A cargo rights fraud real-time blocking system based on physical event anchoring and space-time decoupling

Abstract

The application discloses a kind of based on physical event anchoring and space-time decoupling's sea transport freight right fraud real-time blocking system, comprising: physical event four-dimensional tensor construction module: for building the physical event including space coordinates, time series, equipment signature and event type four-dimensional data tensor representation;Quantum fingerprint anchoring module: for generating anti-copy unique identification to physical event data by 12-bit quantum circuit;Helve space-time topology engine: for calculating the continuity score of the physical event of generating anti-copy unique identification;Three-level risk response execution module: for automatically triggering on-chain blocking operation according to physical event continuity score, three-level risk response is judged;Self-evolution risk control system: for dynamically optimizing rule weight and freight right entropy model by fraud mode clustering, generates the defense mechanism of triggering on-chain blocking operation.The system marks that international trade trust mechanism has entered the trend and possibility of new era of quantum level credibility.

Description

Technical Field

[0001] This invention belongs to the field of international trade security technology and relates to a real-time anti-fraud system for goods ownership based on physical event anchoring and spatiotemporal decoupling. Background Technology

[0002] In the international trade system, the bill of lading, as a document of title to goods, is fundamentally separated from actual control of the goods. This structural defect has led to the long-standing existence of two types of prevalent fraud:

[0003] (1) Fraud involving multiple sales of the same goods: Multiple bills of lading are maliciously issued for the same batch of goods during transportation. Fraudsters take advantage of the time delay in the circulation of paper bills of lading to sell the same batch of goods to multiple buyers, causing significant economic losses to the buyers. The essence of this type of fraud is the disconnect between digital claims of title and physical control of the goods.

[0004] (2) Empty Bill of Lading Fraud: Forging false bills of lading without corresponding physical goods to apply for financing and pledge with banks. Due to the lack of a real-time verification mechanism for the physical state of the goods, banks have difficulty identifying the authenticity of the bills of lading, resulting in huge financial losses. The root cause of this type of fraud lies in the fact that the document verification system cannot anchor to the real state of the physical world.

[0005] Existing technical solutions suffer from systemic deficiencies, creating fatal vulnerabilities in security defenses, and exhibiting the triple flaws of traditional anti-fraud systems. (1) Information black box dilemma: Relying on paper documents or simple electronic systems, it is impossible to obtain key physical event data such as container unlocking status and actual ship location in real time. For example, forged bills of lading can be used to defraud banks of financing through formal review because traditional systems cannot verify whether the container actually exists or has been opened.

[0006] (2) Response delay bottleneck: Manual verification requires coordination among multiple parties, including ship owners, ports, customs, and banks, and takes an average of 24-72 hours. The fraudulent act is only discovered after it has been completed, and the transfer of funds has already become a fait accompli, resulting in a recovery success rate of less than 15%.

[0007] (3) Data silos: Ship owners’ AIS data, port RFID records, and banks’ letter of credit systems are isolated from each other, forming data silos. For example, warehouse entry records and bill of lading flow data cannot be compared in real time, making it difficult to identify empty bill of lading fraud such as “the warehouse detected a container but there was no corresponding bill of lading”.

[0008] Current industry solutions have fundamental flaws: (1) Electronic bill of lading systems (such as Bolero and essDOCS): Although they improve the efficiency of document circulation, they do not solve the problem of real-time decoupling between the physical world and digital title of goods. The system cannot detect physical anomalies such as illegal opening of containers or deviation of ship routes, and still remains at the level of electronic document digitization.

[0009] (2) Blockchain-based evidence storage solutions (such as TradeLens and IBM Blockchain): These solutions only enable on-chain evidence storage of individual documents and lack a verifiable anchoring mechanism for physical events. For example, they cannot prove the authenticity and spatiotemporal consistency of customs inspection photos, resulting in a vulnerability where "the digital record is real but the physical state is false."

[0010] (3) Fragmented IoT monitoring: Independently deployed smart locks and AIS devices generate unstructured data that is not dynamically linked to cargo ownership change events. The system cannot establish a causal chain of "bill of lading transfer → container operation → ship movement", resulting in key risk patterns such as "the container being illegally opened after the bill of lading transfer" not being identified. Summary of the Invention

[0011] To address the aforementioned issues, the technical solution adopted in this invention is: a real-time blocking system for maritime cargo ownership fraud based on physical event anchoring and spatiotemporal decoupling, comprising: a physical event four-dimensional tensor construction module: used to construct physical events based on the three-axis acceleration waveform of container smart locks, the latitude and longitude coordinate sequence of ship AIS, the digital signature image of customs inspection equipment, and the geofence timestamp of warehouse RFID, to construct physical events containing a four-dimensional data tensor representation of spatial coordinates, time series, equipment signature, and event type; Quantum fingerprint anchoring module: Used to generate a unique, copy-resistant identifier for physical event data generated by the physical event four-dimensional tensor construction module through a 12-bit quantum circuit, and anchor it to the blockchain; Hurwitz Spacetime Topology Engine: Used to calculate the continuity score for generating unique anti-copying identifiers for physical events generated by the physical event four-dimensional tensor construction module and anti-counterfeiting identifiers provided by the quantum fingerprint anchoring module; Zero-knowledge cross-validation module: used to verify the physical events generated by the physical event four-dimensional tensor construction module, and to achieve privacy-preserving trusted verification of physical events using a zero-knowledge proof method; Dynamic decoupling rule engine: Based on the physical events generated by the physical event four-dimensional tensor construction module and the acquired bill of lading flow, as well as the trusted verification transmitted by the zero-knowledge cross-validation module, the bill of lading flow and logistics events are compared in real time through the built-in fraud rule library, and fraud patterns are identified based on preset rules. Level 3 Risk Response Execution Module: Based on the continuous scores transmitted by the Herwitz spatiotemporal topology engine and the fraud patterns transmitted by the dynamic decoupling rule engine, it automatically triggers on-chain blocking operations to make judgments on Level 3 risks. Self-evolving risk control system: Using the continuous scores transmitted based on the Hurwitz space-time topology engine, dynamically optimize the rule weights and the cargo right entropy value model through fraud pattern clustering, and generate a defense mechanism for triggering on-chain blocking operations.

[0012] Furthermore, the dynamic decoupling rule engine includes four core rules: space-time conflict detection, path deviation detection, multi-party statement conflict detection, and physical-document disconnection detection, which are specifically as follows: Space-time conflict detection: When the bill of lading transfer timestamp tb and the container opening timestamp to satisfy 0 < to - tb < 1800 seconds, trigger a warning of cargo right theft; Path deviation detection: Real-time calculate the Haversine distance between the ship's position and the declared route , when > 200 nautical miles and lasts for more than 1 hour, trigger the blocking of ghost ships; Multi-party statement conflict detection: Retrieve the ownership statement records of the same container on the blockchain. When there are multiple valid bills of lading, generate a judicial evidence package for selling the same cargo to multiple parties; Physical-document disconnection detection: When the warehouse RFID entry timestamp has no corresponding bill of lading data on the blockchain, mark it as a high-risk event of empty bill of lading pledge; The determination of the dynamic optimization algorithm for the weights of the four rules of space-time conflict detection, path deviation detection, multi-party statement conflict detection, and physical-document disconnection detection is as follows:

[0013]

[0014] Where: FP is the false positive rate of the rule, is the learning rate, represents the current weight value of the i-th rule, represents the weight value of the i-th rule after update, which is the result of dynamic optimization.

