Blockchain-based information security contract signing system

Abstract

The application discloses a blockchain-based information security contract signing system and relates to the technical field of information security. The system first introduces a feature parameter extraction and finite element risk modeling method, extracts signature time deviation values and signature data trajectory similarity values, constructs a two-dimensional finite element analysis model, accurately calculates risk prediction values of random number leakage or reuse in the signature process, and combines the mathematical reversibility of attack samples to evaluate the possibility of private key derivation, thereby forming a comprehensive security level, realizing dynamic monitoring of digital signature security, risk early warning and on-chain signing control, effectively preventing contract forgery and system trust collapse caused by random number weaknesses, and significantly improving the security and credibility of the blockchain contract system.

Description

Technical Field

[0001] This invention relates to the field of information security technology, and more specifically to a blockchain-based information security contract signing system. Background Technology

[0002] Blockchain-based secure contract signing refers to utilizing the decentralized, immutable, and traceable characteristics of blockchain technology to digitally record contract content on the blockchain, ensuring data security and trustworthiness throughout the signing, storage, and execution of the contract. This approach prevents contracts from being forged or altered, enhances trust in contract performance among all parties, and effectively improves the transparency and efficiency of contract management.

[0003] The existing technology has the following shortcomings:

[0004] In the contract signing process based on the ECDSA signature mechanism, if the signing device reuses the same random number or uses a predictable pseudo-random number generation algorithm when signing multiple times, attackers can deduce the private key by obtaining multiple signature samples, thereby forging signatures and tampering with contract content, which seriously threatens the security and trustworthiness of the contract system. Summary of the Invention

[0005] The purpose of this invention is to provide a blockchain-based information security contract signing system to address the shortcomings of the prior art.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a blockchain-based information security contract signing system, including a hash processing module, a digital signature module, a feature parameter extraction module, a finite element risk prediction module, and a risk level determination module;

[0007] The hash processing module is used to perform hash operations on the contract content and generate a unique contract hash value.

[0008] The digital signature module is used to call the signer's private key to digitally sign the contract hash value and generate corresponding signature data;

[0009] The blockchain writing module is used to package the contract hash value and the corresponding signature data as transaction data and write them into the blockchain storage.

[0010] The feature parameter extraction module is used to extract signature-related feature parameters from multiple signature operations, including signature time deviation value and signature data trajectory similarity value;

[0011] The finite element risk prediction module is used to construct a finite element analysis model based on the signature-related characteristic parameters and calculate the predicted risk value of random number duplication or leakage during the signing process.

[0012] The risk level determination module is used to classify risk levels based on the risk prediction value, and, in combination with the mathematical reversibility of known attack samples, assess the possibility that an attacker can deduce the signer's private key, form a comprehensive security level, and trigger a security response mechanism.

[0013] Preferably, the hash processing module specifically includes: parsing the electronic contract content uploaded by the user, encoding and unifying the contract content, performing field extraction and serialization processing on the contract content, concatenating the fields in order to generate a standard input string, and calling the cryptographic hash function to perform a hash operation on the standard string to generate a unique contract hash value.

[0014] Preferably, the digital signature module specifically includes: verifying the signer's identity, including biometrics, certificate chain verification, or password verification; loading the signer's private key through a local key management system or hardware security module; performing a signature operation on the contract hash value using an elliptic curve signature algorithm; using a newly generated high-entropy random number k to ensure the uniqueness of each signature; and outputting a signature pair (r, s), along with a signature timestamp, public key identifier, and signature algorithm information encapsulated into a signature data structure.

[0015] Preferably, the feature parameter extraction module is used to extract the signature time deviation value during continuous signature process;

[0016] Given n consecutive signature operations, the corresponding timestamp sequence is as follows: Difference between adjacent timestamps yields a time interval sequence. ;get: ;

[0017] Divide all ΔT values ​​into several time intervals to form a set of intervals: ;

[0018] Count the number of times ΔT occurs, Z, in each interval, and calculate its probability distribution: ;in, Let m be the probability of the j-th time interval, and m be the total number of time interval distribution intervals.

