Blockchain-based computer motherboard test data security diagnosis system

Abstract

The application discloses a computer mainboard test data security diagnosis system based on a blockchain, relates to the field of the blockchain, and comprises the following steps: a data acquisition module captures and structurally encapsulates analog signals and digital instruction sequences of mainboard testing in real time; a blockchain network stores key information of data fingerprints on a chain to ensure that the key information is inalterable; a secure storage cluster completely saves original data packets, and position information of the original data packets is recorded on the chain; and an intelligent diagnosis module generates a reliable diagnosis report by checking data integrity, calling an analysis model and combining on-chain context. The application has the following advantages: by relying on high-precision data encapsulation and acquisition and blockchain distributed storage technology, the mainboard test data is inalterable, traceable, safe and verifiable, and the problem that traditional test data is easy to fake and difficult to prove is solved. In addition, the application combines a time sequence intelligent analysis model to perform multi-dimensional data comparison and diagnosis, and accurately locates test abnormalities and fault causes.

Description

Technical Field

[0001] This invention relates to the field of blockchain, and in particular to a blockchain-based computer motherboard test data security diagnostic system. Background Technology

[0002] Against the backdrop of increasingly complex digital manufacturing and supply chain management, traditional computer motherboard test data faces core challenges in storage, transmission, and sharing, including susceptibility to data tampering, low transparency, and difficulty in tracing responsibility. Blockchain technology, with its decentralized, immutable, and traceable characteristics, offers an innovative solution to these problems.

[0003] Current mainstream computer motherboard testing and diagnostic solutions suffer from several significant shortcomings. Most employ centralized local data acquisition and storage models, lacking high-precision timing synchronization and layered data encapsulation mechanisms. This results in the inability to completely and accurately retain the underlying electrical signals and instruction sequences of the motherboard, leading to issues such as data corruption, timing misalignment, and missing critical information. Furthermore, traditional methods lack a blockchain-based distributed evidence storage architecture. Test data is stored on a single device or server, lacking encrypted signatures and distributed consensus verification mechanisms, making it highly susceptible to human tampering, overwriting, or accidental loss. Data authenticity and integrity cannot be guaranteed, and the lack of a traceable chain ledger renders quality inspection and auditing without valid and credible evidence. In addition, conventional diagnostics rely heavily on manual experience or basic analysis algorithms, lacking multi-dimensional intelligent timing analysis capabilities. They cannot deeply uncover subtle signal anomalies or timing vulnerabilities across testing phases, nor can they perform cross-comparison across batches or historical test data. This results in vague fault location, inaccurate root cause analysis, and a lack of reliable verification of diagnostic results, making them ill-suited for the high-precision, high-security, and traceable industrial quality inspection requirements of high-end motherboards. Summary of the Invention

[0004] To improve existing systems, a blockchain-based computer motherboard test data security diagnostic system is provided. This method relies on high-precision data encapsulation and acquisition with blockchain distributed evidence storage technology to ensure that motherboard test data is tamper-proof, fully traceable, and securely verifiable, solving the problems of easy falsification and difficulty in obtaining evidence in traditional test data. Combined with a time-series intelligent analysis model, multi-dimensional data comparison and diagnosis are performed to accurately locate test anomalies and root causes of failures.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0006] A blockchain-based computer motherboard test data security diagnostic system includes:

[0007] Data acquisition and encapsulation module: Captures the underlying analog electrical signals and digital instruction sequences generated during the motherboard testing process in real time at the physical testing station, and performs high-precision synchronization, time slicing, structural encapsulation and digital fingerprint generation on the underlying analog electrical signals and digital instruction sequences to obtain a structured raw data packet with a hierarchical nested structure;

[0008] Blockchain Evidence Storage Network Module: Communicates and connects with the data acquisition and packaging module to form a blockchain network. It receives and verifies the structured raw data packets from the data acquisition and packaging module, and packages the data fingerprints, key features, and digital signatures into immutable blocks in chronological order and links them into a chain through a consensus algorithm.

[0009] Secure storage service cluster: Provides physical or logical storage space for persistent storage of complete original test data packets associated with the blockchain evidence storage network. It stores the structured original data packets generated by the data acquisition and encapsulation module. Its storage location information is recorded on the blockchain, and the on-chain fingerprint is associated with the complete original data off-chain.

[0010] Intelligent Diagnosis and Service Module: Authorizes access to the evidence records on the blockchain and the complete data packets in secure storage. It performs integrity verification of on-chain and off-chain data, calls pre-trained intelligent analysis models to perform in-depth analysis of the data, and combines the context information associated on the blockchain to generate a diagnostic report with verifiable credibility.