[0015] Furthermore, the process of constructing a physical event characterized by a four-dimensional data tensor including spatial coordinates, time series, device signatures, and event types based on the three-axis acceleration waveform of the container intelligent lock, the longitude and latitude coordinate sequence of the ship AIS, the digital signature image of the customs inspection equipment, and the geographical fence timestamp of the warehouse RFID is as follows: S11: Perform joint time-frequency analysis on the container intelligent lock waveform: In the time domain, use acceleration amplitude mutation detection. When the waveform peak value > 1.5g, mark a violent unlocking event; In the frequency domain, use fast Fourier transform to extract the main frequency component. When the main frequency deviation rate > 15%, mark illegal interference; S12: The abnormal detection of ship trajectories adopts the Haversine distance model: 1.

[0016] Where: is the latitude, is the longitude, R is the radius of the earth. When three consecutive points > 200 nautical miles, a route deviation warning is triggered; is the spherical distance between two geographical locations, R is the radius of the earth, is the latitude difference between two points, is the longitude difference between two points, represents the latitude value of the starting point, represents the latitude value of the ending point..

[0017] S13: Establish a spatio-temporal consistency verification mechanism for the customs inspection timestamp tc and the smart lock operation timestamp tl: It is determined that the verification is passed when |tc - tl| < n seconds and the container numbers are the same.

[0018] Furthermore, the quantum fingerprint anchoring module includes: Quantum fingerprint generation unit: Used to calculate a unique anti-copy quantum fingerprint identifier for each original physical event; Multi-source spatio-temporal alignment unit: Used to receive the event data with fingerprint identifiers transmitted by the quantum fingerprint generation unit, verify whether it belongs to the same coherent physical process by checking whether the timestamp deviation of multiple related events is within the threshold, and output an associated event group that passes the spatio-temporal consistency verification; Blockchain evidence storage unit: Used to construct a causal chain with a Merkle root based on the associated event group transmitted by the multi-source spatio-temporal alignment unit in chronological order, and package it together with the previous block hash, quantum fingerprint and digital signature to form an immutable permanent blockchain evidence storage; The quantum fingerprint generation unit: Performs the following operations through a quantum circuit: S21: After hashing the original event data, intercept it as a 48-bit classical binary string and output it as an initialization parameter; S22: Based on the binary string, apply the X gate to the corresponding quantum bits to encode the classical information into a definite initial quantum state; S23: Apply the Hadamard gate to the initial quantum state to convert the initial quantum state after applying the Hadamard gate into a uniform superposition state, and output a quantum superposition state containing all possibilities; S24: Apply the CX entanglement gate to the quantum superposition state to make the qubits undergo quantum entanglement and output a highly correlated entangled state; S25: Finally, the entangled state is measured multiple times to collapse the quantum information into a classical probability distribution, and the classical bit sequence with the highest probability is output as the final fingerprint identifier. The execution process of the blockchain evidence storage unit is as follows: The complete block data to be signed is composed of the event causal chain, the preceding hash, and the quantum fingerprint. Then, a digital signature overlaying this data is generated using a private key.

[0019] Furthermore, for the physical events generated based on the physical event four-dimensional tensor construction module, a zero-knowledge proof method is used to achieve trusted verification of physical events under privacy protection; S51: Generate the customs inspection certificate. The specific process is as follows: S511: Obtain inspection photos of the container; S512: Extract container number hash value and geofence location stamp based on container inspection photos; S513: Construct a zk-SNARK circuit verification declaration based on hash value and geofence location stamp: S52: Bill of Lading Premature Transaction Detection: When the timestamp of transfer of title to and the timestamp of container operation tc satisfy |to -tc|>1800 seconds, a fraud evidence vector is generated.

[0020] Furthermore, the Hurwitz spacetime topology engine performs the following process for calculating a continuity score to generate a unique, copy-resistant identifier for physical events generated by the physical event four-dimensional tensor construction module and the anti-counterfeiting identifier provided by the quantum fingerprint anchoring module: 5.1 Relativistic Time Dilation Correction: Introducing the Speed ​​of Light Calculate the time dilation factor corresponding to the velocity v:

[0021] Corrected time difference ; 5.2 Continuity break determination: The score is calculated by the mathematical formula of the Hurwitz spatiotemporal topology model. When the score value is <0.8, a three-dimensional alarm vector containing the break location, time deviation and velocity factor is generated.

[0022] Furthermore, the three-tiered risk classification includes a three-tiered response, a two-tiered response, and a one-tiered response, as detailed below: Level 3 response: Continuity score < 0.5. When a response is classified as Level 3, perform the following actions: Freezing the flow of funds in relevant bank accounts or suspending the transfer of bills of lading on the blockchain network; or generating a judicial evidence package that complies with the Hague-Visby Rules. The generated judicial evidence package conforming to the Hague-Visby Rules includes: a blockchain transaction ID Merkle path, a timestamped event sequence causal chain, and an ECDSA signature verification tree; Level 2 response: 0.5 ≤ continuous score < 0.8; When judged as a Level 2 response, perform the following operations: restrict bill of lading transfer operation permissions or initiate a multi-party video verification mechanism or generate a risk warning evidence chain; Level 1 Response: Consecutive scores ≥ 0.8: When a response is determined to be Level 1, perform the following actions: Send AES-256 encryption risk warnings to regulators and update historical pattern records in the risk knowledge base.

[0023] Furthermore, the self-evolving risk control system uses continuous scores transmitted based on the Hurwitz spatiotemporal topology engine to dynamically optimize rule weights and cargo entropy models through fraud pattern clustering, generating a defense mechanism that triggers on-chain blocking operations. The process is as follows: S71: Calculate the entropy value of cargo ownership based on the dynamic model of cargo ownership entropy value; S72: Based on the entropy value of goods ownership, a fraud pattern clustering engine is adopted, and the historical event feature vector is used for cluster analysis to discover fraud patterns and generate defense rules. The feature vector includes continuous scores, entropy values ​​of goods ownership, and rule weights. S73: Utilize the rule base evolution mechanism to optimize the weights and expand the rules of the rule base, thereby completing the system evolution.

[0024] The fraud pattern clustering engine is implemented as follows: (1) Extract the historical event feature vector X = [continuity score, cargo entropy value, rule weight]; (2) Identify new fraud patterns using the DBSCAN clustering algorithm; (3) When ghost container fraud is detected, generate defense rules; The expression for the dynamic model of cargo ownership entropy is as follows:

[0025]

[0026] Where: pk is the probability of the cargo's safe state. is the standard deviation of the continuous score; wi is the event type weight; ft is the time decay factor; si is the spatial dispersion adjustment coefficient. The rule base evolution mechanism automatically performs weight optimization and rule expansion after processing every 100 events.