[0019] Calculate information entropy using the Shannon entropy formula: ;in: The entropy value of the signature time deviation. To calculate the signature time deviation value based on the probability of occurrence for each time interval, the expression is as follows: ; For maximum entropy, This is the signature time deviation value, ranging from [0,1].

[0020] Preferably, the method for obtaining the similarity value of the signature data trajectory is as follows:

[0021] Given c consecutive signatures, each signature generates a value pair. This constitutes a set of trajectories: ;

[0022] Choose any two signatures Calculate their Euclidean distance: ; , The signature vector is the standardized form; distance is calculated for all signature pair combinations to obtain the distance set: ; Calculate the average distance The expression is: Based on the maximum distance Inverse normalization is performed to obtain the similarity value of the signature data trajectory. The expression is: ;in: ∈[0,1].

[0023] Preferably, the finite element model constructed by the finite element risk prediction module is a two-dimensional behavioral space model, including:

[0024] Map the signature time deviation value and the signature data trajectory similarity value to the two-dimensional plane coordinate axis;

[0025] Using coordinates as perturbation nodes, the perturbation propagation behavior is simulated in a constructed two-dimensional mesh;

[0026] Integrating the grid response yields a global stability function.

[0027] Normalize the integral results to obtain the risk prediction value. .

[0028] Preferably, the risk prediction value output by the finite element analysis model Classify, if If the value is in the range of 0 to 0.2, it is considered secure, indicating that the signature behavior is distributed naturally and has sufficient randomness; if... A value between 0.2 and 0.5 is considered a weak warning, indicating an abnormal trend in a certain behavioral dimension but not yet posing a serious threat; when... If the value falls within the range of 0.5 to 0.8, it is considered high-risk, indicating a pattern or repetitive characteristic in the signature process; if If the value exceeds 0.8, it is classified as a critical anomaly, and there is a high suspicion that the private key or random number has been reused, leaked, or attacked during the signing process. Security response measures will be automatically taken.

[0029] Preferably, in the risk level determination module, the attack reversibility factor is obtained. The method of obtaining the signature is as follows: Suppose the attacker obtains two signature samples: Signature pair 1: Corresponding hash h1; Signature pair 2: Corresponding hash h2; construct a reversibility sub-index, including: Sub-index A: r-value repeatability factor ; Sub-index B: Time interval factor Define the time interval as ΔT = T2 - T1, and normalize it as follows: ;in The maximum secure signature interval is set; Sub-index C: s-value similarity factor. ; ;in The order of the elliptic curve or the empirical maximum value;

[0030] The attack reversibility factor is obtained by weighting and summing the sub-indices A, B, and C. .

[0031] Preferred comprehensive evaluation indicators It can be represented as: ;in, and For weighting coefficients, according to The value of is used to classify and judge the overall signature security; if If the value is greater than or equal to a preset threshold ST, the signing action will be judged as having a high risk of private key leakage, marked as high-risk, and the account will be immediately locked or subsequent signing processes will be blocked; if... If the risk level is less than the preset threshold ST, the system is considered to be in good security condition, marked as low risk level, and recorded only for future reference.

[0032] The technical effects and advantages provided by the present invention in the above technical solution are as follows:

[0033] 1. This invention innovatively solves the problem in existing ECDSA-based signature mechanisms where private keys can be deduced by attackers due to random number reuse or low-quality random number generation by introducing signature behavior feature analysis and a finite element risk prediction model. By extracting signature time deviation values ​​and signature data trajectory similarity values, the system can dynamically perceive abnormal behavior patterns in the signature process. Furthermore, it utilizes two-dimensional finite element perturbation simulation to construct a behavior stability evaluation mechanism, calculating the predicted risk value of random number usage during the signature process, thus achieving early identification and tiered management of security risks.