[0011] Preferably, the data acquisition and packaging module specifically includes:

[0012] Signal monitoring and capture unit: Real-time monitoring of the test equipment's data bus and debugging interface via a programmable gate array chip, synchronously capturing the underlying analog electrical signal flow and digital instruction sequence during the test process;

[0013] Timestamp and synchronization unit: Each captured low-level analog electrical signal stream and digital instruction sequence is stamped with a hardware timestamp with nanosecond-level precision and global monotonically increasing in real time. The timestamp is synchronized and aligned with the start and end instructions of the test item issued by the host computer of the test equipment.

[0014] Data Slicing and Encapsulation Unit: The captured real-time data stream is sliced ​​according to the start and end time windows of the preset test items. For the data in each time slice, the corresponding high-precision timestamp, the unique identifier of the test item, and the hardware fingerprint of the acquisition device itself are fused to generate a structured raw data packet with a hierarchical nested structure. The structured raw data packet includes a header layer that stores metadata, a data body layer that stores the raw signal and instruction byte arrays that have been losslessly compressed and encoded, and a verification layer that stores the Merkle root hash generated based on the data body content.

[0015] Preferably, the blockchain evidence storage network module specifically includes:

[0016] Blockchain network unit: contains multiple physically distributed nodes, including test terminal nodes, quality inspection center nodes and audit nodes. The test terminal nodes are integrated with a consensus preprocessing unit, which is implemented in the form of a hardware security module or a trusted execution environment, and is used to securely store the node's asymmetric encrypted private key.

[0017] The evidence storage transaction constructor unit receives the structured raw data packets from the data acquisition and encapsulation module, calculates the unique digital fingerprint of the data packets using a cryptographic hash function, digitally signs the fingerprint, key metadata, and Merkle root hash of the data packets using the protected node private key inside the HSM / TEE, and encapsulates them into a lightweight evidence storage transaction request.

[0018] Distributed consensus engine: Based on the Byzantine fault-tolerant consensus algorithm, the quality inspection center node is preset as the leading node, and the audit node and test terminal node are used as verification nodes. Each consensus node votes in multiple rounds to reach a consensus on the legality, signature validity and order of the received evidence storage transaction requests.

[0019] Chain-based ledger repository: After consensus, data is packaged into blocks in chronological order and linked to previous blocks through cryptographic hashes to form an immutable chain-based ledger, completing the on-chain anchoring of test data fingerprints and key features.

[0020] Preferably, in the consensus preprocessing unit of the test terminal node, the hardware security module or trusted execution environment is further configured as follows:

[0021] When calculating the cryptographic hash value, a collision-resistant cryptographic hash function is used to perform calculations on the entire byte sequence of the header layer, data body layer, and verification layer of the structured original data packet;

[0022] When generating a lightweight evidence storage transaction request, the index pointer is constructed as a composite data structure containing a network address, storage path, and unique identifier of the data packet;

[0023] The format of the evidence storage transaction request conforms to the standard transaction structure defined by the blockchain network, and its content fields include at least: transaction type identifier, sender node identifier, signed data fingerprint and Merkle root hash, key metadata, index pointer, and complete digital signature value.

[0024] Preferably, the consensus mechanism of the Byzantine fault-tolerant consensus algorithm specifically includes:

[0025] The consensus process is divided into three main stages: pre-preparation, preparation, and submission.

[0026] During the pre-preparation phase, the leading node sorts the received legitimate evidence storage transaction requests and packages them into a pre-preparation message, which is then broadcast to all verification nodes.

[0027] During the preparation and submission phase, each verification node conducts multiple rounds of interactive voting on the message. The transaction block that receives the approval of more than two-thirds of the verification nodes is finally submitted and written to the local ledger of each node.

[0028] The blockchain network also deploys a view replacement protocol. When a validator node does not receive a legitimate message from the dominant node within a preset time, it will trigger a view replacement process to elect a new dominant node according to a predetermined node order.

[0029] Preferably, the secure storage service cluster specifically includes:

[0030] Encrypted storage node array unit: Composed of multiple physical storage servers. When a complete data packet is received, it uses a one-time session key dynamically obtained from the blockchain network to encrypt the data packet, divides the encrypted data packet into multiple data fragments, attaches erasure codes, and stores them on different physical nodes.

[0031] On-chain indexing and verification interface unit: After the data packet is successfully stored, the interface will create a structured pointer containing its storage location, the key identifier required for decryption, and data fragment verification information. The hash value of the pointer is submitted to the blockchain network as a core element for evidence storage.

