Software Base Measurement Mechanism module, generate corresponding control policies for the abnormal results in the judgment results of step S7. The TPCM secure communication module encrypts the result and sends it to the cloud service instance through the original secure communication channel.

[0028] S9, the cloud service instance-side measurement agent receives instructions, verifies data integrity, decrypts, and then executes them.

[0029] S10. After communication ends or times out, the cloud service instance-side metric agent and the local user-side password resource management module immediately clear all sensitive information of this session, including the temporary private key S_temp and the session master key MasterSecret. When establishing a connection again, step S4 will be repeated to generate a brand new temporary key and session key.

[0030] This completes the security protection process for cloud service rental instances.

[0031] The key point of this invention is that, in cloud service leasing scenarios, it breaks away from the traditional security approach of placing monitoring, analysis, and judgment all in the cloud, decoupling the functions: measurement is performed in an untrusted cloud environment, while analysis and judgment are performed in a trusted user's local environment. This solves user privacy issues to a certain extent by storing all sensitive information locally on the user's device. Cloud service providers cannot snoop on users' security policies and business privacy, thus clearly defining trust boundaries.

[0032] This invention targets cloud service leasing scenarios, enhancing user privacy protection to some extent while ensuring the security and trustworthiness of user data and applications. It addresses the issue of cloud tenants' sensitive business information being transparent to cloud service providers when using cloud services. Utilizing trusted computing dynamic measurement technology, it enables real-time detection of advanced threats such as code injection in memory. Users (especially enterprises dealing with personal information and core algorithms) can deploy their businesses on public clouds without worrying about cloud service providers snooping on their business logic and data. Furthermore, this invention fully adopts national cryptographic algorithms (SM2 / SM3 / SM4) and a trusted computing 3.0 technical architecture, complying with national cybersecurity level protection regulations. It can provide compliant and secure cloud migration solutions for critical industries such as government, finance, energy, and transportation. In terms of versatility, this invention is decoupled from the underlying cloud platform technology, adaptable to various virtualization architectures such as virtual machines and containers. The measurement strategy is configurable, easily expandable, and possesses a certain degree of universality. Attached Figure Description

[0033] Figure 1 This is the overall architecture diagram of the system of the present invention. Detailed Implementation

[0034] The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0035] The technical solution adopted in this invention is an Infrastructure as a Service (IaaS) leasing scenario in a cloud computing context, which aims to solve the core problem of security trust dilemma and data privacy leakage risk faced by cloud tenants (users) because their business processes run on the infrastructure provided by cloud service providers.

[0036] Specifically, in the current technological context, cloud tenants' critical applications and sensitive data run on third-party cloud instances, and their security highly depends on the security measures and business reputation of cloud service providers. This leads to two major pain points: First, cloud service providers' administrator privileges theoretically allow access to instance memory and disk, exposing tenants' process behavior, core algorithms, and privacy data to potential leakage risks; second, existing remote authentication technologies mostly focus on static startup verification, lacking continuous monitoring and measurement of runtime process behavior, and cannot effectively defend against runtime injection, tampering, and other attacks.

[0037] Therefore, the purpose of this invention is to provide a solution that returns the initiative and control over security to cloud tenants, resolving the conflict between privacy protection and security monitoring without trusting the cloud service provider's infrastructure. This allows tenants to perform in-depth monitoring of processes within cloud instances while ensuring that sensitive information such as monitoring policies and baseline data is not exposed to the cloud service provider. Furthermore, it provides a fine-grained, continuous dynamic process measurement mechanism to compensate for the lack of runtime security capabilities and effectively detect runtime threats to processes.

[0038] like Figure 1 The diagram shows the system architecture of this invention. To ensure the security and trustworthiness of user data and application runtime in remote cloud service rental instances, and to protect user privacy, a measurement agent and a secure communication client are deployed within the cloud service rental instances, under the premise of distrusting the cloud service provider's infrastructure. A secure and trustworthy communication channel is established through the secure communication client, connecting to the local user side via the Internet. The local user side deploys a TPCM secure communication module, responsible for secure communication with multiple cloud service rental instances. A trusted software base is responsible for maintaining measurement policies, determining and controlling measurement during user process runtime in the cloud service rental instances. Specifically, this includes a determination mechanism module, a control mechanism module, a measurement mechanism module, and a trusted benchmark library. The vTPCM cryptographic resource management module is responsible for storing cryptographic resources from multiple cloud service rental instances. At the hardware layer, a TCM trusted cryptographic module serves as the starting point of the trust chain, providing hardware cryptographic services to the above modules.