[0034] 2. This invention combines cryptographic reversibility analysis with the design of an attack reversibility factor. By constructing multiple sub-indicators, it quantifies the repeatability, temporal tightness, and data similarity of signature samples, and integrates these with the finite element analysis output to form a comprehensive security level. When the comprehensive security level exceeds a preset threshold, the system can automatically trigger security response mechanisms such as account locking and signature blocking, effectively preventing high-risk events such as signature forgery and contract tampering, and significantly improving the credibility, anti-attack capability, and practicality of the blockchain contract system. Attached Figure Description

[0035] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0036] Figure 1 This is a mind map of the system modules of the present invention. Detailed Implementation

[0037] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0038] For examples, please refer to Figure 1 As shown, the blockchain-based information security contract signing system described in this embodiment includes a hash processing module, a digital signature module, a feature parameter extraction module, a finite element risk prediction module, and a risk level determination module.

[0039] The hash processing module is used to perform hash operations on the contract content and generate a unique contract hash value.

[0040] The digital signature module is used to call the signer's private key to digitally sign the contract hash value and generate corresponding signature data;

[0041] The blockchain writing module is used to package the contract hash value and the corresponding signature data as transaction data and write them into the blockchain storage.

[0042] The feature parameter extraction module is used to extract signature-related feature parameters from multiple signature operations, including signature time deviation value and signature data trajectory similarity value;

[0043] The finite element risk prediction module is used to construct a finite element analysis model based on the signature-related characteristic parameters and calculate the predicted risk value of random number duplication or leakage during the signing process.

[0044] The risk level determination module is used to classify risk levels based on the risk prediction value, and, in combination with the mathematical reversibility of known attack samples, assess the possibility that an attacker can deduce the signer's private key to form a comprehensive security level.

[0045] The intelligent response control module is used to trigger a security response mechanism when the overall security level reaches a preset risk threshold.

[0046] In this embodiment, a hash processing module is provided for use in a blockchain-based information security contract signing system. This module is used to preprocess and hash the content of the electronic contract to be signed, generating a unique contract hash value, which is used as the basis for subsequent digital signatures and blockchain records.

[0047] Users upload an electronic contract through the front-end interface. The system supports contract formats including .pdf, .docx, .json, and .txt. Taking a .pdf contract as an example, the system automatically calls a file parsing tool to extract the PDF content into plain text and recognize fields such as: title, number, body text, and signatory information.

[0048] The extracted text content is uniformly converted to UTF-8 encoding format, and interference items such as BOM headers, hidden characters, and control characters are removed. The system further standardizes the text paragraphs, replacing all paragraph marks with a uniform line break format and compressing redundant spaces to ensure consistent text content standards.

[0049] The system extracts the structured information from the contract into fields and constructs the input string to be hashed in the following order: contract title + contract number + contract text + signatory ID + current timestamp;

[0050] For structured contracts, such as JSON format electronic contracts, the system uses Canonical JSON encoding rules to serialize fields in lexicographical order to prevent differences in hash results caused by changes in field order.

[0051] The system calls the SHA-256 hash algorithm in the encryption library, reads the serialized contract input string, and performs a hash operation: contract_hash = SHA256(serialized_contract_text); the hash value is a 64-bit hexadecimal string, which serves as a unique identifier for the contract content.

[0052] The generated contract_hash value is verified locally once (to confirm length, character set, etc.), and then passed to the signing module through an internal system interface for digital signing using the signer's private key. This hash value is also included as part of the contract digest and is packaged and uploaded to the blockchain along with the signing result.

[0053] Through this embodiment, the system can generate highly consistent and secure hash values ​​when processing contract texts from different sources and in different formats, ensuring that subsequent digital signatures and blockchain evidence storage are based on the same content, thus significantly enhancing the legality and non-repudiation of the entire contract signing process.

[0054] The digital signature module is used to receive the contract hash value, call the signer's private key to perform digital signature operation, generate legally valid signature data (r,s) or an equivalent signature pair, and submit the signature data as a signing certificate to the subsequent processing module.