[0032] Preferably, the intelligent diagnosis and service module specifically includes:

[0033] Diagnostic Engine: Responding to diagnostic requests, it queries and retrieves relevant on-chain evidence records by calling smart contracts on the blockchain evidence storage network, and locates and retrieves complete structured raw data packets from the secure storage service cluster based on the index pointers in the evidence records.

[0034] Data integrity verification unit: Recalculates the cryptographic hash value of the retrieved structured original data packet and compares it with the corresponding data fingerprint obtained from the blockchain ledger. At the same time, it uses the public key of the test terminal node to verify the digital signature of the on-chain evidence.

[0035] Multidimensional time-series data analysis model unit: The model adopts an architecture that integrates deep convolutional neural networks and long short-term memory networks. The original signal flow and instruction sequence in the verified data packet are input into the model to learn local abnormal patterns in the signal waveform and capture time-series dependencies and state transition anomalies across test phases. Based on the on-chain evidence feature data related to the motherboard's historical test items or test items of the same batch of motherboards, cross-comparison and context analysis are performed to output structured diagnostic conclusions, including anomaly localization, root cause inference and confidence score.

[0036] Human-computer interaction interface: used to receive diagnostic requests, display diagnostic reports, allow users to trigger real-time hash and signature verification of the on-chain evidence data corresponding to key conclusions in the report, and display the verification results and related blockchain transaction hashes and block height information.

[0037] Preferably, the smart contract on the blockchain evidence storage network includes the following logic:

[0038] A publicly available mapping table data structure is used to store the association between the motherboard serial number or test batch number and a series of evidence-based transaction hashes;

[0039] An access-controlled registration function is provided for authorized test terminal nodes to call after a successful on-chain notarization transaction, in order to update the mapping table and establish the association between the test object and the on-chain notarization.

[0040] A public query function, which the diagnostic engine can call, takes a motherboard or batch identifier as input and returns a list of all associated evidence-based transaction hashes, serving as the entry point for data retrieval.

[0041] Preferably, the operating mechanism of the multidimensional time-series data analysis model specifically includes:

[0042] The deep convolutional neural network layer is composed of multiple one-dimensional convolutional layers and pooling layers stacked alternately, which automatically extract local spatiotemporal features from the original signal stream;

[0043] The long short-term memory network layer receives high-level feature sequences extracted by the convolutional layer, which are used to model long-distance temporal dependencies during the entire test item execution period and identify anomalies in the time dimension.

[0044] The cross-comparison and context analysis specifically refer to the following: when analyzing the current test item data, the diagnostic engine queries the smart contract and blockchain ledger to obtain the on-chain evidence characteristics of other test items that have been completed in the previous test process on the same motherboard and have electrical or logical connections, or performs distribution similarity calculations on the test data characteristics of the current motherboard and the same test item characteristics of other motherboards in the same batch to determine whether it is a systemic batch problem or an individual defect.

[0045] Compared with the prior art, the advantages of the present invention are:

[0046] Leveraging nanosecond-level high-precision synchronous acquisition and layered nested data encapsulation technology, the system fully preserves the underlying electrical signals and instruction timing data of the motherboard test, ensuring the integrity and standardization of the original test data. Through a multi-node distributed blockchain network and Byzantine fault-tolerant consensus mechanism, combined with hardware security module encrypted signatures, the system achieves chain-based solidification of test data fingerprints and key features, preventing data tampering and forgery, and ensuring full traceability and verification. Simultaneously, it employs an architecture of on-chain index storage and off-chain sharded encrypted erasure coding storage, balancing storage security and retrieval efficiency for massive test data. Based on a multi-dimensional temporal intelligent model fusion of convolutional and long short-term memory, combined with cross-comparison of on-chain historical data and test data from the same batch, it can accurately identify local signal anomalies and timing defects, precisely locate the root cause of faults, and output confidence scores. Coupled with real-time on-chain data verification, the system ensures authoritative and verifiable motherboard test diagnostic results, meeting the high-precision and high-security quality inspection requirements of industrial motherboards. Attached Figure Description

[0047] Figure 1 This is a schematic diagram of the blockchain-based computer motherboard test data security diagnostic system proposed in this invention;

[0048] Figure 2 This is a diagram of the blockchain evidence storage network module proposed in this invention;

[0049] Figure 3 This is a diagram of the intelligent diagnosis and service module proposed in this invention. Detailed Implementation

[0050] The following description is intended to disclose the invention and enable those skilled in the art to implement it. The preferred embodiments described below are merely examples, and other obvious variations will occur to those skilled in the art.

