A trusted monitoring and auditing system fusing tee and blockchain

Abstract

The application relates to the technical field of transaction processing systems, and particularly discloses a trusted monitoring and auditing system integrating a TEE and a blockchain, which comprises the following steps: generating a transaction instantaneous state digest containing a key information field at a key node of a business process through a trusted execution environment, signing the transaction instantaneous state digest, anchoring a hash value of the transaction instantaneous state digest to a blockchain network, and storing the signed digest off-chain; and when auditing, verifying the integrity of the hash on the chain and the off-chain digest and the TEE signature, so as to realize double authentication of transaction authenticity and environment reliability. Through the collaborative architecture of the TEE and the blockchain, the application deeply integrates the transaction core data state and the trusted environment proof, so that the auditing verification can be independently completed without relying on a centralized database, and the problem of relying on the log credibility in the traditional auditing is fundamentally avoided, and a lightweight trusted verification basis is provided for cross-institution cooperation scenarios.

Description

TECHNICAL FIELD

[0001] The application is a trusted monitoring and auditing system integrating TEE and blockchain, belonging to the technical field of transaction processing system. BACKGROUND

[0002] In key business processes such as administrative supervision, financial transactions and commercial contract execution, ensuring the auditability and trustworthiness of the entire operation link has always been a core challenge in the technical field. Traditional auditing systems usually rely on centralized databases to store transaction logs and ensure data integrity through digital signatures. However, this model has two inherent defects: 1. The centralized storage architecture is vulnerable to single-point tampering risks, and it is difficult to establish mutual trust consensus in cross-institutional collaboration scenarios; 2. The ex-post tracing mechanism relies on the authenticity of log records, but there is no effective verification method for the security of the log generation environment and the physical operating state, making it difficult to deal with attack scenarios where malicious nodes fake environmental information.

[0003] In recent years, the industry has attempted to introduce blockchain technology to achieve non-tamperable storage of audit data, but pure on-chain storage has problems such as data inflation and low verification efficiency, and cannot penetrate the trusted state of the local computing environment when the transaction occurs. In addition, existing solutions generally ignore the precision of capturing the instantaneous state of key nodes in business processes, resulting in a disconnect between audit granularity and business logic, making it difficult to meet the regulatory needs of high-sensitivity scenarios such as administrative approval decision tracing and supply chain financial compliance verification.

[0004] More deeply, existing technologies have systemic design flaws in the coordination mechanism between trusted environments and distributed ledgers: Trusted Execution Environment (TEE) can ensure the verifiability of key code execution, but the trusted proof generated by it lacks dynamic association with business processes and has not been built with the cryptographic binding of physical environment state; the blockchain network provides anti-tamperability for data storage, but the verification path of on-chain and off-chain data is fragmented, and end-to-end closed-loop verification of transaction logic, environment state and stored data cannot be achieved. Such limitations are particularly pronounced in cross-domain multi-party collaboration scenarios, such as cross-border trade settlement, where participants need to verify the correctness of contract execution logic, the trustworthiness of clearing instruction generation environment, and the compliance of device operating state. However, existing technical solutions cannot achieve multi-dimensional trusted capture of key transaction nodes in a distributed architecture while ensuring verification efficiency. Therefore, how to achieve full-dimensional trusted capture of key transaction nodes in a distributed architecture, build a lightweight and environment-aware auditing and verification system, and establish a multi-level cryptographic association mechanism between transaction logic, trusted environment and physical state, thereby avoiding the technical bottlenecks of traditional auditing systems in real-time, complete trusted proof and cross-system verification efficiency, has become a technical problem to be solved by the present application. SUMMARY

[0005] The application provides a trusted monitoring and auditing system combining TEE and a blockchain, which mainly aims to solve the problems of incomplete instantaneous state capture at key transaction nodes, single trusted proof dimension and split verification path on and off the chain in the existing auditing system, thereby causing auditing blind spots and low efficiency.

[0006] To achieve the above-mentioned purpose, the application provides a trusted monitoring and auditing system combining TEE and a blockchain, which comprises:

[0007] A transaction processing module configured with a trusted execution environment, which is used to generate a transaction instantaneous state digest containing preset key information fields based on the current transaction context when the business process runs to a predefined key transaction node, and digitally sign the transaction instantaneous state digest by using the trusted execution environment; the transaction processing module is also used to perform hash calculation on the digitally signed transaction instantaneous state digest to generate an instantaneous state hash value;

[0008] An on-chain anchoring module connected with the transaction processing module, which is used to submit the instantaneous state hash value to a blockchain network for recording;

[0009] An off-chain storage module connected with the transaction processing module, which is used to store the digitally signed transaction instantaneous state digest;

[0010] An auditing verification module, which is used to obtain the instantaneous state hash value of a target transaction from the blockchain network and the corresponding digitally signed transaction instantaneous state digest from the off-chain storage module; the auditing verification module is also used to compare the obtained instantaneous state hash value with the hash value of the re-calculated transaction instantaneous state digest to verify the integrity of the transaction instantaneous state digest, and verify the digital signature by using the corresponding trusted public key of the trusted execution environment to confirm the authenticity of the source of the transaction instantaneous state digest and the reliability of the generation environment.