[0055] Before signing, the system verifies the identity of the signer, including: private key ownership authentication (such as fingerprint, facial recognition, password verification); digital certificate validity check (such as X.509 format certificate chain verification); after successful verification, the private key signing engine can be called.

[0056] The system securely loads the private key of the corresponding signer from the local key management system (KMS) or hardware security module (HSM); at the same time, it calls the signature algorithm parameters, such as: elliptic curve name (e.g., secp256k1) and signature scheme (e.g., ECDSA, EdDSA, or RSA-PSS).

[0057] The system uses the loaded private key to perform an encrypted signature operation on the contract hash value to generate a signature value; the system ensures that a new, sufficiently entropy random number k is used in each signing process to prevent reuse.

[0058] The generated signature result and related auxiliary data are encapsulated into a standard signature data structure, including but not limited to: the original contract hash value; the signature value (r, s) or signature string; the signer's public key or address identifier; the signature timestamp; and the algorithm identifier.

[0059] This module enables the system to generate unforgeable and verifiable signature data while ensuring the secure isolation of the signer's identity and private key. It also provides a strong cryptographic foundation for blockchain notarization, effectively preventing attacks such as signature replay, forgery, and random number reuse.

[0060] The blockchain writing module is used to encapsulate the generated contract hash value and its corresponding signature data into transaction data according to a preset format, send it to the blockchain network, and complete the writing and storage through a consensus mechanism, so as to realize the immutable and traceable on-chain evidence storage of the contract signing process.

[0061] Receive structured data from the signature module, including the following data fields:

[0062] Contract hash (contract_hash);

[0063] Signature values ​​(such as r and s in ECDSA);

[0064] Signatory identification (e.g., address, public key, certificate number);

[0065] Auxiliary fields such as timestamp and signature algorithm identifier;

[0066] Construct the on-chain transaction payload.

[0067] To reduce on-chain storage costs, data compression or encoding is performed, such as: using Base58 or RLP encoding for hexadecimal data; removing redundant fields and using hash indexes instead of the original descriptions; and converting encoded transaction data into an on-chain acceptable format, such as JSON, CBOR, or Solidity ABI encoding.

[0068] Build a transaction structure and attach the necessary on-chain metadata (such as gas, nonce, chain ID).

[0069] The transaction is signed using the system's private key (not the contract signer) to prevent tampering; the transaction is broadcast to the network via a blockchain node interface (such as Ethereum's JSON-RPC, Hyperledger Fabric SDK, Polkadot Substrate API).

[0070] Network nodes verify the legality of transactions using a pre-defined consensus algorithm (such as PoW, PoS, PBFT); transactions are packaged into new blocks and form an immutable record on the chain; the system monitors the transaction hash and block height to confirm the completion of the transaction writing.

[0071] After a transaction is confirmed, the system automatically records: transaction hash (tx_hash); block number (block_number); block timestamp (onchain_timestamp); and simultaneously creates a local index (such as contract ID → block address) to facilitate subsequent retrieval and verification.

[0072] The feature parameter extraction module is used to extract key behavioral feature parameters related to security from each signature data during the process of the system performing multiple digital signature operations. The feature parameters include, but are not limited to, signature time deviation value and signature data trajectory similarity value, for subsequent risk prediction model analysis.

[0073] After each digital signature is completed, the system stores the following data in the signature log: signing timestamp (T); signature value (r, s); signer identifier; contract hash value; the data is organized in a time series manner to form a complete set of signature behavior records.

[0074] The method for obtaining the signature time deviation value is as follows:

[0075] Given n consecutive signature operations, the corresponding timestamp sequence is as follows: Difference between adjacent timestamps yields a time interval sequence. ;get: ;

[0076] Divide all ΔT values ​​into several time intervals (interval width can be customized), for example, 1 second as one interval, forming a set of intervals: ;

[0077] Count the number of times ΔT occurs, Z, in each interval, and calculate its probability distribution: ;in, Let m be the probability of the j-th time interval, and m be the total number of time interval distribution intervals.