[0051] See Figure 1 As shown, the blockchain-based computer motherboard test data security diagnostic system includes:

[0052] Data acquisition and encapsulation module: Captures the underlying analog electrical signals and digital instruction sequences generated during the motherboard testing process in real time at the physical testing station, and performs high-precision synchronization, time slicing, structural encapsulation and digital fingerprint generation on the underlying analog electrical signals and digital instruction sequences to obtain a structured raw data packet with a hierarchical nested structure;

[0053] Blockchain Evidence Storage Network Module: Communicates and connects with the data acquisition and packaging module to form a blockchain network. It receives and verifies the structured raw data packets from the data acquisition and packaging module, and packages the data fingerprints, key features, and digital signatures into immutable blocks in chronological order and links them into a chain through a consensus algorithm.

[0054] Secure storage service cluster: Provides physical or logical storage space for persistent storage of complete original test data packets associated with the blockchain evidence storage network. It stores the structured original data packets generated by the data acquisition and encapsulation module. Its storage location information is recorded on the blockchain, and the on-chain fingerprint is associated with the complete original data off-chain.

[0055] Intelligent Diagnosis and Service Module: Authorizes access to the evidence records on the blockchain and the complete data packets in secure storage. It performs integrity verification of on-chain and off-chain data, calls pre-trained intelligent analysis models to perform in-depth analysis of the data, and combines the context information associated on the blockchain to generate a diagnostic report with verifiable credibility.

[0056] In the data acquisition and packaging module, the signal monitoring and acquisition unit is responsible for the synchronous acquisition of raw data. The programmable gate array (FPGA) chip directly connects to the data bus, debugging interface, and key analog test points of the test equipment through its programmable input / output pins and high-speed transceiver. The hardware logic implemented inside the FPGA drives multiple high-speed ADCs in parallel, digitizing the analog electrical signals at a preset sampling rate much higher than the characteristic frequency of the signal under test. Simultaneously, for the digital bus, the FPGA logic latches and records all transmitted instruction or data sequences in real time using the same system clock reference.

[0057] The timestamp and synchronization unit assigns a precise time stamp to each captured data unit. A high-stability, low-jitter temperature-compensated crystal oscillator integrated within the FPGA chip provides the base clock. An N-bit free-running counter continuously accumulates its count value under this clock drive. This serves as the system time base. When the data capture logic samples a data point, it synchronously latches the current counter value. The absolute count value is then converted into a timestamp with nanosecond precision. The key synchronization operation involves the test equipment's host computer simultaneously sending a hardware trigger pulse to the FPGA when sending the software command to start or end each test item. Upon detecting the edge of this trigger pulse, the FPGA latches the current counter value again. This is recorded as the synchronization anchor for the test item. The timestamps of all data points are synchronized using their latched count values. anchor point count value The difference is converted into a precise time offset relative to the start point of the test item. The formula is:

[0058]

[0059] in, The system clock frequency provided for the TCXO and This is the value latched by the counter.

[0060] The data slicing and encapsulation unit divides the continuous data stream into test item data blocks with independent semantics according to a preset test process configuration file. The configuration file defines the start and end time windows for each test item. The encapsulation logic is based on the timestamps of the data points. The data points are then categorized into the corresponding test item time window. For all data points belonging to the same test item, the original sample point array and the digital instruction byte array belonging to that test item are merged and compressed using a lossless compression algorithm to obtain the compressed data byte stream. The content of the data body is then used to calculate the root hash value of its Merkle tree using the SHA-256 hash function, serving as a strong checksum for data integrity.

[0061] Generate a hierarchical data packet with a fixed format. The data packet includes: a header layer: storing metadata, including unique identifiers for test items. The unique hardware fingerprint of the data acquisition device Absolute start time of test items Data body length, compression algorithm identifier, etc. Data body layer: stores the compressed original data byte stream. Verification layer: Stores the calculated Merkle root hash value. , and a packet tail check sequence used to verify the integrity of the packet structure.

[0062] Finally, the unit outputs a complete, self-contained structured raw data packet for each test item. This data packet can be formally represented as:

[0063]

[0064] Here, || represents byte concatenation. The data packet is then sent to secure storage, where its cryptographic hash is... This will be calculated and used in the subsequent blockchain evidence storage process.

[0065] In the blockchain evidence storage network module, a set of physically distributed server nodes interconnected via a network forms a permissioned consortium blockchain network. Nodes in the network are divided into three categories based on their roles: test terminal nodes bound to test workstations, quality inspection center nodes located in the quality control center, and audit nodes located in independent audit departments. All nodes jointly maintain a synchronous, immutable distributed ledger used to anchor the fingerprints of test data.