[0011] Preferably, the transaction processing module is also used to obtain a unique identifier of the current trusted execution environment instance when generating the transaction instantaneous state digest in the trusted execution environment, and encapsulate the unique identifier into the transaction instantaneous state digest to enhance the traceability and uniqueness of the transaction instantaneous state digest.

[0012] Preferably, the key transaction node is at least one of the following: a decision node in an administrative approval process, a performance confirmation node in a commercial contract execution, or a clearing instruction issuing node in a financial transaction.

[0013] Preferably, the blockchain network is a consortium chain or a lightweight private chain optimized for recording the instantaneous state hash value, and the consensus mechanism of the lightweight private chain is configured to prioritize recording efficiency and low-cost operation.

[0014] Preferably, the transaction processing module further comprises an environment sensing unit configured to acquire raw signals of a current physical environment state via at least one physical sensor deployed on a device hosting the trusted execution environment; the transaction processing module is further configured to process the raw signals into a standardized physical environment perturbation signature; the physical environment perturbation signature is included in the transaction instantaneous state digest or is cryptographically associated with the transaction instantaneous state digest or its instantaneous state hash value, and is processed or signed by the trusted execution environment, so that the instantaneous state hash value anchored to the blockchain network can reflect the state of the physical environment perturbation signature at the same time.

[0015] Preferably, when a plurality of critical transaction nodes in the processing system constitute a transaction cluster with a preset execution sequence, for any current transaction node in the transaction cluster except the first one, the transaction processing module is further configured to, when generating the transaction instantaneous state digest corresponding to the current transaction node, acquire a hash value of a transaction instantaneous state digest prequel generated by a previous critical transaction node adjacent to the current transaction node; the hash value of the transaction instantaneous state digest prequel is included in the transaction instantaneous state digest as prequel anchoring information; the on-chain anchoring module is configured to submit only the instantaneous state hash value of the transaction instantaneous state digest generated by the last critical transaction node in the transaction cluster, or a hash value of an aggregated digest representing all transaction instantaneous state digests of the entire transaction cluster, to the blockchain network for recording.

[0016] Preferably, the audit verification module judges the integrity of the transaction instantaneous state digest by the following condition:

[0017] ,

[0018] wherein, represents a hash value recalculated from the transaction instantaneous state digest acquired from the off-chain storage module; represents an instantaneous state hash value of a target transaction acquired from the blockchain network.

[0019] Preferably, the transaction processing module is configured to dynamically adjust the composition of the critical information field according to the business type corresponding to the critical transaction node when generating the transaction instantaneous state digest, so as to ensure that the transaction instantaneous state digest can accurately capture the core trusted elements in different business scenarios.

[0020] Preferably, the audit verification module is configured to trigger an audit or early warning mechanism for the transaction according to the indication of the physical environment perturbation signature; the judgment condition for triggering the threshold is: wherein, represents an abnormality level indicated by the physical environment perturbation signature, represents a preset minimum abnormality level threshold for triggering the audit or early warning mechanism.

[0021] The beneficial effects of the present application compared to the prior art are:

[0022] 1. Through the collaborative architecture of TEE and blockchain, the system generates and signs the transaction instantaneous state digest (TISD) in real time at the key nodes of the business process, and only anchors its hash value to the chain. This mechanism deeply integrates the core data state at the time of transaction occurrence with the trusted environment proof, so that the audit verification does not need to rely on centralized database, but can verify the authenticity and integrity of the transaction independently through the comparison of on-chain hash and off-chain digest. This design avoids the problem of relying on the credibility of logs in traditional audit, and provides a lightweight trusted verification basis for cross-institution and cross-system collaboration scenarios.

[0023] 2. By integrating physical sensor data into TEE to generate perturbation feature code (MPS), and cryptographically associating it with the transaction digest, the system not only ensures logical level trust, but also synchronously captures abnormal signals of the device running environment. This dual-dimension trusted binding mechanism enables the audit process to not only verify the correctness of transaction logic, but also trace the physical environment state at the time of transaction occurrence, effectively identifying potential physical attacks or abnormal operation scenarios, and providing multi-dimensional trusted protection for high-sensitive administrative decisions or financial transactions. For sequentially related transaction clusters, the system embeds the hash value of the previous transaction when generating each transaction digest, forming an endogenous time sequence chain. By verifying the anchor hash at the end of the chain, the integrity of the entire transaction chain can be traced back during audit. This chain structure, combined with the tamper-proof nature of blockchain, enables the full-cycle state of complex business processes to be quickly penetrated and verified, significantly improving the regulatory efficiency of multi-step administrative approval, supply chain finance, and other scenarios. By predefining key information field templates and dynamically adjusting strategies, the system can automatically capture core trusted elements in different scenarios based on business types, such as decision-making basis for administrative approval and clearing instruction parameters for financial transactions. This flexibility ensures that transaction digests accurately reflect business characteristics while avoiding redundant data storage, maintaining the lightweight of the overall architecture while ensuring audit granularity, and adapting to diverse scenario needs from government regulation to commercial contracts. BRIEF DESCRIPTION OF DRAWINGS

[0024] Figure 1 A transaction digest generation and dual verification time sequence diagram based on a multi-layer structure in the trusted monitoring and auditing system of the present application;

[0025] Figure 2 A transaction digest generation and blockchain anchoring and audit verification interaction flowchart based on a trusted execution environment in the present application;

[0026] Figure 3 A transaction instantaneous state digest generation flowchart based on business type and environment perception at key transaction nodes in the present application;

[0027] Figure 4 A schematic diagram of a transaction digest chain generation and storage structure based on hash chaining and physical perception of the present application.