[0039] Secure Communication Client: This module is deployed in the cloud service rental instance and adopts a C / S architecture. After the cloud service instance starts, it actively establishes a connection with the local user-side TPCM secure communication module. The underlying transport layer protocol uses TCP, and a challenge-response mechanism is added on top of TCP to achieve robust two-way authentication, i.e., the client authenticates the server and the server authenticates the client. The specific process involves a six-way handshake operation: connection initiation and challenge (cloud service rental instance side -> local user side), server challenge request, client response to challenge and challenge the server (cloud service rental instance side -> local user side), server response to challenge and generation of temporary key pair, certificate issuance (local user side -> cloud service rental instance side), client verification and confirmation (cloud service rental instance side -> local user side), session key derivation and secure communication (two-way communication phase).

[0040] Metrics Proxy: This module is deployed within the cloud service rental instance. After a secure communication channel is established, it provides security protection for user processes within the cloud service rental instance during its operation, based on the metrics policy issued by the local user side. Specifically, it collects information about the process through the `task_struct` pointer within a certain time period, calculates a hash value as the metric, and then sends it to the local user side for real-time trusted metric measurement of the process. The local user-side decision-making mechanism module determines whether to stop the process based on the metric result. This module works in conjunction with the local user-side trusted software base to achieve a privacy architecture that separates metric and decision-making processes in the cloud service rental scenario, enabling collaboration between the cloud and local environments.

[0041] TPCM Secure Communication Module: This module is deployed on the local user side. As the server for both cloud service instances, it provides secure data communication between the local user and the client. The communication protocol design adds a challenge-response mechanism to the TCP protocol. Specifically, it uses the SM2 asymmetric encryption algorithm for key exchange and session key generation, the SM3 algorithm to ensure data integrity during communication, and the SM4 symmetric encryption algorithm to ensure data encryption. In particular, because it cannot be guaranteed that cloud service providers will not spy on user privacy or be secure and trustworthy, a temporary session key is generated for each client-server communication session to ensure forward security. This means that even if the long-term private key of the cloud instance is leaked in the future, it cannot be used to decrypt any historical communication sessions.

[0042] Trusted Software Base: This module is deployed on the local user side and includes a judgment mechanism module, a control mechanism module, a measurement mechanism module, and a trusted benchmark library. This module is used to judge the measurement values ​​of user processes from remote cloud service rental instances, generate judgment results, and manage measurement policies for different cloud service rental instances. The trusted benchmark library is responsible for storing the trusted benchmark values ​​of user processes for each cloud service instance.

[0043] Cryptographic Resource Management Module: This module is deployed on the local user side, providing cryptographic algorithm interfaces to the Trusted Software Base and TPCM secure communication module and managing secure communication session keys for multiple cloud service leased instances. Specifically, it involves session key generation, session key deletion, certificate signing, certificate challenge, etc.

[0044] TCM Trusted Cryptography Module: This module is installed on the local user side via hardware. It serves as the source of the overall system's trusted chain, providing hardware-based trusted cryptography services to upper-layer applications.

[0045] A method for enhancing cloud instance privacy and security based on dynamic measurement via a secure channel includes establishing a secure communication channel between the cloud service instance and the local user, distributing measurement policies, dynamically measuring user processes on the cloud service instance, determining the measurement results on the local user side, and distributing the measurement determination results. Its features include the following steps: A unique device identifier is generated for each cloud service rental instance. Local users register the cloud service rental instance to the trusted whitelist of the local trusted software base based on the unique device identifier.

[0046] 2. Users configure trusted policies for cloud service rental instances. The cloud service rental instance-side measurement agent performs the first complete measurement on the specified business process according to the trusted policy, generates an initial measurement value as a baseline value, sends it to the local user side, and stores it in the trusted baseline library.

[0047] 3. The cloud service instance metrics agent locally generates a pair of long-term SM2 asymmetric keys, and the user issues a digital certificate. The certificate and public key are synchronized to the local user-side password resource management module and stored in association with the instance's device identifier.

[0048] 4. The cloud service instance metrics agent and the local user side execute a six-way handshake protocol, mainly including Hello and challenge exchange: Both parties exchange random number challenges (Nonce_C, Nonce_S) and long-term certificates (Cert_client). Two-way authentication: Both parties use their respective long-term private keys to sign and verify each other's challenges, completing robust two-way authentication. Temporary key distribution: After successful authentication, the local user side generates a temporary SM2 key pair (S_temp, P_temp), encrypts S_temp with the public key corresponding to the previously stored cloud service instance device identifier, and sends it to the cloud instance. Session key derivation: Both parties use the temporary key pair and the SM2 key negotiation protocol to derive a symmetric master key, MasterSecret, specific to this session.

[0049] 5. The cloud service instance measurement agent periodically traverses the `task_struct` structure pointers of user business processes with kernel privileges to collect their information, calculates hash values ​​as measurement values, and performs real-time reliable measurement. Then, it assembles the measurement results, process ID, timestamp, and other information into a measurement message.