[0066] The notarized transaction constructor unit is the starting point for data on-chain, and its core is a Hardware Security Module (HSM) or a CPU-based Trusted Execution Environment (TEE). When the test terminal node receives the structured raw data packet from the data acquisition module... Then, the HSM / TEE first calculates its globally unique digital fingerprint. The calculation formula is as follows:

[0067]

[0068] Wherein, SHA-256(⋅) represents the SHA-256 cryptographic hash function, and the input is a data packet. All bytes, from Extract the Merkle root hash of its data body from the verification layer. Subsequently, HSM / TEE uses its internally stored, non-exportable node private keys. A digital signature σ is generated for key information. The signing process can be represented as follows:

[0069]

[0070] Here, Sign(⋅) represents a digital signature algorithm such as ECDSA or SM2. It includes key metadata such as timestamps and test item IDs. Finally, a notarization transaction request is constructed. Its data structure includes: Transaction Type identifier (Type), Sender Node Public Key identifier. Hash value and metadata The transaction request includes a Pointer index pointing to the secure storage location of the original data packet, and a digital signature σ. The transaction request is then broadcast to the blockchain network.

[0071] The distributed consensus engine implements a consensus mechanism based on the Practical Byzantine Fault Tolerance (PBFT) algorithm. The network pre-determines a quality control center node as the initial leading node, with audit nodes and a subset of high-reputation test terminal nodes serving as verification nodes. The consensus process consists of three phases:

[0072] During the pre-preparation phase, the leading node will receive legitimate... Sort by the order of receipt and package them into a candidate block. Generate a pre-preparation message, including the sequence number n and the current view number v. The broadcast is sent to all validator nodes. Where H(⋅) is the block hash. Sign the master node.

[0073] During the preparation and submission phase, after receiving the pre-preparation message, the verification node verifies the signature and transaction validity. If the verification passes, it broadcasts the preparation message. When a node receives requests from different verification nodes for the same... When the number of valid prepare messages exceeds 2f (where f is the maximum tolerable number of Byzantine nodes, and the total number of nodes is 3f+1), the commit phase begins, and a commit message is broadcast. .

[0074] When a node receives more than 2f valid commit messages for the same block, it will immediately commit that block. Marked as final confirmation.

[0075] The final confirmed block in the chained ledger repository is appended to the locally maintained blockchain. Each block The block header contains the following key fields: the hash of the previous block. Merkle root hash of all notarized transactions in this block , timestamp, consensus information, etc. The linking relationships between blocks are guaranteed by hash pointers:

[0076]

[0077] This structure makes it possible to access any historical blocks. Any alteration of the content will affect its hash value. The change causes the PrevHash recorded in the header of all subsequent blocks to not match, and the tampering is immediately exposed.

[0078] To ensure liveness, a view replacement protocol is triggered when a validator node does not receive a legitimate message from the dominant node within a preset timeout period, or detects malicious behavior by the dominant node. After the protocol is triggered, the validator node broadcasts a view replacement request and elects a new dominant node according to a predefined node order list. Once the new dominant node takes over, it resumes control of the consensus process from the latest stability checkpoint, ensuring continued network operation even when some nodes fail.

[0079] In a secure storage service cluster, an encrypted storage node array unit consists of multiple physical storage server nodes deployed within a data center. These nodes are interconnected via a network to form a unified storage resource pool. When the cluster receives a structured raw data packet from the data acquisition module... Then, proceed as follows:

[0080] The storage cluster management service first queries the blockchain network to obtain a temporarily generated symmetric session key that is uniquely linked to the data packet. This key is dynamically distributed via a key management smart contract within the blockchain network and features a one-time pad key. Subsequently, data packets are encrypted using Advanced Encryption Standards in Galois / counter mode. Encryption is performed to generate ciphertext C. The encryption process can be formally represented as:

[0081]

[0082] Here, IV is a randomly generated initialization vector, and Tag is an authentication tag generated by GCM mode.

[0083] The encrypted ciphertext C is divided into n data fragments. Subsequently, erasure coding algorithm is applied to generate m checksum fragments. These n+m fragments are distributed and stored across different physical nodes in the cluster. Even if any m fragments are lost or corrupted, the original ciphertext C can still be completely reconstructed from the remaining fragments. The data recovery process can be represented as:

[0084]

[0085] Where S and P are the sets of indices of the successfully acquired data fragments and the verification fragments, respectively, and .