[0028] The object implementation, functional features and advantages of the present application will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION

[0029] It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.

[0030] The embodiment of the present application provides a trusted monitoring and auditing system fusing TEE and block chain, and the system comprises:

[0031] A transaction processing module configured with a trusted execution environment, configured to generate a transaction instantaneous state digest containing preset key information fields based on a current transaction context when a business process runs to a predefined key transaction node, and to digitally sign the transaction instantaneous state digest by using the trusted execution environment; the transaction processing module is further configured to perform hash calculation on the digitally signed transaction instantaneous state digest to generate an instantaneous state hash value;

[0032] An on-chain anchoring module connected with the transaction processing module, configured to submit the instantaneous state hash value to a block chain network for recording;

[0033] An off-chain storage module connected with the transaction processing module, configured to store the digitally signed transaction instantaneous state digest;

[0034] An auditing verification module configured to obtain an instantaneous state hash value of a target transaction from the block chain network, and to obtain a corresponding digitally signed transaction instantaneous state digest from the off-chain storage module; the auditing verification module is further configured to compare the obtained instantaneous state hash value with a hash value of a re-calculated transaction instantaneous state digest to verify the integrity of the transaction instantaneous state digest, and to verify the digital signature by using a corresponding trusted public key of the trusted execution environment to confirm the authenticity of the source of the transaction instantaneous state digest and the reliability of the generation environment.

[0035] Preferably, the transaction processing module, when generating the transaction instantaneous state digest in the trusted execution environment, is further configured to obtain a unique identifier of a current trusted execution environment instance, and to encapsulate the unique identifier into the transaction instantaneous state digest to enhance the traceability and uniqueness of the transaction instantaneous state digest.

[0036] Preferably, the key transaction node is at least one of the following: a decision node in an administrative approval process, a performance confirmation node in a commercial contract execution, or a clearing instruction issuing node in a financial transaction.

[0037] Preferably, the blockchain network is a consortium chain, or a lightweight private chain optimized for recording instantaneous state hash values, and a consensus mechanism of the lightweight private chain is configured to prioritize recording efficiency and low-cost operation.

[0038] Preferably, the transaction processing module further comprises an environment sensing unit configured to acquire original signals of a current physical environment state through at least one physical sensor deployed on a device carrying a trusted execution environment; and the transaction processing module is further configured to process the original signals into a standardized physical environment perturbation feature code; the physical environment perturbation feature code is contained in the transaction instantaneous state digest or is cryptographically associated with the transaction instantaneous state digest or an instantaneous state hash value thereof, and is then processed or signed by the trusted execution environment, so that the instantaneous state hash value anchored to the blockchain network can reflect the state of the physical environment perturbation feature code.

[0039] Preferably, when a plurality of key transaction nodes in the processing system constitute a transaction cluster with a preset execution sequence, for any current transaction node in the transaction cluster except the first one, the transaction processing module is further configured to, in the trusted execution environment, generate a corresponding transaction instantaneous state digest current, and acquire a hash value of a transaction instantaneous state digest previous generated by a previous key transaction node adjacent to the current transaction node in the transaction cluster; the hash value of the transaction instantaneous state digest previous is contained in the transaction instantaneous state digest current as previous anchoring information; and the on-chain anchoring module is configured to submit only an instantaneous state hash value of a transaction instantaneous state digest generated by a last key transaction node in the transaction cluster, or a hash value of an aggregated digest representing all transaction instantaneous state digests of the entire transaction cluster, to the blockchain network for recording.

[0040] Preferably, the audit verification module verifies the integrity of the transaction instantaneous state digest by the following condition:

[0041] ,

[0042] Among them, represents a hash value recalculated from the transaction instantaneous state digest acquired from the off-chain storage module; represents an instantaneous state hash value of a target transaction acquired from the blockchain network.

[0043] Preferably, the transaction processing module is configured to dynamically adjust the composition of the key information field according to the business type corresponding to the key transaction node when generating the transaction instantaneous state digest, so as to ensure that the transaction instantaneous state digest can accurately capture core trusted elements in different business scenarios.