[0050] 6. The cloud service instance measurement agent uses the session key derived in step 4 to encrypt the measurement message and sends it to the local user via the Internet.

[0051] 7. The local user-side TPCM secure communication module receives the measurement message, decrypts it in a secure environment, and the trusted software base determination mechanism module queries the trusted benchmark library to obtain the benchmark value of the corresponding process of the corresponding cloud service instance.

[0052] 8. Based on the trusted policy in the Trusted Software Base Measurement Mechanism module, generate corresponding control policies (such as alarms, process termination, forced shutdown, etc.) for the abnormal results in the judgment results of step 7. The TPCM secure communication module encrypts the result and sends it to the cloud service instance through the original secure communication channel.

[0053] 9. The cloud service instance-side measurement agent receives the instruction, verifies the data integrity, decrypts it, and then executes it.

[0054] 10. Upon communication termination or timeout, the cloud service instance-side metric agent and the local user-side password resource management module immediately clear all sensitive information from this session, including the temporary private key S_temp and the session master key MasterSecret. Step 4 will be repeated the next time a connection is established to generate a completely new temporary key and session key.

[0055] This completes the security protection process for cloud service rental instances.

[0056] The key point of this invention is that, in cloud service leasing scenarios, it breaks away from the traditional security approach of placing monitoring, analysis, and judgment all in the cloud, decoupling the functions: measurement is performed in an untrusted cloud environment, while analysis and judgment are performed in a trusted user's local environment. This solves user privacy issues to a certain extent by storing all sensitive information locally on the user's device. Cloud service providers cannot snoop on users' security policies and business privacy, thus clearly defining trust boundaries.

[0057] This invention targets cloud service leasing scenarios, enhancing user privacy protection to some extent while ensuring the security and trustworthiness of user data and applications. It addresses the issue of cloud tenants' sensitive business information being transparent to cloud service providers when using cloud services. Utilizing trusted computing dynamic measurement technology, it enables real-time detection of advanced threats such as code injection in memory. Users (especially enterprises dealing with personal information and core algorithms) can deploy their businesses on public clouds without worrying about cloud service providers snooping on their business logic and data. Furthermore, this invention fully adopts national cryptographic algorithms (SM2 / SM3 / SM4) and a trusted computing 3.0 technical architecture, complying with national cybersecurity level protection regulations. It can provide compliant and secure cloud migration solutions for critical industries such as government, finance, energy, and transportation. In terms of versatility, this invention is decoupled from the underlying cloud platform technology, adaptable to various virtualization architectures such as virtual machines and containers. The measurement strategy is configurable, easily expandable, and possesses a certain degree of universality.

Claims

1. A method for enhancing cloud instance privacy and security based on dynamic measurement of secure channels, characterized in that, A secure and reliable communication channel is established through a secure communication client, connecting to the local user side via the Internet; the local user side deploys the TPCM secure communication module, which is responsible for secure communication with multiple cloud service leased instances; The Trusted Software Base is responsible for maintaining the measurement policy, determining and controlling the measurement of user processes during runtime in cloud service rental instances; the TPCM cryptographic resource management module is responsible for storing the cryptographic resources of multiple cloud service rental instances; the hardware layer has a TCM trusted cryptographic module as the starting point of the trust chain to provide hardware cryptographic services for the TPCM secure communication module.

2. The cloud instance privacy and security enhancement method based on dynamic measurement of secure channels according to claim 1, characterized in that, The system modules that implement this method include: Secure Communication Client: The secure communication client is deployed in the cloud service rental instance. It adopts a C / S architecture and actively establishes a connection with the local user-side TPCM secure communication module after the cloud service instance starts. The underlying transport layer protocol adopts the TCP protocol. A challenge-response mechanism is added on top of the TCP protocol to achieve robust two-way authentication, that is, the client authenticates the server and the server authenticates the client. Measurement Proxy: Deployed in the cloud service rental instance, after the secure communication channel is established, it performs security protection on user processes in the cloud service rental instance during the operation of the cloud service rental instance according to the measurement policy issued by the local user side; the measurement proxy and the local user-side trusted software base work together to achieve the separation of measurement and judgment in the cloud service rental scenario, and the privacy architecture of cloud and local collaboration. TPCM Secure Communication Module: The TPCM secure communication module is deployed on the local user side and acts as the server for both parties to communicate, providing secure data communication between multiple cloud service instances and local users; Trusted Software Base Module: Deployed on the local user side, it includes a decision mechanism module, a control mechanism module, a measurement mechanism module, and a trusted benchmark library; Cryptographic Resource Management Module: The cryptographic resource management module is deployed on the local user side. It provides cryptographic algorithm interfaces to the Trusted Software Base and TPCM secure communication module and manages secure communication session keys for multiple cloud service leased instances. Specifically, it involves session key generation, session key deletion, certificate signing, and certificate challenge. TCM Trusted Cryptography Module: Installed on the local user side of the hardware device; as the source of the overall system's trusted chain, it provides hardware trusted cryptography services to upper-layer applications.