[0027] Furthermore, it also includes a thought-decision output module: Connect to the Hurwitz spatiotemporal topology engine: Receive continuous scores as core input data for building risk visualizations and assessing global ownership entropy values; Connect to the dynamic decoupled rule engine: Receive the specific triggered fraud rule type, rule ID, and judgment details, which are used to mark abnormal relationships in the knowledge graph and generate rule trigger decomposition paths in the auditable evidence chain; Connect to the self-evolving risk control system: Receive updated cargo ownership entropy value model parameters and rule weight changes, which are used to refresh the risk heat map and display the system's own evolution timeline; Connect with quantum fingerprint anchoring and blockchain evidence storage modules: call on transaction records, quantum fingerprints and Merkel paths on the blockchain to provide underlying tamper-proof data traceability for auditable evidence chains; Connect to the physical event four-dimensional tensor construction module: Obtain the original spatiotemporal event sequence as the basic data for constructing nodes and edges in the spatiotemporal knowledge graph.

[0028] A real-time blocking method for maritime cargo title fraud based on physical event anchoring and spatiotemporal decoupling includes the following processing steps: Real-time acquisition phase: Container acceleration waveforms, ship trajectory coordinates, customs digital signatures, and RFID timestamps are synchronously acquired through an IoT device network to construct a four-dimensional data tensor containing spatial coordinates, time series, device signatures, and event types; Quantum anchoring phase: Generate an event fingerprint containing device signature, GPS coordinates and millisecond-level timestamp, generate a copy-resistant identifier through quantum circuits and write it into the blockchain; The spatiotemporal decoupling phase is carried out as follows: (1) The continuity score is calculated using the Hurwitz model; (2) Dynamic rule matching is performed: when a bill of lading transfer is detected, there is illegal unlocking, route deviation > 200 nautical miles or multiple bill of lading conflicts, a high-risk response is triggered. Zero-knowledge verification stage: Generate zk-SNARK proofs for customs inspection evidence and verify spatiotemporal consistency; Conduct risk prevention and control assessment and implementation: Level 3 Response: Freeze Account + Suspend Bill of Lading Processing + Generate Legal Evidence Package Level 2 Response: Initiate video verification + restrict bill of lading operations Level 1 Response: Send encrypted alert + update knowledge base; Evolutionary stage: Historical events are analyzed through DBSCAN clustering. When the false alarm rate of a rule is less than 5%, the weight is increased by 0.05.

[0029] This invention provides a real-time cargo ownership fraud prevention system based on physical event anchoring and spatiotemporal decoupling. This system deeply integrates IoT sensing, blockchain evidence storage, quantum encryption, and self-evolving risk control technologies. Addressing the issues of multiple sales of the same goods and fraudulent pledging of empty bills of lading arising from the separation of bills of lading (B / L) from physical control in maritime trade, it constructs a full-stack solution covering physical event collection, spatiotemporal decoupling analysis, on-chain real-time blocking, and dynamic defense optimization. By establishing a judicially-grade trusted verification channel between the physical world and digital cargo ownership, this system provides real-time, accurate, and arbitrable cargo ownership security for the global supply chain, signifying a trend and possibility of international trade trust mechanisms entering a new era of quantum-level trust. This invention has the following advantages and benefits: Trustworthy digitization of the physical world: Transforming physical events such as container status and ship trajectory into tamper-proof on-chain evidence through quantum-level anti-counterfeiting tags; Millisecond-level decoupling of title to goods and logistics: Building a dynamic rule engine to achieve real-time comparison of bill of lading transfer events and physical control events; Judicial-level enforcement blocking: Automatically triggers account freezing, bill of lading locking, and generation of arbitrable evidence packages based on risk classification; Continuous Evolution Defense: Dynamically optimize defense strategies through machine learning to counter ever-evolving fraud tactics. Attached Figure Description

[0030] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0031] Figure 1 This is a flowchart illustrating the overall architecture of a real-time blocking system for maritime cargo title fraud based on physical event anchoring and spatiotemporal decoupling, as described in this invention. Figure 2 This is a flowchart illustrating the generation of quantum fingerprints for physical events in a factual case of the present invention. Figure 3 This is a flowchart of the spatiotemporal continuity scoring calculation model in the factual case of this invention; Figure 4 This is a flowchart illustrating the three-level risk response process for fraud prevention in a factual case of this invention. Figure 5 This is a flowchart of the self-evolving risk control system in an embodiment of the present invention; Figure 6 This is a flowchart of zero-knowledge cross-validation in an embodiment of the present invention; Figure 7This is a waveform analysis diagram used in an embodiment of the present invention for preliminary diagnosis of the legality of a single physical event (such as a single unlocking or locking operation); Figure 8 A blockchain structure diagram that provides intuitive proof of data immutability and complete traceability in the implementation examples of this invention; Figure 9 This invention provides a heatmap of cargo ownership entropy values ​​in an implementation example to enable regulatory personnel to gain a global risk situational awareness, quickly identify high-risk areas, and make strategic decisions. Figure 10 This is a three-dimensional spatiotemporal distribution map of the cargo ownership entropy value, used in an embodiment of the present invention to trace the dynamic changes in risk of a single cargo throughout its entire transportation lifecycle. Figure 11 This invention provides a goods ownership-logistics alignment verification diagram for visually diagnosing fraudulent behavior and for the rule engine's determination. Figure 12 This is a timeline of the system's self-healing actions used in the implementation examples of this invention for post-event auditing and review, demonstrating the effectiveness and response speed of the system's complete measures in response to specific risk events; Figure 13 This invention provides a pie chart showing the distribution of system self-healing actions to quickly understand the distribution ratio of system response measures and assess the overall risk level and resource consumption trend in the implementation examples of this invention. Figure 14 This is a timeline of rule evolution in the implementation examples of the present invention, illustrating the learning and growth process of the risk control model itself. Figure 15 This is a rule matrix analysis diagram used as an analysis tool for deep self-optimization of the rule base in an embodiment of the present invention. Detailed Implementation It should be noted that, unless otherwise specified, the embodiments and features in the embodiments of the present invention can be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and embodiments.

[0032] 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, and not all embodiments. The following description of at least one exemplary embodiment is merely illustrative and is in no way intended to limit the present invention or its application or use. 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.