[0044] Preferably, the type of trusted execution environment (TEE) is selected from: Intel SGX compatible technology on the server side, or ARM TrustZone in embedded devices; the audit verification module is configured to trigger an audit or warning mechanism for the transaction according to the indication of the physical environment perturbation signature; the judgment condition for triggering the threshold is: wherein, represents the abnormality level indicated by the physical environment perturbation signature, represents the preset minimum abnormality level threshold for triggering the audit or warning mechanism; meanwhile, in order to improve the consistency and structural readability of the system technical expression, in the description of each module of the present application, all intermediate data related to the state capture of the key nodes of the business process are uniformly referred to as transaction instantaneous state digest (hereinafter referred to as TISD), and the TISD covers all transaction state snapshot contents generated, signed and involved in the on-chain anchoring and off-chain verification by the trusted execution environment. The expressions such as digest and signed digest in different application contexts all refer to the same concept. In order to ensure the integrity of the time sequence chain, the system adopts the method of embedding the previous digest hash value into the current TISD when constructing the hash nested structure of the transaction cluster to realize the internal anchoring association. The terms such as internal time sequence chain and previous anchoring information mentioned in the text are used to describe this structured hash linking mechanism, which aims to enhance the continuity and traceability of the transaction state at each stage. In addition, the core judgment formula for integrity comparison in the audit verification module is: wherein represents the hash function calculated again, represents the target hash recorded on the chain, and this formula is applicable to the verification scene of the consistency of the data at both ends of the chain for any target transaction, which ensures that the mathematical basis of the verification process is clear and clear, and is convenient for engineering implementation.

[0045] Embodiment 1: To further clarify the specific engineering implementation path of the trusted monitoring and auditing system integrating TEE and blockchain, the running logic, data interaction mechanism and configuration basis of key parameters of each functional module of the system are described in detail in combination with a typical application example in the financial transaction clearing scenario. In this embodiment, the system is deployed in a collaborative network jointly participated by banks, third-party clearing institutions and regulatory authorities. Each participant achieves account book consistency through a blockchain network based on a consortium chain. At the same time, all core transaction processing terminals are integrated with a trusted execution environment supporting Intel SGX technology to generate, sign and anchor the key state of a transaction in real time during the processing of each clearing instruction. When a cross-bank fund transfer instruction with a unique business number is received by a clearing node, the system enters the trusted monitoring process. First, the transaction processing module deployed in the node generates a transaction instantaneous state summary in accordance with the current instruction business context data (including but not limited to the originating bank number, the receiving bank number, the transfer amount, the timestamp, etc.) within its trusted execution environment. The summary is constructed according to the built-in field template of the system, and the field contents are directly derived from the transaction context being processed. After generation, the trusted execution environment digitally signs the above-mentioned summary content and calculates the hash value of the signature summary. The hash value is then submitted to the consortium chain network by the on-chain anchoring module for account book recording by each node in the network according to the pre-set consensus mechanism. It should be noted that, in order to avoid data expansion on the chain and improve processing efficiency, the original text of the summary is not directly uploaded to the chain in the system design, but only the hash value is submitted. The signature original text of the summary is saved in a local secure container through an off-chain storage module and is regularly backed up off-site to ensure that it can be completely retrieved and verified in future auditing processes.

[0046] To further improve the reliability of the summary data, the embodiment introduces an environment perception unit, which is deployed in a device carrying a trusted execution environment. The unit automatically collects typical physical disturbance information (such as electromagnetic interference, vibration spectrum characteristics, or environmental temperature and humidity, etc.) during the operation of the device through physical sensors, and forms a physical environment micro-disturbance feature code after standardization processing of the collected results. The feature code is packaged in the summary structure during summary generation, or is associated with the summary or its hash value through cryptographic methods at the key level, and is then uniformly processed and signed by the trusted execution environment, ensuring that external environmental changes are reflected at the summary level and cannot be forged. After the completion of the entire clearing process, the audit verification module can initiate authenticity verification of the target transaction at any point in time. The verification process includes obtaining the hash value of the target transaction from the blockchain network, and simultaneously retrieving the corresponding signed summary text from the off-chain storage module. The hash value of the summary text is recalculated and compared. If they are consistent, the verification engine is called to verify the summary signature using the preset trusted execution environment public key to confirm that the summary is generated by the trusted environment and the content has not been tampered with. If the audit process requires verification of the physical environment state, the system can further extract the physical disturbance feature code from the summary and compare it with the pre-set model parameters to determine whether there is an abnormal physical state.

[0047] In addition, considering that some business processes have multiple steps and strong sequential continuous transaction situations, the embodiment introduces a transaction cluster mechanism. Taking cross-period batch fund transfer as an example, the process includes multiple transaction steps, which are executed at different time nodes. When generating the summary of each sub-transaction, the hash value of the previous sub-transaction is embedded in the current summary, thereby constructing an internal time sequence chain. Finally, the system only anchors the hash value of the last sub-transaction to the blockchain network. During auditing, the end anchor node is verified, and the pre-sequence chain relationship embedded in the summary is combined, thereby achieving sequential integrity verification of the entire transaction cluster. For the various key parameters involved in the embodiment, their specific settings are based on the results of extensive testing and data analysis on a real financial transaction simulation platform during the system development phase. For example, the five-dimensional combined structure is used for the encoding method of the physical disturbance feature code, which is the optimal structure determined after verifying its anti-interference and identifiability in cross-device and multi-scenario environments. The fixed-length structure design is adopted for the summary field template, which is the engineering choice obtained after considering the summary calculation efficiency and audit clarity based on multiple rounds of optimization experiments.