[0078] Calculate information entropy using the Shannon entropy formula: ;in: The entropy value of the signature time deviation. To calculate the signature time deviation value based on the probability of occurrence for each time interval, the expression is as follows: ; For maximum entropy, This is the signature time deviation value, ranging from [0,1]. The larger the value, the more abnormal the deviation.

[0079] The method for obtaining the similarity value of signature data trajectory is as follows:

[0080] Given c consecutive signatures, each signature generates a value pair. This constitutes a set of trajectories: ;

[0081] Choose any two signatures Calculate their Euclidean distance: ; , The signature vector is the standardized form; distance is calculated for all signature pair combinations to obtain the distance set: ; Calculate the average distance The expression is: Based on the maximum possible distance The signature data trajectory similarity value is obtained by inverse normalization of the maximum value under normalization. The expression is: ;in: ∈[0,1], the closer to 1, the more similar the signature trajectories (the higher the potential risk).

[0082] The finite element risk prediction module is used to construct a finite element analysis model for simulating behavioral stability based on key behavioral characteristic parameters in the signature process. It models and simulates possible random number reuse, weak random numbers, or signature forgery behaviors, and outputs the corresponding risk prediction values.

[0083] The module receives the following key parameters provided by the feature extraction module:

[0084] The signature time deviation value and signature trajectory similarity value are both normalized to the [0,1] interval and used as input variables.

[0085] Model the behavior space as a two-dimensional finite element region:

[0086] The horizontal axis represents the time deviation dimension (representing the stability of the signature rhythm);

[0087] The vertical axis represents the trajectory similarity dimension (representing the regularity of signature data);

[0088] A two-dimensional mesh consisting of multiple units and nodes is constructed within this region to simulate the stable behavior corresponding to different combinations of features.

[0089] The received signature time deviation value and signature data trajectory similarity value are used as the key nodes to apply perturbation sources;

[0090] The stability / instability of simulated behavioral offset propagation in the mesh;

[0091] By integrating the response functions (such as strain and deformation) within the model elements, a global stability index is obtained.

[0092] Convert finite element simulation results into numerical scoring This indicates that random numbers use secure stability: ;in: The higher the value, the lower the stability of the signature process, and the greater the risk of duplication / leakage; the model supports precision adjustment and can train different mapping functions based on empirical samples.

[0093] Risk prediction values ​​output by the finite element analysis model Classify, if A value between 0.0 and 0.2 is considered secure, indicating a natural distribution of signature activity and sufficient randomness; if... A value between 0.2 and 0.5 is considered a weak warning, indicating that an abnormal trend has emerged in a certain behavioral dimension (such as signature interval or signature result) but has not yet posed a serious threat; when If the value falls within the range of 0.5 to 0.8, the system identifies it as high-risk, indicating that the signature process exhibits significant regularity or repetitive characteristics, suggesting the use of insecure random numbers; if If the value exceeds 0.8, it is classified as a critical anomaly, and it is highly suspected that the private key or random number has been reused, leaked, or attacked during the signing process. The system will automatically take security response measures such as blocking the signing and locking the account.

[0094] Risk prediction values ​​output by the finite element analysis model Furthermore, the system combines the mathematical reversibility characteristics of attack samples to jointly assess the possibility of attackers deducing the signer's private key, thereby forming a more accurate comprehensive security level.

[0095] In the specific implementation, the first step is to calculate... This reflects the stability and abnormality of the signature behavior; subsequently, an attack reversibility factor is introduced. This factor scores based on cryptographic reversibility conditions such as whether existing signature samples contain duplicate random numbers k, whether signature values ​​r are consistent, and whether the time interval is extremely short. If the system finds the same r value or extremely close (r,s) pairs in two signatures, and the signature time intervals are abnormally concentrated, a higher score is calculated. This means that an attacker could use this formula to deduce the private key.