[0086] The on-chain indexing and verification interface unit is a crucial bridge connecting off-chain storage and on-chain evidence storage, responsible for generating and managing verifiable data location pointers. When data packets... After successful encryption and fragmented storage, the interface creates a structured index pointer. This pointer is a machine-readable data structure containing a list recording the physical node network address and storage path within the node where each data fragment and verification fragment is stored, a reference to the identifier or decryption credential of the corresponding session key in the key management smart contract on the blockchain, and a list recording erasure coding parameters and fragment hashes. It will be serialized into a byte stream.

[0087] The interface calculates the cryptographic hash value of the index pointer. And it is used as one of the core evidence elements, along with data packet fingerprints. Together they are submitted to the blockchain network to constitute a notarized transaction. Part of it. That is: Includes:

[0088] in, A complete pointer It is then stored off-chain, and its hash is... It serves as its digital seal on the blockchain.

[0089] When services such as intelligent diagnostics need to access raw data, they first obtain the stored transaction information from the blockchain. and Then, submit by querying the pointer management service. To obtain the complete Verify SHA-256 ( Is it equal to the chain? This confirms that the pointer has not been tampered with. According to The system collects data fragments from various storage nodes based on the list information. During collection, the integrity of each fragment's transmission is verified by comparing its hash. Once enough fragments are collected, erasure coding is used to recover the original ciphertext C. Finally, C is decrypted using the session key obtained from the on-chain smart contract via an identifier to obtain the original data packet. And finally calculate its hash value. With on-chain storage Perform comparisons to complete end-to-end integrity closed-loop verification.

[0090] In the intelligent diagnostics and service module, the diagnostic engine and data integrity verification unit are the core drivers of the diagnostic process. When the engine receives a data entry with the motherboard serial number... Or test batch number Upon receiving the diagnostic request, the system first invokes a smart contract deployed on the blockchain. This contract maintains a public, immutable mapping state variable M, whose functionality can be formally represented as:

[0091]

[0092] in, This is a list of associated notarized transaction hashes. The contract's publicly available query function is invoked, returning all relevant transaction hashes. The engine then retrieves the complete notarized record from the blockchain node based on the hashes. The record contains the data fingerprint, digital signature, and index pointer hash. The engine uses the corresponding test terminal node's public key. Verify signature validity:

[0093]

[0094] After successful verification, the engine resolves the index pointer. Retrieves and reassembles complete structured raw data packets from the secure storage cluster. Finally, the engine calculates... hash value and stored on the chain A final comparison is performed to complete end-to-end integrity verification. Any verification failure will immediately terminate the process and trigger an alarm.

[0095] The multidimensional time-series data analysis model unit contains a pre-trained hybrid deep learning model for in-depth analysis of validated data. (Data package) The multi-dimensional time-series signal, parsed as timestamp-aligned, is first fed into a one-dimensional convolutional neural network (CNN). This network consists of multiple alternating convolutional and pooling layers, with convolutional kernels sliding along the time dimension to extract local anomalous patterns. The network can automatically detect local transient anomalies such as voltage glitches and signal overshoot. The high-level feature sequences extracted by the CNN are then fed into a long short-term memory (LSTM) network. LSTM units learn long-term dependencies through their gating mechanism. This network is used to identify cross-cycle anomalous patterns such as start-up delays and response timing errors.

[0096] The model's analytical capabilities are enhanced by the global context provided by the blockchain. The diagnostic engine queries the historical provenance of all relevant test items for this motherboard via smart contracts, or retrieves the characteristics of other motherboards in the same batch. It then calculates the characteristics of the current motherboard. Similarity or deviation from historical features or intra-batch feature distributions, such as calculating the Mahalanobis distance to determine whether anomalies are common to the batch:

[0097]

[0098] in, and These are the mean vector and covariance matrix of the characteristics of motherboards in the same batch, respectively. This distance is used to help determine whether the defect is a systematic batch problem or an individual random failure.

[0099] Finally, the model integrates the local anomaly features of CNN, the temporal dependency features of LSTM, and the results of contextual comparison analysis to output a structured diagnostic conclusion, including the specific anomaly time point, the possible faulty component or circuit module, and a confidence score indicating the credibility of the conclusion.

[0100] The human-computer interaction interface provides a graphical user interface. After the user enters query conditions, the interface calls the diagnostic engine and displays a diagnostic report. Each key conclusion in the report has a verification button next to it. When the user clicks it, the interface backend will automatically perform a lightweight on-chain verification: retrieve the original evidence-based transaction associated with the conclusion from the blockchain, re-verify the signature, and display the transaction hash and its block height, thereby directly binding the immutability of the blockchain to the final diagnostic report.