[0033] Figure 1This is a flowchart illustrating the overall architecture of a real-time blocking system for maritime cargo title fraud based on physical event anchoring and spatiotemporal decoupling, as described in this invention. Figure 2 This is a flowchart illustrating the generation of quantum fingerprints for physical events in a factual case of the present invention. Figure 3 This is a flowchart of the spatiotemporal continuity scoring calculation model in the factual case of this invention; Figure 4 This is a flowchart illustrating the three-level risk response process for fraud prevention in a factual case of this invention. Figure 5 This is a flowchart of the self-evolving risk control system in an embodiment of the present invention; Figure 6 This is a flowchart of zero-knowledge cross-validation in an embodiment of the present invention; A real-time prevention system for maritime cargo title fraud based on physical event anchoring and spatiotemporal decoupling includes: The physical event four-dimensional tensor construction module is used to construct physical events based on the three-axis acceleration waveform of container smart locks, the latitude and longitude coordinate sequence of ship AIS, the digital signature image of customs inspection equipment, and the geofence timestamp of warehouse RFID, and to construct physical events containing four-dimensional data tensor representations of spatial coordinates, time series, equipment signatures and event types. Quantum fingerprint anchoring module: Used to generate a unique, copy-resistant identifier for physical event data generated by the physical event four-dimensional tensor construction module through a 12-bit quantum circuit, and anchor it to the blockchain; The Hurwitz Spatiotemporal Topology Engine is used to calculate a continuity score for generating unique, anti-copying identifiers from physical events generated by the physical event four-dimensional tensor construction module and anti-counterfeiting identifiers provided by the quantum fingerprint anchoring module. The Hurwitz Spatiotemporal Topology Engine is the system's core risk assessor. It receives structured event sequences from the physical event four-dimensional tensor construction module and anti-counterfeiting identifiers from the quantum fingerprint anchoring module as input, and calculates the continuity score of the events through a spatiotemporal model. Its output score is a quantified risk indicator, primarily fed directly to the three-level risk response execution module as the core basis for triggering on-chain blocking operations. Simultaneously, this score is also provided as a key feature to the self-evolving risk control system for subsequent fraud pattern clustering analysis and model optimization. The zero-knowledge cross-validation module is used to verify the credibility of physical events generated by the physical event four-dimensional tensor construction module under privacy protection using zero-knowledge proof methods. This module is the system's privacy protection and credibility verifier, specifically processing sensitive raw data (such as customs inspection photos) from the physical event four-dimensional tensor construction module and generating zero-knowledge proofs that do not reveal details. This proof is primarily provided to the dynamically decoupled rule engine, enabling it to ensure credibility while protecting privacy when verifying key claims such as "authenticity of inspection." This strengthens the credibility foundation of the entire rule judgment and the subsequently generated judicial evidence package.

[0034] Dynamic decoupling rule engine: Based on the physical events generated by the physical event four-dimensional tensor construction module, the obtained bill of lading flow, and the trustworthy verification transmitted by the zero-knowledge cross-verification module, it compares the bill of lading flow with the logistics events in real time through the built-in fraud rule library, and identifies fraud patterns based on preset rules; the dynamic decoupling rule engine is the pattern recognition and logical judgment center of the system; it receives two data streams in parallel: one is the "logistics event" stream from the physical event four-dimensional tensor construction module, and the other is the "bill of lading flow" from the external business system. It performs real-time comparison through the built-in fraud rule library (such as spatio-temporal conflict, path deviation). The specific fraud patterns identified by it directly drive the three-level risk response execution module to make accurate responses; at the same time, its rule trigger records and effect feedback are given to the self-evolving risk control system, serving as the data basis for dynamic optimization of rule weights and expansion of the rule library; Three-level risk response execution module: Based on the continuous score transmitted by the Hurwitz spatio-temporal topology engine and the fraud pattern transmitted by the dynamic decoupling rule engine, it automatically triggers an on-chain blocking operation to make a judgment on the three-level risk response; Self-evolving risk control system: Using the continuous score transmitted by the Hurwitz spatio-temporal topology engine, it dynamically optimizes the rule weights and the cargo right entropy value model through fraud pattern clustering, and generates a defense mechanism for triggering an on-chain blocking operation.

[0035] The self-evolving risk control system is the core of the system's self-learning and optimization. It obtains continuous scores from the Hurwitz spatio-temporal topology engine, rule historical records from the dynamic decoupling rule engine, and calculates the cargo right entropy value from the global data, so as to perform fraud pattern clustering and model analysis. The optimized rule weights and newly discovered defense rules output by it are directly fed back and updated to the rule library of the dynamic decoupling rule engine, thereby enhancing the real-time detection ability of the system; at the same time, the updated entropy value model also provides a more accurate basis for the risk visualization of the four-dimensional decision-making output module.

[0036] Furthermore, the dynamic decoupling rule engine includes four core rules: spatio-temporal conflict detection, path deviation detection, multi-party statement conflict detection, and physical-document disconnection detection, which are specifically as follows: Spatio-temporal conflict detection: When the bill of lading transfer timestamp tb and the container opening timestamp to satisfy 0 < to - tb < 1800 seconds, a warning of cargo right theft is triggered; Path deviation detection: Calculate the Haversine distance between the ship's position and the declared route in real time. When > 200 nautical miles continuously for more than 1 hour, a ghost ship blocking is triggered; Multi-party statement conflict detection: Retrieve the ownership statement records of the same container on the blockchain. When there are multiple valid bills of lading, a judicial evidence package for selling the same cargo to multiple parties is generated; Physical-Document Chain Break Detection: When the warehouse RFID entry timestamp has no corresponding bill of lading data on the blockchain (not just the timestamp), it is marked as a high-risk event of empty bill of lading pledging; The dynamic optimization algorithm for the weights of the four rules—spatiotemporal conflict detection, path deviation detection, multi-party declaration conflict detection, and physical-document chain break detection—is determined as follows:

[0037]

[0038] Where: FP is the false alarm rate of the rule. For learning rate, This represents the current weight value of the i-th rule. This represents the updated weight value of the i-th rule, which is the result of dynamic optimization.

[0039] Furthermore, the process of constructing a physical event containing a four-dimensional data tensor representation of spatial coordinates, time series, device signature, and event type based on the three-axis acceleration waveform of the container smart lock, the latitude and longitude coordinate sequence of the ship's AIS, the digital signature image of the customs inspection equipment, and the geofence timestamp of the warehouse RFID is as follows: S11: Perform joint time-frequency domain analysis on the waveform of the container smart lock: In the time domain, abrupt changes in acceleration amplitude are detected, and a forced unlocking event is marked when the waveform peak value is greater than 1.5g. In the frequency domain, the main frequency component is extracted using Fast Fourier Transform, and illegal interference is marked when the main frequency offset rate is >15%. S12: Ship trajectory anomaly detection uses the Havessing distance model: 1.

[0040] in: Latitude R is the longitude, and R is the Earth's radius. When three consecutive points are... A course deviation warning is triggered when the distance exceeds 200 nautical miles. R is the spherical distance between two geographical locations, where R is the Earth's radius. The difference in latitude between two points. The difference in longitude between the two points. The latitude value representing the starting point Indicates the latitude value of the endpoint.

[0041] S13: Establish a spatiotemporal consistency verification mechanism between the customs inspection timestamp tc and the smart lock operation timestamp tl: the verification is deemed successful when |tc - tl| < 300 seconds and the container numbers are consistent.

[0042] S11, S12, and S13 are parallel and collaborative, but they are not sequentially connected. Instead, they are independent analysis units that execute in parallel within the physical event four-dimensional tensor construction module, each processing different data sources. They each extract high-level semantic information from the raw data (such as event type "forced unlocking," status label "course deviation," and verification conclusion "spatiotemporal consistency passed"), and output this information to a unified tensor construction framework for integration. From a logical synergy perspective, they are complementary: for example, the "course deviation alarm" output by S12 can provide a high-risk context for the event (unlocking analysis) and S13 verification of containers on the same ship; while the "verification passed" conclusion of S13 can strengthen or challenge the legitimacy of the event determined by S11. Therefore, there is no linear information transmission among the three; rather, they enrich the event attributes and associated contexts in the "four-dimensional tensor" together, providing high-quality, mutually verifiable input for subsequent quantum anchoring and rule analysis.