[0096] Attack reversibility factor The aim is to quantify the likelihood that an attacker can deduce the private key from the signature data. High reversibility ( →1) → This indicates that the signature behavior is highly reversible; reversibility is low ( →0)→ This indicates that the signature process is secure and irreversible.

[0097] The acquisition method is as follows: Suppose the attacker obtains two signature samples (ECDSA as an example):

[0098] Signature pair 1: Corresponding hash h1;

[0099] Signature pair 2: Corresponding hash h2;

[0100] Extract and analyze features from them:

[0101] Do two identical r values ​​exist? If so... = The message indicates that the random number k is repeated.

[0102] Does an extremely small time interval exist (supporting attack inference): for example, a time difference of <1 second;

[0103] Does an approximate s-value or linear relationship exist? It can be determined by linearly fitting s~h.

[0104] Construct reversibility sub-indices, including:

[0105] Sub-index A: r-value repeatability factor ; ;

[0106] Sub-index B: Time interval factor ;

[0107] Define the time interval as ΔT = T2 - T1, and normalize it as follows: ;in The maximum secure signature interval is set (e.g., 10 seconds).

[0108] Sub-index C: s-value similarity factor ; ;in The order is the elliptic curve order or the empirical maximum value to ensure normalization;

[0109] The attack reversibility factor is obtained by weighting and summing the sub-indices A, B, and C. .

[0110] Comprehensive evaluation indicators It can be represented as: ;in, and The weighting coefficient reflects the system's emphasis on behavioral risk versus mathematically reversible risk.

[0111] according to The value of is used to classify and assess the overall signature security. If If the value is greater than or equal to a preset threshold ST, the signing action will be judged as having a high risk of private key leakage, marked as high-risk, and the account will be immediately locked or subsequent signing processes will be blocked; if... If the risk level is below a preset threshold ST, the system is considered to be in good security condition, marked as low-risk, and recorded only for future reference. This method combines behavioral patterns with cryptographic principles, achieving more refined security control for contract signing.

[0112] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0113] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