[0101] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. Additionally, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some embodiments, multitasking and parallel processing are possible or may be advantageous.

[0102] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

[0103] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A blockchain-based computer motherboard test data security diagnostic system, characterized in that, include: Data acquisition and encapsulation module: Captures the underlying analog electrical signals and digital instruction sequences generated during the motherboard testing process in real time at the physical testing station, and performs high-precision synchronization, time slicing, structural encapsulation and digital fingerprint generation on the underlying analog electrical signals and digital instruction sequences to obtain a structured raw data packet with a hierarchical nested structure; Blockchain Evidence Storage Network Module: Communicates and connects with the data acquisition and packaging module to form a blockchain network. It receives and verifies the structured raw data packets from the data acquisition and packaging module, and packages data fingerprints, key features, and digital signatures into immutable blocks in chronological order and links them into a chain through a consensus algorithm. Secure storage service cluster: Provides physical or logical storage space for persistent storage of complete original test data packets associated with the blockchain evidence storage network. It stores the structured original data packets generated by the data acquisition and encapsulation module. Its storage location information is recorded on the blockchain, and the on-chain fingerprint is associated with the complete original data off-chain. Intelligent Diagnosis and Service Module: Authorizes access to the evidence records on the blockchain and the complete data packets in secure storage. It performs integrity verification of on-chain and off-chain data, calls pre-trained intelligent analysis models to perform in-depth analysis of the data, and combines the context information associated on the blockchain to generate a diagnostic report with verifiable credibility.

2. The blockchain-based computer motherboard test data security diagnosis system of claim 1, wherein, The data acquisition and packaging module specifically includes: Signal monitoring and capture unit: Real-time monitoring of the test equipment's data bus and debugging interface via a programmable gate array chip, synchronously capturing the underlying analog electrical signal flow and digital instruction sequence during the test process; Timestamp and synchronization unit: Each captured low-level analog electrical signal stream and digital instruction sequence is stamped with a hardware timestamp with nanosecond-level precision and global monotonically increasing in real time. The timestamp is synchronized and aligned with the start and end instructions of the test item issued by the host computer of the test equipment. Data Slicing and Encapsulation Unit: The captured real-time data stream is sliced ​​according to the start and end time windows of the preset test items. For the data in each time slice, the corresponding high-precision timestamp, the unique identifier of the test item, and the hardware fingerprint of the acquisition device itself are fused to generate a structured raw data packet with a hierarchical nested structure. The structured raw data packet includes a header layer that stores metadata, a data body layer that stores the raw signal and instruction byte arrays that have been losslessly compressed and encoded, and a verification layer that stores the Merkle root hash generated based on the data body content.

3. The blockchain-based computer motherboard test data security diagnostic system of claim 1, wherein, The blockchain-based evidence storage network module specifically includes: Blockchain network unit: contains multiple physically distributed nodes, including test terminal nodes, quality inspection center nodes and audit nodes. The test terminal nodes are integrated with a consensus preprocessing unit, which is implemented in the form of a hardware security module or a trusted execution environment, and is used to securely store the node's asymmetric encrypted private key. The evidence storage transaction constructor unit receives the structured raw data packets from the data acquisition and encapsulation module, calculates the unique digital fingerprint of the data packets using a cryptographic hash function, digitally signs the fingerprint, key metadata, and Merkle root hash of the data packets using the protected node private key inside the HSM / TEE, and encapsulates them into a lightweight evidence storage transaction request. Distributed consensus engine: Based on the Byzantine fault-tolerant consensus algorithm, the quality inspection center node is preset as the leading node, and the audit node and test terminal node are used as verification nodes. Each consensus node votes in multiple rounds to reach a consensus on the legality, signature validity and order of the received evidence storage transaction requests. Chain-based ledger repository: After consensus, data is packaged into blocks in chronological order and linked to previous blocks through cryptographic hashes to form an immutable chain-based ledger, completing the on-chain anchoring of test data fingerprints and key features.

4. The blockchain-based computer motherboard test data security diagnostic system of claim 3, wherein, In the consensus preprocessing unit of the test terminal node, the hardware security module or trusted execution environment is further configured as follows: When calculating the cryptographic hash value, a collision-resistant cryptographic hash function is used to perform calculations on the entire byte sequence of the header layer, data body layer, and verification layer of the structured original data packet; When generating a lightweight evidence storage transaction request, the index pointer is constructed as a composite data structure containing a network address, storage path, and unique identifier of the data packet; The format of the evidence storage transaction request conforms to the standard transaction structure defined by the blockchain network, and its content fields include at least: transaction type identifier, sender node identifier, signed data fingerprint and Merkle root hash, key metadata, index pointer, and complete digital signature value.