[0043] Furthermore, the quantum fingerprint anchoring module includes: Quantum fingerprint generation unit: used to calculate a unique, copy-resistant quantum fingerprint identifier for each original physical event; Multi-source spatiotemporal alignment unit: used to receive event data with fingerprint identifiers transmitted by the quantum fingerprint generation unit, verify whether multiple related events belong to the same coherent physical process by checking whether the timestamp deviations are within the threshold, and output a group of related events that have passed the spatiotemporal consistency verification; the maximum time deviation of ship AIS location, warehouse RFID record, and customs inspection evidence is required to be <30 minutes. Blockchain evidence storage unit: used to construct a causal chain with Merkle tree roots in chronological order based on the associated event group transmitted by the multi-source spatiotemporal alignment unit, and package it together with the previous block hash, quantum fingerprint and digital signature to form an immutable permanent blockchain evidence storage unit. The quantum fingerprint generation unit performs the following operations via quantum circuitry: S21: Hash the original event data and extract it into a 48-bit classic binary string, which is then output as the initialization parameter. S22: Based on the binary string, apply the X gate to the corresponding qubit to encode the classical information into a definite initial quantum state; S23: Apply a Hadamard gate to the initial quantum state, convert the initial quantum state with the Hadamard gate applied into a uniform superposition state, and output a quantum superposition state containing all possibilities; S24: Apply the CX entanglement gate to the quantum superposition state to induce quantum entanglement between bits and output a highly correlated entangled state; S25: Finally, the entangled state is measured multiple times to collapse the quantum information into a classical probability distribution, and the classical bit sequence with the highest probability is output as the final fingerprint identifier. The execution process of the blockchain evidence storage unit is as follows: The complete block data to be signed consists of the event causal chain, the preceding block SHA-256 and quantum fingerprint identifier, and the ECDSA digital signature path. Then, a digital signature overlaying this data is generated using a private key.

[0044] The Chinese meaning of CX is "Controlled-NOT Gate". In the quantum fingerprint generation process, the purpose of applying CX (controlled-NOT gate) to adjacent qubits is to associate these bits into an indivisible whole, ensuring that the identifier obtained by the final measurement is highly bound to the original input data and is extremely difficult to forge.

[0045] Furthermore, the zero-knowledge cross-validation module is used to generate verifiable proofs of customs inspection evidence based on zk-SNARK technology. The execution process of the zero-knowledge cross-validation module is as follows: S51: The process for generating a customs inspection certificate is as follows: Obtain photos of the container inspection; Based on the container inspection photos, extract the container number hash H_{container} and the geofence location stamp (lat,lon); Based on the hash H_{container} and the geofence location stamp, construct the zk-SNARK circuit verification declaration: S52: Perform advance bill of lading transaction detection: when the timestamp of the transfer of ownership is t o With container operation timestamp t c When |to - tc|>1800 seconds, a fraud evidence vector is generated.

[0046] There is a close collaborative relationship between S51 and S52, namely, "providing basic data" and "analyzing upper-level risks". Simply put, S51 is the process of ensuring that "the timestamp we receive is genuine", while S52 is the process of using "this genuine timestamp" to determine whether the business logic is abnormal.

[0047] The formula for calculating the continuity score of an event is:

[0048] in, =0.5、 =0.3、 =0.2 is the empirical weighting coefficient. =1800s (30-minute critical value). =50000m (50 km critical value). =100m / s (critical value of 360 km / h). Indicates the time difference between consecutive events (such as the interval between bill of lading transfer and container unlocking). This represents the distance difference between consecutive locations (based on Havesien distance), and v represents the movement speed. The time difference (in seconds) between consecutive events. The Hurwitz spacetime topology engine also includes: 5.1 Relativistic Time Dilation Correction: Introducing the speed of light c = 299792458 m / s, the time dilation factor corresponding to velocity v is calculated:

[0049] Corrected time difference ; 5.2 Continuity Break Determination: When the score value is <0.8, a three-dimensional alarm vector containing the break location, time deviation, and velocity factor is generated. The score value is calculated using the mathematical formula of the Hurwitz spacetime topology model. The score value S is a quantitative result calculated by a deterministic mathematical model that comprehensively considers the three physical dimensions of time, distance, and velocity and assigns weights. The formula is described in the specific implementation details.

[0050] Furthermore, the three levels of risk are divided as follows: Level 3 response: Continuity score < 0.5. When a response is classified as Level 3: (1) Freeze the flow of funds in the relevant bank accounts; (2) Suspend the right to transfer bills of lading on the blockchain network; (3) Generate a judicial evidence package that conforms to the Hague-Visby Rules, including: Blockchain transaction ID Merkle path; Timestamped event sequence causal chain; ECDSA signature verification tree; (1) Freezing the flow of funds in relevant bank accounts, (2) suspending the transfer rights of bills of lading on the blockchain network, and (3) generating a judicial evidence package that complies with the Hague-Visby Rules are three actions that are triggered in parallel and executed in concert. As the core actions of the highest risk level (Level 3 response), they aim to simultaneously and quickly achieve the isolation of fund risks, the freezing of the transfer of ownership, and the fixation of judicial evidence, so as to maximize the fraud prevention effect and prepare for subsequent legal proceedings. The three actions are concurrent, but there is logical coordination and connection at the evidentiary level. Information flows from the emergency response instructions to the evidence package construction process, ensuring the completeness, timeliness, and enforceability of the final judicial evidence.

[0051] Level II response, 0.5 ≤ continuous score < 0.8; when judged as a Level II response: (1) Restrict the authority to process bills of lading; (2) Initiate a multi-party video verification mechanism; (3) Generate a risk warning evidence chain; Level 1 Response: Consecutive scores ≥ 0.8: When judged as a Level 1 response: (1) Send an AES-256 encryption risk warning to the regulator; (2) Update the historical pattern records of the risk knowledge base.

[0052] Furthermore, the self-evolving risk control system, which dynamically optimizes rule weights and asset entropy models through fraud pattern clustering to generate a defense mechanism that triggers on-chain blocking operations, follows this process: S71: Calculate the entropy value of cargo ownership based on the dynamic model of cargo ownership entropy value; S72: Based on the entropy value of goods ownership, a fraud pattern clustering engine is used to perform clustering analysis using this feature vector. This identifies new fraud patterns and generates defense rules. The feature vector includes a continuity score, the entropy value of goods ownership, and rule weights. "This feature vector" refers to the "historical event feature vector" extracted and constructed by the self-evolving risk control system from historical event data within the fraud pattern clustering engine, used for machine learning. "This feature vector" refers to the specific input data object used by the fraud pattern clustering engine for clustering analysis in the above steps, i.e., the "historical event feature vector."

[0053] The eigenvector is clearly defined in the original text, and its standard structure is as follows: Eigenvector X = [Continuity score, Cargo entropy value, Rule weight] Continuity score: Calculated by the Hurwitz spacetime topology engine, it characterizes the spatiotemporal plausibility of a single or continuous event.

[0054] Cargo entropy value: Calculated by the cargo entropy value dynamic model, it represents the uncertainty and risk disorder of the entire cargo flow.

[0055] Rule weight: derived from the current rule library of the dynamically decoupled rule engine, each rule has its dynamically adjusted weight value.

[0056] The dynamic model of cargo ownership entropy:

[0057]

[0058] Where: pk is the probability of the cargo's safe state. is the standard deviation of the continuity score; wi is the event type weight (change of ownership = 1.2), ft is the time decay factor (30-day linear decay), and si is the spatial dispersion adjustment coefficient; Rule base evolution mechanism: After processing every 100 events, weight optimization and rule expansion are automatically performed.