Claims

1. A blockchain-based information security contract signing system, characterized in that: The system comprises a hash processing module, a digital signature module, a feature parameter extraction module, a finite element risk prediction module, and a risk level determination module. The hash processing module is configured to perform hash operation on the contract content to generate a unique contract hash value. The digital signature module is configured to perform digital signature on the contract hash value by calling a private key of a signing party to generate corresponding signature data, specifically including: verifying the identity of the signing party, including biometric identification, certificate chain verification, or password verification; loading the private key of the signing party through a local key management system or a hardware security module; performing signature operation on the contract hash value using an elliptic curve signature algorithm; using a newly generated high-entropy random number k to ensure the uniqueness of each signature; outputting the signature pair (r, s) and attaching the signature timestamp, public key identifier, and signature algorithm information to encapsulate as a signature data structure; The blockchain writing module is configured to pack the contract hash value and the corresponding signature data as transaction data and write them into a blockchain storage. The feature parameter extraction module is configured to extract signature-related feature parameters, including signature time deviation value and signature data trajectory similarity value, in multiple signature operations. The signature data track similarity value is obtained by the following method: assuming that there are continuous c times of signature data, a value pair is generated for each signature , to form a track set ; the Euclidean distance of two optional signatures is calculated ; the distance set is obtained by calculating the distance of all signature pairs ; the average distance is calculated ; the maximum distance is inversely normalized to obtain the signature data track similarity value ; The finite element risk prediction module is configured to construct a finite element analysis model based on the signature-related feature parameters, and calculate a risk prediction value of random number repetition or leakage in the signature process. The risk level determination module is configured to divide the risk level according to the risk prediction value, and evaluate the possibility of deducing the private key of the signing party by the attacker based on the mathematical reversibility of the known attack samples, form a comprehensive security level, and trigger a security response mechanism. 2.The blockchain-based information security contract signing system of claim 1, wherein: The hash processing module specifically includes: parsing the electronic contract content uploaded by the user, uniformly processing the contract content, performing field extraction and serialization processing on the contract content, and splicing a standard input string according to the field order; calling an encryption hash function to perform hash operation on the standard string to generate a unique contract hash value. 3.The blockchain-based information security contracting system of claim 1, wherein: The feature parameter extraction module is configured to extract the signature time deviation value in continuous signature process. There are n continuous signature operations, and the corresponding timestamp sequence is: ; The adjacent difference of the time stamp is made, and the time interval sequence is obtained: ; the following is obtained: ; Divide all the ΔT values into several time intervals to form an interval set: ; The number of occurrences Z of ΔT in each interval is counted, and the probability distribution is calculated: ; wherein is the probability of the jth time interval interval, and m is the total number of time interval distribution intervals. The information entropy is calculated according to a Shannon entropy formula: ; wherein: is an entropy value of the signature time deviation, is a probability of occurrence of each time interval, the signature time deviation value is calculated, and the expression is: ; is a maximum entropy, is the signature time deviation value, and the range is [0, 1]. 4.The blockchain-based information security contract signing system of claim 3, wherein: The signature data trajectory similarity value is obtained by: The finite element model constructed by the finite element risk prediction module is a two-dimensional space model, including: There are continuous c times signature data, each time signature generates a value pair , constitutes a track set: ; Optionally twice signed , calculate its Euclidean distance: ; , The normalized signature vector; the distance calculation is performed for all signature pairs, and the distance set is obtained: ; Calculate the average distance , the expression is: ; According to the maximum distance , the signature data track similarity value is obtained by reverse normalization , the expression is: ; Where: ∈[0,1]. 5.The blockchain-based information security contracting system of claim 4, wherein: Map the signature time deviation value and the signature data trajectory similarity value to the two-dimensional plane coordinate axes; Take the coordinates as perturbation nodes to simulate perturbation propagation behavior in the constructed two-dimensional grid; Integrate the grid response quantity to form a global stability function; ​ The integral result is normalized to obtain a risk prediction value . 6.The blockchain-based information security contracting system of claim 5, wherein: Risk prediction value output by the finite element analysis model Classification, if 0 to 0.2 interval, determined to be safe, indicating that the signature behavior distribution is natural, and the randomness is sufficient; if 0.2 to 0.5, determined to be a weak warning, indicating that an abnormal trend has appeared in a certain behavior dimension, but has not yet constituted a serious threat; when 0.5 to 0.8, identified as high risk, indicating that the signature process has regularity or repetition characteristics; if More than 0.8, classified as a critical anomaly, highly suspicious that the private key or random number in the signature process has been reused, leaked or attacked, and security response measures will be automatically taken. 7.The blockchain-based information security contracting system of claim 6, wherein: In the risk level determination module, an attack reversibility factor is obtained , and the obtaining method is as follows: assuming that an attacker obtains two signature samples: signature pair 1: corresponding to hash h1; signature pair 2: corresponding to hash h2; a reversibility sub-index is constructed, including: sub-index A: r value repetition factor ; ; sub-index B: time interval factor ; defining the time interval as ΔT = T2 - T1, and normalizing as: ; wherein is a set maximum secure signature interval; sub-index C: s value similarity factor ; ; wherein is an elliptic curve order or an empirical maximum value; The attack reversibility factor is obtained by weighted summation calculation of sub-indexes A, B and C . 8.The blockchain-based information security contracting system of claim 7, wherein: Comprehensive evaluation index Can be expressed as: ; wherein, And Is a weight coefficient, according to The value of the overall signature security is classified; if Greater than or equal to the preset threshold ST, it is judged that there is a high risk of private key leakage in the signature behavior, marked as high risk level, immediately lock the account or block the subsequent signing process; if Less than the preset threshold ST, it is considered that the system security state is good, marked as low risk level, only record for reference.

Images