5. The blockchain-based computer motherboard test data security diagnostic system of claim 3, wherein, The consensus mechanism of the Byzantine Fault Tolerant consensus algorithm specifically includes: The consensus process is divided into three main stages: pre-preparation, preparation, and submission. During the pre-preparation phase, the leading node sorts the received legitimate evidence storage transaction requests and packages them into a pre-preparation message, which is then broadcast to all verification nodes. During the preparation and submission phase, each verification node conducts multiple rounds of interactive voting on the message. The transaction block that receives the approval of more than two-thirds of the verification nodes is finally submitted and written to the local ledger of each node. The blockchain network also deploys a view replacement protocol. When a validator node does not receive a legitimate message from the dominant node within a preset time, it will trigger a view replacement process to elect a new dominant node according to a predetermined node order.

6. The blockchain-based computer motherboard test data security diagnostic system of claim 1, wherein, The security storage service cluster specifically includes: Encrypted storage node array unit: Composed of multiple physical storage servers. When a complete data packet is received, it uses a one-time session key dynamically obtained from the blockchain network to encrypt the data packet, divides the encrypted data packet into multiple data fragments, attaches erasure codes, and stores them on different physical nodes. On-chain indexing and verification interface unit: After the data packet is successfully stored, the interface will create a structured pointer containing its storage location, the key identifier required for decryption, and data fragment verification information. The hash value of the pointer is submitted to the blockchain network as a core element for evidence storage.

7. The blockchain-based computer motherboard test data security diagnostic system according to claim 1, characterized in that, The intelligent diagnosis and service module specifically includes: Diagnostic Engine: Responding to diagnostic requests, it queries and retrieves relevant on-chain evidence records by calling smart contracts on the blockchain evidence storage network, and locates and retrieves complete structured raw data packets from the secure storage service cluster based on the index pointers in the evidence records. Data integrity verification unit: Recalculates the cryptographic hash value of the retrieved structured original data packet and compares it with the corresponding data fingerprint obtained from the blockchain ledger. At the same time, it uses the public key of the test terminal node to verify the digital signature of the on-chain evidence. Multidimensional time-series data analysis model unit: The model adopts an architecture that integrates deep convolutional neural networks and long short-term memory networks. The original signal flow and instruction sequence in the verified data packet are input into the model to learn local abnormal patterns in the signal waveform and capture time-series dependencies and state transition anomalies across test phases. Based on the on-chain evidence feature data related to the motherboard's historical test items or test items of the same batch of motherboards, cross-comparison and context analysis are performed to output structured diagnostic conclusions, including anomaly localization, root cause inference and confidence score. Human-computer interaction interface: used to receive diagnostic requests, display diagnostic reports, allow users to trigger real-time hash and signature verification of the on-chain evidence data corresponding to key conclusions in the report, and display the verification results and related blockchain transaction hashes and block height information.

8. The blockchain-based computer motherboard test data security diagnostic system of claim 7, wherein, The logic of the smart contract on the blockchain evidence storage network includes: A publicly available mapping table data structure is used to store the association between the motherboard serial number or test batch number and a series of evidence-based transaction hashes; An access-controlled registration function is provided for authorized test terminal nodes to call after a successful on-chain notarization transaction, in order to update the mapping table and establish the association between the test object and the on-chain notarization. A public query function, which the diagnostic engine can call, takes a motherboard or batch identifier as input and returns a list of all associated evidence-based transaction hashes, serving as the entry point for data retrieval.

9. The blockchain-based computer motherboard test data security diagnostic system of claim 7, wherein, The operational mechanism of the multidimensional time series data analysis model specifically includes: The deep convolutional neural network layer is composed of multiple one-dimensional convolutional layers and pooling layers stacked alternately, which automatically extract local spatiotemporal features from the original signal stream; The long short-term memory network layer receives high-level feature sequences extracted by the convolutional layer, which are used to model long-distance temporal dependencies during the entire test item execution period and identify anomalies in the time dimension. The cross-comparison and context analysis specifically refer to the following: when analyzing the current test item data, the diagnostic engine queries the smart contract and blockchain ledger to obtain the on-chain evidence characteristics of other test items that have been completed in the previous test process on the same motherboard and have electrical or logical connections, or performs distribution similarity calculations on the test data characteristics of the current motherboard and the same test item characteristics of other motherboards in the same batch to determine whether it is a systemic batch problem or an individual defect.

Images