[0059] Furthermore, the system also includes a decision-making output module. The function of this module is to integrate, transform, and elevate the complex and abstract risk data and judgment results generated by the various analysis modules within the system into decision-making knowledge that human decision-makers can intuitively understand, comprehensively control, and directly apply to judicial and auditing matters. Essentially, it is a decision support center whose core value lies in empowering managers, regulators, and judicial personnel with the intelligent analytical capabilities of machines through visualized, graphical, and structured evidence. It has direct connections and data interaction relationships with the following modules: Connect to the Hurwitz spatiotemporal topology engine: Receive continuous scores as core input data for building risk visualizations and assessing global ownership entropy values; Connect to the dynamic decoupled rule engine: Receive the specific triggered fraud rule type, rule ID, and judgment details, which are used to mark abnormal relationships in the knowledge graph and generate rule trigger decomposition paths in the auditable evidence chain; Connect to the self-evolving risk control system: Receive updated cargo ownership entropy value model parameters and rule weight changes, which are used to refresh the risk heat map and display the system's own evolution timeline; Connect with quantum fingerprint anchoring and blockchain evidence storage modules: call on transaction records, quantum fingerprints and Merkel paths on the blockchain to provide underlying tamper-proof data traceability for auditable evidence chains; Connect to the physical event four-dimensional tensor construction module: Obtain the original spatiotemporal event sequence as the basic data for constructing nodes and edges in the spatiotemporal knowledge graph.

[0060] The spatiotemporal knowledge graph is constructed as follows: (1) Nodes: cargo entity (container ID), bill of lading entity (B / L number), port entity; (2) Edge relationships: control transfer (gold star), spatial movement path (red deviation warning), time dependency chain; The system also includes a risk visualization system: (1) Heat map of cargo ownership entropy: Areas with entropy values ​​> 0.5 are marked with a red warning; (2) Three-dimensional spatiotemporal distribution map: marking the continuity breakpoints; (3) Dynamic decision matrix: including detour plan, acceleration plan, and transfer plan; The auditable evidence chain includes: (1) a blockchain transaction traceability tree; (2) a rule-triggered decomposition path; and a quantum fingerprint verification log.

[0061] A real-time blocking method for maritime cargo title fraud based on physical event anchoring and spatiotemporal decoupling includes the following processing steps: Real-time acquisition phase: Container acceleration waveforms, ship trajectory coordinates, customs digital signatures, and RFID timestamps are synchronously acquired through an IoT device network to construct a four-dimensional data tensor containing spatial coordinates, time series, device signatures, and event types; Quantum anchoring phase: Generate an event fingerprint containing device signature, GPS coordinates and millisecond-level timestamp, generate a copy-resistant identifier through quantum circuits and write it into the blockchain; Perform the spatiotemporal decoupling phase: (1) The continuity score was calculated using the Hurwitz model; (2) Execute dynamic rule matching: When a bill of lading transfer is detected, if there is illegal unlocking, route deviation >200 nautical miles, or multiple bill of lading conflicts, a high-risk response is triggered; Zero-knowledge verification stage: Generate zk-SNARK proofs for customs inspection evidence and verify spatiotemporal consistency; Conduct risk prevention and control assessment and implementation: Level 3 Response: Freeze Account + Suspend Bill of Lading Processing + Generate Legal Evidence Package Level 2 Response: Initiate video verification + restrict bill of lading operations Level 1 Response: Send encrypted alert + update knowledge base; Evolutionary stage: Historical events are analyzed through DBSCAN clustering. When the false alarm rate of a rule is less than 5%, the weight is increased by 0.05.

[0062] Example 1 A real-time prevention system for maritime cargo title fraud based on physical event anchoring and spatiotemporal decoupling includes: The physical event four-dimensional tensor construction module is used to construct physical events based on the three-axis acceleration waveform of container smart locks, the latitude and longitude coordinate sequence of ship AIS, the digital signature image of customs inspection equipment, and the geofence timestamp of warehouse RFID, and to construct physical events containing four-dimensional data tensor representations of spatial coordinates, time series, equipment signatures and event types. Quantum fingerprint anchoring module: used to generate a unique, copy-resistant identifier for physical event data using a 12-bit quantum circuit; Hurwitz Spacetime Topology Engine: Used to calculate the continuity score of physical events that generate unique, copy-resistant identifiers; Level 3 Risk Response Execution Module: Used to automatically trigger on-chain blocking operations based on the physical event continuity score, and to make judgments on Level 3 risk responses; Self-evolving risk control system: used to dynamically optimize rule weights and cargo entropy models through fraud pattern clustering, generating a defense mechanism that triggers on-chain blocking operations.

[0063] Furthermore, the physical event four-dimensional tensor construction module is used to solve the problems of acquisition, standardization, and fusion of multi-source heterogeneous physical data. Container smart locks provide comprehensive monitoring: 1) Deploy an industrial-grade triaxial accelerometer (sampling rate ≥100Hz) to continuously collect high-precision waveform data of the lock / unlock operation.

[0064] 2) Temporal domain depth analysis: A sliding window algorithm is used to detect the peak acceleration of the waveform. When the instantaneous acceleration exceeds 1.5g (equivalent to a violent impact), it is determined to be a violent unlocking event by combining it with duration spectrum analysis. At the same time, the waveform kurtosis index is calculated. When the kurtosis is >3.0, it is identified as an abnormal impact mode.

[0065] 3) Frequency Domain Intelligent Diagnosis: Frequency domain features from 0-100Hz are extracted using a 4096-point Fast Fourier Transform (FFT), and the energy proportion of the dominant frequency component is calculated. When the dominant frequency offset exceeds 15% (e.g., spectrum broadening during electronic interference attacks), it is marked as an illegal operation event. Simultaneously, the spectral entropy value is calculated; an abnormally high entropy value indicates potential interference behavior.

[0066] Ship trajectory dynamic verification system: 1) Real-time access to AIS broadcast data stream, parsing latitude and longitude coordinates updated every minute. Establishing a ship kinematics model, and predicting the next position using Kalman filtering.

[0067] 2) A precise calculation model for the Havesien distance based on spherical geometry:

[0068] in Latitude (radians) Longitude (in radians). When three consecutive points... When the distance exceeds 200 nautical miles, it is identified as a "ghost ship" fraud mode, triggering a course deviation alarm; (3) Customs-Warehouse Spatiotemporal Alignment Engine: 1) Customs inspection terminals are equipped with digital signature cameras; the image metadata includes: Container number SHA-256 hash value, BeiDou / GPS dual-mode positioning coordinates (accuracy ≤ 0.5 meters), and trusted timestamp (synchronized atomic clock, error < 1 millisecond).

[0069] 2) Warehouse UHF RFID access control system records: Container EPC code, geofence entry timestamp, reader location coordinates.

[0070] 3) Establish a spatiotemporal consistency verification protocol: If the customs inspection time tc and the warehouse entry time tw satisfy |tc - tw|>30 minutes, the video audit process will be automatically triggered.