[0048] Embodiment 2: In a typical financial clearing network, multiple key nodes of business processing (including clearing instruction initiation, fund arrival confirmation, cross-system settlement execution, etc.) have very high requirements for the credibility of the operating environment. To verify whether the system of the present application, which integrates a trusted execution environment (TEE) and a blockchain mechanism, has the ability to achieve high-trust capture and verification of instantaneous state at key transaction nodes, this test was carried out on the basis of combining a simulation platform with an entity device deployment environment, and the data used was desensitized example data.

[0049] The system deployment environment constructed a test system under a simulated financial transaction clearing platform, with a simulated transaction flow of about 250 per second. The system deployment included nodes supporting Intel SGX and ARM TrustZone TEEs, and standard non-TEE nodes were set as a control group. The blockchain network structure used a consortium chain architecture with 5 accounting nodes running the PBFT consensus mechanism. The MPS perception and acquisition module selected electromagnetic interference (EMI), environmental temperature and humidity changes, and micro-mechanical vibration as the input sources of MPS, with a sampling frequency of 200 Hz and a five-dimensional combined encoding structure for disturbance feature codes. The transaction model was set to construct a complete transaction cluster process consisting of four consecutive key nodes, each generating an independent TISD and embedding a previous summary hash value. Under the set business concurrency conditions, the system continuously processed 1000 valid clearing instruction transactions, and the following performance data was recorded:

[0050] Table 1: shows the time statistical data table of different task processing steps.

[0051]

[0052] Analysis: The average total time of TEE environment internal execution of TISD generation, signature and hash processing is less than 10 milliseconds, which can meet the processing capacity requirement of more than 50 per second. The test method is to inject known interference sources into the system under standard working environment and different disturbance scenarios, and repeat the test 100 times. The anti-interference stability index is the RMS variation rate of MPS in normal state. The abnormal recognition rate is the probability of correct identification and triggering of the audit mechanism by MPS.

[0053] Table 2: shows the sensor data analysis table under different environments.

[0054]

[0055] Analysis: MPS exhibits an extremely low mutation rate under normal conditions and an identification rate of over 90% under complex physical interference conditions. It has good stability and sensitivity, meeting the system's perception of physical disturbance changes in highly sensitive business environments. The test method is for the system to construct a transaction cluster containing 4 consecutive key transaction nodes. Each node generates an independent TISD and embeds the hash value of the previous node. Finally, only the hash value of the last node TISD is anchored to the chain; the verification process is for the audit module to trace back step by step from the last node to compare whether the previous hash chain in the summary is continuous, and verify the validity of the signature and the consistency of MPS.

[0056] Table 3: Verification result table for each node in the transaction process.

[0057]

[0058] It can be seen that the transaction chain constructed under the nested hash mechanism does not experience chain breaks or verification failures in the multi-step transaction process. The chain summary structure combined with the blockchain can achieve complete and continuous backtracking verification, with good sequentiality and consistency guarantees. This experiment comprehensively verifies the practical engineering feasibility of the solution of the present invention from three core dimensions: transaction summary generation efficiency, physical disturbance feature code recognition capability, and cross-node transaction chain consistency. The experiment shows that: the system has the ability to generate TISD in real time and stably and complete signature and hash processing in a high-concurrency environment; the MPS mechanism can effectively capture environmental anomaly information, with good anti-interference ability and recognition accuracy; the summary chain structure, combined with the on-chain anchoring design, has efficient backtracking and consistency verification capabilities in multi-step processes.

[0059] Example 3: This example combines Figures 1 to 4 , describes the implementation of a trusted monitoring and auditing system that integrates TEE and blockchain. Figure 1 As shown in the figure, the physical perception layer includes a temperature sensor and a power monitoring module, which respectively collect temperature signals (±0.5°C) and voltage fluctuation data as the basic input of the physical environment status; these data are then transmitted to the trusted execution environment of the trusted computing layer. In this trusted execution environment, the summary generator generates a summary containing a feature code based on the transaction context + environmental data, and calculates its hash value (SHA-256); the generated signature summary and hash value will be sent to the alliance chain node of the blockchain layer for block confirmation; the alliance chain layer adopts the alliance chain configuration: consensus mechanism: PBFT, block interval: 2 seconds to ensure the immutability and real-time nature of the data; after completing the on-chain record, the audit verification module on the audit end obtains the anchor hash from the blockchain layer, and performs double verification in combination with the off-chain data, that is, verifying the validity of the summary hash value and signature, and finally returning the verification result.