3. The cloud instance privacy and security enhancement method based on dynamic measurement of secure channels according to claim 2, characterized in that, The specific process of secure communication client involves six handshake operations: connection initiation and challenge (cloud service rental instance side -> local user side), server challenge request, client response to challenge and challenge server (cloud service rental instance side -> local user side), server response to challenge and generate temporary key pair, certificate issuance (local user side -> cloud service rental instance side), client verification and confirmation (cloud service rental instance side -> local user side), session key derivation and secure communication (two-way communication phase).

4. The cloud instance privacy and security enhancement method based on dynamic measurement of secure channels according to claim 2, characterized in that, The measurement agent collects information about a process through the process's task_struct structure pointer within a certain time period, calculates a hash value as a measurement value, and then sends it to the local user side to perform real-time trusted measurement of the process. The local user side decision mechanism module determines whether to stop the process based on the measurement result.

5. The cloud instance privacy and security enhancement method based on dynamic measurement of secure channels according to claim 2, characterized in that, The TPCM secure communication module adds a challenge-response mechanism to the TCP protocol. Specifically, it uses the SM2 asymmetric encryption algorithm to achieve key exchange and session key generation between the two parties, the SM3 algorithm to ensure the integrity of data during communication, and the SM4 symmetric encryption algorithm to ensure data encryption during communication. The client and server temporarily generate a session key for each communication session to ensure forward security.

6. The cloud instance privacy and security enhancement method based on dynamic measurement of secure channels according to claim 2, characterized in that, The Trusted Software Module is used to determine the user process metrics from the remote cloud service rental instance, generate the determination results, and manage the metric policies for different cloud service rental instances. Its subordinate Trusted Benchmark Library is responsible for storing the trusted benchmark values ​​of user processes for each cloud service instance.

7. A cloud instance privacy and security enhancement method based on dynamic measurement of secure channels according to any one of claims 1-6, characterized in that, Specifically, the following steps are included: S1. A unique device identifier is generated for the cloud service rental instance. The local user side registers the unique device identifier of the cloud service rental instance into the trusted whitelist of the local trusted software base. S2. The user configures a trusted policy for the cloud service rental instance. The cloud service rental instance-side measurement agent completes the first full measurement of the specified business process according to the trusted policy, generates an initial measurement value as a baseline value, sends it to the local user side, and stores it in the trusted baseline library. S3, cloud service rental instance metrics agent locally generates a pair of long-term SM2 asymmetric keys, and the user issues a digital certificate; The certificate and public key are synchronized to the local user-side password resource management module and stored in association with the device identifier of the cloud service rental instance; S4. The cloud service instance metrics agent and the local user side execute a six-way handshake protocol, including Hello and challenge exchange: both parties exchange random number challenges (Nonce_C, Nonce_S) and long-term certificates Cert_client; two-way authentication: both parties use their respective long-term private keys to sign and verify each other's challenges, completing robust two-way authentication; temporary key distribution: after successful authentication, the local user side generates a temporary SM2 key pair (S_temp, P_temp), encrypts S_temp with the public key corresponding to the previously stored cloud service instance device identifier, and sends it to the cloud instance; session key derivation: both parties use the temporary key pair to derive the symmetric master key MasterSecret used for this session through the SM2 key negotiation protocol; S5, the cloud service instance measurement agent periodically traverses the task_struct structure pointer of the user's business process with kernel privileges to collect its information, calculates the hash value as the measurement value for real-time reliable measurement; and then assembles the measurement result, process ID, and timestamp information into a measurement message. S6. The cloud service instance measurement agent uses the session key derived in step 4 to encrypt the measurement message and sends it to the local user side via the Internet. S7. The local user-side TPCM secure communication module receives the measurement message and decrypts it in a secure environment. The trusted software base judgment mechanism module queries the trusted benchmark library to obtain the benchmark value of the corresponding process of the corresponding cloud service instance. S8. Generate corresponding control strategies for abnormal results in the judgment results of step S7 based on the trust policies in the Trusted Software Base Measurement Mechanism module. The TPCM secure communication module encrypts the result and sends it to the cloud service instance through the original secure communication channel; S9. The cloud service instance-side measurement agent receives the instruction, verifies the data integrity, decrypts it, and then executes it. S10. After communication ends or times out, the cloud service instance-side metric agent and the local user-side password resource management module immediately clear all sensitive information of this session, including the temporary private key S_temp and the session master key MasterSecret; when the connection is established again, step S4 will be repeated to generate a brand new temporary key and session key. This completes the security protection for cloud service rental instances.

Images