[0071] 2.2 Quantum Fingerprint Blockchain Anchoring Module This module enables the preservation of physical events in a non-replicable and tamper-proof manner. (1) Quantum fingerprint generation algorithm (innovative anti-counterfeiting mechanism): 1) Data preprocessing: Perform SHA-256 hash on the raw event data (such as acceleration waveform + GPS coordinates + timestamp), output a 256-bit digest, and truncate the first 48 bits as the binary input sequence.

[0072] 2) Quantum state preparation: Initialize a 12-bit quantum register and apply an X gate (bit flip operation) to the bits with a value of '1' in the binary sequence.

[0073] 3) Quantum superposition: Applying a Hadamard gate (H gate) to all qubits creates a quantum superposition state. .

[0074] 4) Quantum entanglement: Applying a CX gate (controlled NOT gate) to adjacent bits generates entangled states such as... .

[0075] 5) Probability measurement: Perform 1024 measurements under the calculated ground state, count the probability of each state, and take the state with the highest probability as the final fingerprint identifier (e.g., "110010101101").

[0076] (2) Multi-source spatiotemporal alignment verification: Establish an event correlation matrix: For the same container, calculate the maximum time deviation max(|t1-t2|, |t2-t3|, |t1-t3|) for ship arrival event E1, warehouse entry event E2, and customs inspection event E3. When the deviation is greater than 30 minutes, it is marked as a "spatiotemporal discontinuity event" and multi-party review is initiated.

[0077] (3) Blockchain-based evidence storage architecture: 1) Adopting a layered blockchain design: Physical layer chain: Each container is assigned an independent chain, and the block header contains: version number (fixed to 0x01), previous block hash (SHA-256 digest), Merkle root (event dataset hash), timestamp (UNIX millisecond level), quantum fingerprint (12-bit measurement state), and nonce value (proof of work).

[0078] Business layer chain: Stores business events such as bill of lading transfer and change of ownership, and interacts with the physical layer chain through cross-chain protocol.

[0079] 2) Data anchoring mechanism: After physical events are processed by quantum fingerprints, they are written into the physical layer chain and simultaneously generated into a degenerate hash anchored to the business layer chain, forming a dual-chain mutual verification structure.

[0080] 2.3 Hurwitz Spacetime Topology Engine This module quantifies the spatiotemporal continuity risks of the flow of ownership and logistics: (1) Mathematical model for continuous scoring:

[0081] in: =0.5、 =0.3、 =0.2 is the empirical weighting coefficient. =1800s (30-minute critical value). =50000m (50 km critical value). =100m / s (critical value of 360 km / h). Indicates the time difference between consecutive events (such as the interval between bill of lading transfer and container unlocking). represents the distance difference between consecutive locations (based on Havesing distance), and v represents the movement speed.

[0082] (2) Relativistic correction mechanism (optimization for high-speed scenarios) When air transport is detected (v > 100 m / s), a special relativistic time dilation correction is introduced:

[0083] Replace the original time difference with the corrected time difference to avoid misjudgment caused by high-speed movement.

[0084] (3)Continuous break alarm generation: When the score value is lower than 0.8, generate a three-dimensional alarm vector:

[0085] The alarm vector is input into the dynamic rule engine to trigger the subsequent risk analysis process.

[0086] 2.4 Dynamic decoupling rule engine This module realizes the intelligent identification of fraud patterns and weight optimization: (1)Core rule library and detection logic: 1) Temporal and spatial conflict rule: Monitor the temporal relationship between the bill of lading transfer operation and the physical events of the container. When it is detected that the bill of lading transfer time tb and the container opening timestamp to satisfy 0 < to - tb < 1800 seconds, it is determined as a high-risk event of "theft of goods ownership", and an evidence vector containing the time deviation value is generated.

[0087] 2) Path deviation rule: Real-time calculate the Haversine distance between the ship's position and the declared route. When three consecutive points > 200 nautical miles and the duration > 1 hour, it is determined as "phantom ship" fraud, and the satellite image verification process is automatically triggered.

[0088] 3) Multi-party statement conflict rule: Retrieve the container ownership statement records on the blockchain. When multiple valid bills of lading are associated with the same container number, a judicial evidence package of "selling the same goods to multiple parties" is automatically generated, including the hash values and timestamp sequences of all associated bills of lading.

[0089] 4) Physical-document disconnection rule: Compare the warehouse RFID entry records with the blockchain bill of lading data. When the warehouse detects the entry of a container but there is no corresponding bill of lading record, it is marked as the highest risk level of "pledging an empty bill", and the relevant accounts are immediately frozen.

[0090] (2)Rule weight dynamic optimization system: After processing every 1000 events, count the false positive rate (FPR) of each rule:

[0091] Weight update algorithm: For example: The initial weight of the path deviation rule is 0.9. When the false positive rate drops to 2.8% for three consecutive periods, the weight is increased to 0.95 to enhance its influence in decision-making.

[0092] Furthermore, regarding the zero-knowledge cross-validation module, this module enables trusted verification of physical events while protecting privacy: (1) Customs inspection zero-knowledge certificate process: 1) Customs officers use specialized equipment to take photos of the container during inspection, and the system automatically extracts them: Box number hash value: Geofence location stamp: (lat, lon), timestamp: tc (synchronized with BeiDou time synchronization system); 2) Construct a zk-SNARK arithmetic circuit to verify the three core declarations: Declaration 1 (Data Integrity): The photo pixel hash Hphoto is consistent with the original data, satisfying the requirement...

[0093] Statement 2 (Location Credible): The shooting location was within 100 meters of the declared geofence.

[0094] Statement 3 (box number matches): It matches the container number hash exactly as stated on the bill of lading.

[0095] Generate a concise proof (Proof π) with a size of <256 bytes, which can be publicly verified without revealing the original photo.

[0096] (2) Bill of lading advance transaction detection mechanism: 1) Monitor the deviation between the timestamp 'to' for the transfer of ownership and the timestamp 'tc' for container operations in real time. When |to - tc| > 1800 seconds (i.e., the change of ownership is more than 30 minutes earlier or later than the physical operation), generate a fraud evidence vector:

[0097] 2) Input the evidence vector into the risk response system to trigger a response at level 2 or higher.

[0098] Furthermore, regarding the Level 3 risk response execution module, this module implements tiered and automated blocking operations: (1) Level 3 response (continuity score < 0.5, extremely high risk): 1) Funds Freeze: Send an account freeze command via the SWIFT connection to lock the flow of funds in the relevant bank account in real time.

[0099] 2) Suspension of Bill of Lading Circulation: The status of the bill of lading is updated to "frozen" on the blockchain network, prohibiting operations such as ownership transfer and pledging.

[0100] 3) Generation of judicial evidence packages: Blockchain Transaction Traceability Tree: A Complete Merkle Path from the Genesis Block to the Current Event Timestamped event causal chains: for example, "2025-08-14T08:30: Bill of Lading Transfer → 2025-08-14T08:35: Container Opening → 2025-08-14T09:00: Location Offset" Multi-layered digital signature verification path: Device signature → Customs officer signature → Bank authorization signature (2) Level II response (0.5 ≤ score < 0.8, medium risk): 1) Reduce the authority of bill of lading operations: restrict bill of lading operations to "read-only mode", allowing viewing but prohibiting transfer, pledge and other change operations.