[0060] like Figure 2As shown, first, in the trusted execution environment, the system collects temperature sensor signals, power monitoring data, and device fingerprint collection, etc. original data through the physical environment perception module; combined with the transaction context data in the business input and the preset key field, the core processing module generates the instantaneous state digest, and embeds the environment feature code, then performs digital signature; finally, the cryptographic output generates the hash value (SHA-256) and the signature digest (ECC); the above signature digest is provided to the audit verification end, and the audit verification end further obtains the on-chain hash, verifies the off-chain digest, and verifies the environmental reliability; at the same time, the hash value (SHA-256) calculated in the trusted execution environment is anchored to the blockchain network, and the hash value is received and stored by the blockchain network, and a verification interface is provided; finally, the hash comparison verification is completed in the audit verification end, and by comparing the on-chain and off-chain data and verifying the legality of the signature digest (ECC) and the consistency of the generated environment, the integrity of the transaction and the reliability of the trusted execution environment are ensured.

[0061] As shown in Figure 3 , first, the system distinguishes administrative approval, financial transaction and other businesses in the business type judgment stage: for the administrative approval path, the key information field is dynamically adjusted to capture the decision basis; for the financial transaction path, the key information field is dynamically adjusted to capture the clearing instruction parameter; and for other businesses, the default or general key information field is used, then the transaction instantaneous state digest step is entered, and it is judged whether the physical environment perception unit is enabled, if enabled, the physical environment original signal is obtained, then the original signal is processed to generate the physical environment perturbation feature code, which is associated to the digest by cryptography; if not enabled, the TEE processing is directly performed, finally, the transaction instantaneous state digest generation is completed by the trusted execution environment processing or signature.

[0062] As shown in Figure 4 , the figure shows the generation and organization method of the transaction digest chain structure realized by combining the environment perception unit and the off-chain storage module in the transaction processing module (TEE environment) of the present application. In the transaction processing module, the environment perception unit first collects environmental data, including the temperature and power physical disturbance information at the time of the device, which is embedded in the transaction instantaneous state digests TISD-T1, TISD-T2, TISD-T3 and TISD-T4 generated in each stage, forming data encapsulation related to the actual running environment, at the same time, each digest (such as TISD-T1) will calculate its hash value H(T1) after generation, and the hash value is embedded into the next digest (such as TISD-T2) as the previous anchor information, realizing the chain hash link, this chain structure is TISD-T1, H(T1), TISD-T2, H(T2), TISD-T3, H(T3), TISD-T4 from top to bottom.

[0063] Embodiment 4: In the government process supervision scenario, a set of monitoring and auditing system integrating trusted execution environment and blockchain is deployed to implement full-process trusted recording and dynamic auditing verification on the key nodes in the construction project approval process. The system is configured in administrative approval terminal, on-site survey equipment and third-party evaluation platform and other core processing nodes. This embodiment takes the decision node in the major infrastructure project approval process as a typical application scenario, and systematically describes the specific operation process, parameter response logic and collaboration mechanism between modules of each component of the system in the scenario. In this scenario, when the project application completes the preliminary material review and enters the final approval stage, the system automatically activates the transaction processing module deployed in the approval terminal when the approval personnel starts the confirmation approval interface. The module first calls the transaction summary construction engine in the trusted execution environment, extracts seven core information fields including approval number, project code, fund size, evaluation rating, approval personnel identification, terminal location code and approval timestamp from the current approval context according to the field template preset for infrastructure approval scenario in the system. The above field template system is formed through field frequency analysis, audit reference value evaluation and data consistency modeling optimization in the early process test stage. The field order and type are uniformly set by the system to ensure that the data structure has clarity and consistency in the audit analysis process.

[0064] Thereafter, the environment-aware sub-module deployed in the trusted execution environment automatically collects disturbance signals in the current running environment of the approval terminal. In the embodiment, the selected disturbance sources include three physical indicators: power supply voltage fluctuation, background electromagnetic intensity, and device case micro-vibration. The system performs continuous data collection at a sampling frequency of one hundred times per second for two seconds, obtaining six hundred groups of original signal data. The disturbance processing engine processes the collected data based on Fourier transform and amplitude deviation analysis method to generate a set of five-dimensional disturbance feature codes, including voltage fluctuation distribution code, electromagnetic spectrum main peak code, acceleration waveform density code, device stability score code, and disturbance balance index code. The five-dimensional feature code structure is determined by the system through disturbance distinguishability evaluation and abstract signature compatibility analysis in more than three hundred typical physical interference scenarios, ensuring that the embedding of disturbance data in the abstract structure has stability and traceability; Subsequently, the transaction processing module encapsulates the above seven field data and disturbance feature code as a transaction instantaneous state abstract, which adopts a fixed-length splicing and field order signature strategy to ensure the consistency of cross-platform verification. After the abstract structure is constructed, the private key signature engine embedded in the trusted execution environment is called to digitally sign it. The signature algorithm used is a verifiable signature algorithm based on the elliptic curve cryptography system, and the overall signature process takes no more than five milliseconds to ensure the interactive experience during user operation. The signed transaction abstract is then converted into a unique hash value by the hash calculator, which is broadcast to the consortium chain network through the on-chain anchoring module. The consortium chain adopts a delegated Byzantine fault tolerance consensus mechanism to optimize the processing delay of government data, with a record time control within three seconds. At the same time, the off-chain storage module generates an index path according to the hash value, writes the original signed abstract package into an encrypted storage container with access control, and records the approval event identifier, terminal identifier, signature timestamp, and field template version number in the metadata index table. This design supports subsequent combined queries through approval number, hash value, or timestamp. The system internally uses an inverted index tree structure to improve retrieval efficiency, and supports fast positioning and calling of any abstract package in one thousand data within two seconds in actual application.

[0065] In a regulatory sampling process, auditors initiate a true audit request for a certain infrastructure approval event through the regulatory platform. The system calls the off-chain storage interface to retrieve the signature digest package corresponding to the approval matter by the audit verification module, and synchronously retrieves the corresponding anchor hash value from the alliance chain network. First, the audit module recalculates the hash value of the digest package in the verification engine and compares it with the chain record value for consistency. When the comparison result is consistent, the system continues to call the recordable trusted execution environment public key to verify the digest signature. If the physical disturbance verification function is enabled for the audit task, the system further extracts the disturbance feature code embedded in the digest and compares it with the standard disturbance model recorded by the approval terminal device. If the matching rate is higher than the preset trusted threshold, the system returns the physical state trusted identifier. Otherwise, it is marked as an environment abnormality to be reviewed and the warning log is recorded synchronously. For the parallel sub-processes involving finance, technology and other departments in the approval process of the project, the system integrates them using the transaction cluster organization strategy. Each sub-transaction node automatically embeds the hash value of the previous sub-transaction node digest as the previous anchor information when generating its digest, forming a sequential nested structure. Finally, only the digest hash value generated by the summary node (i.e. the decision approval node) is submitted to the blockchain for anchoring. This mechanism builds a clear and reasonably nested chain digest structure. When audited, the audit verification module traces back the nested hash chain in the digest from the end node, and the verification path is clear. The system can support up to eight levels of transaction nesting, and ensure that the verification depth does not affect the front-end operation response performance. The test results show that under the condition of six levels of transaction nesting, the average time for the system to complete the full-chain backtracking and verification process is not more than five seconds, and the verification accuracy is 100%. In summary, by deploying a trusted execution environment and a physical disturbance perception mechanism at the high-sensitivity nodes of government approval, this embodiment builds a transaction trust record process with clear structure, controllable traceability and real-time response capability. Combined with the on-chain anchoring and off-chain verification dual-path mechanism, it effectively realizes the full-process authenticity verification and environment state review of the approval behavior without relying on a centralized log system, which belongs to the extended implementation mode known to those skilled in the art.

[0066] Embodiment 5: The present application provides a deterministic procedure for determining the physical environment micro-disturbance feature code and the audit threshold, to ensure the physical environment integrity of the unattended financial clearing server at the time of issuing critical transaction instructions, thereby effectively preventing attacks on the internal operation of the trusted execution environment through physical means such as power transient disturbance or causing cabinet micro-vibration. When a server deployed in a remote data center processes large amount of financial clearing instructions, it must establish a quantifiable and reproducible mechanism to determine whether the physical environment is disturbed by malicious attacks at the moment of generating the transaction instantaneous state digest.

[0067] To this end, the transaction processing module performs a one-time offline environment baseline calibration procedure after the first deployment of the system or major hardware changes. When the procedure is started, the environment perception unit continuously collects more than ten thousand sample periods of multi-source physical sensor raw signals under the benchmark conditions of the server being in standard idle load and the environment being stable. The sources include electromagnetic interference, environmental temperature and humidity, and slight mechanical vibration. For the raw time series signals collected by each sensor, the calibration procedure uses short-time Fourier transform to calculate the energy mean and standard deviation in the preset frequency band. These statistical values are solidified into a device-specific structured standard physical environment model and securely stored in the encrypted storage area of the trusted execution environment. The establishment process of the model provides a stable and statistically significant reference benchmark for subsequent quantitative comparison of real-time environment status.

[0068] In the system operation, when the business process triggers the clearing instruction to issue this critical transaction node, the environment perception unit starts real-time disturbance analysis. This analysis captures the current physical sensor signals in a sliding time window of two seconds and performs the same feature extraction algorithm as in the offline calibration phase to calculate the real-time energy mean. Then, the system calculates the difference between the real-time energy mean and the corresponding benchmark mean in the standard physical environment model, and divides it by the standard deviation in the model to obtain the standard score of each sensor deviating from the benchmark state, i.e., the normalized deviation score. This score is quantized and mapped to a certain integer interval to form the dimension codes of the physical environment disturbance feature code. The system calculates the overall physical environment abnormality level based on these codes. This level is determined by the following weighted summation formula: In this formula, is the total number of physical sensors, is the quantized encoding of the normalized deviation score of the th sensor, is the weight of the th sensor. This weight is not empirically set, but is determined by applying a set of known type simulation attacks that inject power noise and analyzing the resulting signal response sensitivity during the offline calibration phase. This amplifies the influence of the sensors that are most sensitive to known attack patterns.

[0069] Correspondingly, the judgment threshold for triggering the review or warning mechanism also follows a deterministic process setting. The system uses the ten thousand sets of baseline environment data collected during the offline calibration phase to calculate their corresponding historical distribution. is set to the 99.9th percentile value of this historical distribution. This setting ensures that the system false positive rate is at an acceptable extremely low level under normal environmental fluctuations, while maintaining high sensitivity to significant abnormal disturbances that exceed this statistical boundary. Finally, when the clearing instruction is calculated and confirmed to be lower than The rank value is encapsulated into the transaction transient state digest together with the feature code itself, and is signed by the trusted execution environment. This mechanism inextricably binds the physical environment state at the moment with the transaction logic by cryptography, so that the subsequent audit can not only verify the authenticity of the transaction content, but also trace back the integrity of the physical environment at the moment of generation, thereby effectively covering the blind area of traditional audit and providing deep trusted security for high-security scenarios.

[0070] It is apparent for those skilled in the art that the present application is not limited to the details of the foregoing exemplary embodiments, and the present application can be implemented in other concrete forms without departing from the spirit or essential characteristics of the present application.

[0071] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application and are not limiting. Although the present application has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that the technical solutions of the present application can be modified or replaced equivalently without departing from the spirit and scope of the technical solutions of the present application.

Claims

1. A trusted monitoring and auditing system integrating TEE and blockchain, characterized by: The system comprises: A transaction processing module, configured with a trusted execution environment, is configured to generate a transaction instantaneous state summary containing preset key information fields based on the current transaction context when the business process runs to a predefined key transaction node; and digitally sign the transaction instantaneous state summary using the trusted execution environment. The transaction processing module is further configured to perform a hash calculation on the digitally signed transaction instantaneous state summary to generate an instantaneous state hash value; An on-chain anchoring module, connected to the transaction processing module, for submitting the instantaneous state hash value to the blockchain network for recording; An off-chain storage module, connected to the transaction processing module, for storing digitally signed transaction instantaneous status summaries; An audit verification module is configured to obtain a target transaction's instantaneous state hash value from the blockchain network and a corresponding digitally signed transaction instantaneous state summary from the off-chain storage module; the audit verification module is further configured to compare the obtained instantaneous state hash value with the hash value of the recalculated transaction instantaneous state summary to verify the integrity of the transaction instantaneous state summary; and to verify the digital signature using a trusted public key corresponding to a trusted execution environment to confirm the authenticity of the source of the transaction instantaneous state summary and the reliability of the generation environment; The transaction processing module also includes an environment perception unit, which is used to obtain the original signal of the current physical environment state through at least one physical sensor deployed on the device that carries the trusted execution environment; the transaction processing module is also used to process the original signal into a standardized physical environment perturbation signature code; the physical environment perturbation signature code is included in the transaction instantaneous state summary, or is cryptographically associated with the transaction instantaneous state summary or its instantaneous state hash value, and then processed or signed by the trusted execution environment.

2. A trusted monitoring and auditing system integrating TEE and blockchain according to claim 1, characterized in that: When generating the transaction transient state summary in the trusted execution environment, the transaction processing module is further configured to obtain a unique identifier of the current trusted execution environment instance and encapsulate the unique identifier into the transaction transient state summary.

3. A trusted monitoring and auditing system integrating TEE and blockchain according to any one of claims 1 or 2, characterized in that: A key transaction node is at least one of the following: a decision-making node in an administrative approval process, a performance confirmation node in commercial contract execution, or a settlement instruction issuance node in a financial transaction.

4. A trusted monitoring and auditing system integrating TEE and blockchain according to claim 1, characterized in that: The blockchain network is a consortium chain or a lightweight private chain optimized for recording instantaneous state hash values.

5. A trusted monitoring and auditing system integrating TEE and blockchain according to claim 1, characterized in that: The transaction processing module is configured to, when a transaction cluster having a preset execution order is formed by a plurality of predefined key transaction nodes in the processing system, generate a corresponding transaction transient state summary for any current transaction node other than the first in the transaction cluster within the trusted execution environment, and further to obtain a hash value of a preceding transaction transient state summary generated by a key transaction node immediately preceding the current transaction node in the transaction cluster; The hash value of the preceding transaction state summary is included in the current transaction state summary as the preceding anchor information; the on-chain anchoring module is configured to submit only the instantaneous state hash value of the transaction state summary generated by the last key transaction node in the transaction cluster, or the hash value of an aggregate summary representing the instantaneous state summaries of all transactions in the entire transaction cluster, to the blockchain network for recording.

6. A trusted monitoring and auditing system integrating TEE and blockchain according to claim 1, characterized in that: The audit verification module verifies the integrity of the transaction instantaneous state summary based on the following conditions: ， in, Represents the hash value recalculated from the transaction instantaneous state summary obtained from the off-chain storage module; Represents the instantaneous state hash value of the target transaction obtained from the blockchain network.

7. A trusted monitoring and auditing system integrating TEE and blockchain according to claim 1, characterized in that: The audit verification module is configured to trigger a review or early warning mechanism for a transaction based on the indication of the physical environment perturbation signature code; the judgment conditions for the trigger threshold are: ,in, Indicates the abnormality level indicated by the physical environment perturbation signature code, Indicates the preset minimum abnormality level threshold that triggers the review or warning mechanism.

Images