[0101] 2) Multi-party video verification: Automatically initiate Zoom / Teams video conferences to invite ship owners, cargo owners, and banks to verify the real-time status of goods online and record video evidence.

[0102] 3) Risk warning evidence chain: Generate a simplified judicial package containing key event hashes and timestamp sequences.

[0103] (3) Level 1 Response (score ≥ 0.8, low risk): 1) Encrypted warning notification: Send an AES-256 encrypted risk warning message to the regulator, including the risk type, location coordinates, and timestamp.

[0104] 2) Incremental update of knowledge base: Write the event feature vector $[continuity score, cargo entropy value, rule weight]$ into the risk knowledge base to optimize subsequent decision-making.

[0105] Furthermore, regarding the self-evolving risk control system, this module enables continuous optimization and upgrading of defense strategies: (1) Fraud pattern clustering engine: 1) Feature Engineering: Extracting multi-dimensional features of historical events:

[0106] 2) Density clustering: The DBSCAN algorithm (parameters eps=0.5, min_samples=10) is used to identify outlier clusters. Calculate the Euclidean distance between samples:

[0107] Mark core points, boundary points, and noise points; merge density to form sample clusters. 3) Rule generation: When a new fraud pattern is discovered (such as "ghost container" - forging non-existent container numbers), a defense rule is automatically generated: "If the container has no genesis event on the blockchain, it is marked as high risk".

[0108] (2) Dynamic model of cargo entropy value: Entropy calculation is performed in two steps: Step 1: Calculate the state probability distribution

[0109] Where: pk is the probability of cargo safety status, is the standard deviation of the continuity score, wi is the event type weight (change of ownership = 1.2, usually 1.0), and ft is the time decay factor (30-day linear decay). ,in ), si is the spatial dispersion adjustment coefficient (when the standard deviation of the location point is... >100,000 meters =1.2, otherwise =1.0); Step 2: Calculate Shannon entropy and standard deviation compensation

[0110] in: The standard deviation of continuous scores; Entropy application: When the entropy value is >0.5, a system warning is triggered; when it is >0.7, the response level is automatically upgraded.

[0111] (3) Rule base evolution mechanism: Triggering condition: Activated when 100 events are processed or a new cluster is detected.

[0112] Execution process: Weight optimization → Cluster analysis → Incremental update of rule base → Entropy model parameter calibration.

[0113] Furthermore, regarding the four-dimensional decision output system, this module provides a panoramic risk governance interface and decision support: The four-dimensional decision output system includes: (1) Spatiotemporal knowledge graph architecture: 1) Node Entities: Cargo Entity (Attributes: Container ID, Cargo Type, Weight); Bill of Lading Entity (Attributes: B / L Number, Cargo Owner, Issuance Time); Port Entity (Attributes: Port Code, Geographic Location, Throughput) 2) Edge relationships: Control transfer (edge ​​attributes: transfer time, transaction parties, visually marked with a gold star); Spatial movement path (edge ​​attributes: distance, time consumption, abnormal paths are marked with a red warning); Time-dependent chain (edge ​​attributes: time difference, breakpoints are marked with a flashing dashed line); (2) Risk visualization subsystem: 1) Cargo ownership entropy heatmap: using a global map as the base map, colored by port region: Green (entropy < 0.3): Safe zone; Yellow (0.3 ≤ entropy < 0.5): Low risk; Red (entropy value ≥ 0.5): High risk (e.g., Shanghai Port's entropy value = 0.73 on August 14, 2025) 2) Three-dimensional spatiotemporal distribution map: Mark the continuity break points in the three-dimensional coordinate system: X-axis: Longitude (-180°~180°); Y-axis: Latitude (-90°~90°); Z-axis: Time deviation (seconds) 3) Dynamic Decision Matrix: Outputs multiple contingency plans and strategies. Detour options: Alternative routes to high-risk sea areas (saving 12-48 hours); expedited options: Applying for priority berthing rights at ports (reducing transshipment time by 30%); transshipment contingency plans: List of backup ports and contact persons (3-5 nearby ports). 4) Auditable chain of evidence system: Blockchain Transaction Traceability Tree: Visually displays the complete Merkle path from the genesis block to the current event, and supports clicking to view the quantum fingerprint and event details of each block.

[0114] Rule-triggered decomposition path: For example: "Path deviates from rule (weight 0.93) → Δd=215 nautical miles detected → trigger LEVEL 3 response → generate evidence package 2025-0814-EVID-01".

[0115] Quantum fingerprint verification log: Records the matching process between the original event data and the on-chain fingerprint, including hash value, quantum state probability distribution and verification result.

[0116] Figure 7 This is a waveform analysis diagram used in an embodiment of the present invention for preliminary diagnosis of the legality of a single physical event (such as a single unlocking or locking operation); Figure 8 A blockchain structure diagram that provides intuitive proof of data immutability and complete traceability in the implementation examples of this invention; Figure 9 This invention provides a heatmap of cargo ownership entropy values ​​in an implementation example to enable regulatory personnel to gain a global risk situational awareness, quickly identify high-risk areas, and make strategic decisions. Figure 10 This is a three-dimensional spatiotemporal distribution map of the cargo ownership entropy value, used in an embodiment of the present invention to trace the dynamic changes in risk of a single cargo throughout its entire transportation lifecycle. Figure 11 This invention provides a goods ownership-logistics alignment verification diagram for visually diagnosing fraudulent behavior and for the rule engine's determination. Figure 12 This is a timeline of the system's self-healing actions used in the implementation examples of this invention for post-event auditing and review, demonstrating the effectiveness and response speed of the system's complete measures in response to specific risk events; Figure 13This invention provides a pie chart showing the distribution of system self-healing actions to quickly understand the distribution ratio of system response measures and assess the overall risk level and resource consumption trend in the implementation examples of this invention. Figure 14 This is a timeline of rule evolution in the implementation examples of the present invention, illustrating the learning and growth process of the risk control model itself. Figure 15 This is a rule matrix analysis diagram used as an analysis tool for deep self-optimization of the rule base in an embodiment of the present invention. Example 2: Real-time blocking of illegal unlocking of containers at Shenzhen Port (1) Example description Container number CONT_001 (Event ID: EVENT_0000CONT_001) experienced an abnormal operation at Shenzhen Port (114.05°E, 22.55°N) on October 1, 2023. The system detected that the peak value of the three-axis acceleration waveform of the smart lock reached 2.1g, exceeding the forced unlocking threshold (1.5g). At the same time, the system recorded that the deviation between the customs inspection timestamp (2023-10-01T08:15:00Z) and the unlocking timestamp (2023-10-01T08:30:25Z) was within 300 seconds.

[0117] (2) Processing flow 1) Physical event acquisition and feature extraction The smart lock sensor collected a waveform sequence [0.2g, 1.8g, 0.5g, 2.1g, 0.3g]. Time domain analysis identified a peak value of 2.1g (exceeding the threshold of 1.5g), and frequency domain analysis showed a main frequency offset rate of 18% (exceeding the 15% threshold), which was marked as a composite event of "forced unlocking + electronic interference".

[0118] Associated with the ship's AIS track (ship ID: MV_STAR, location 114.05°E, 22.55°N) and customs inspection digital signature (container number hash: a3d5f8, Beidou positioning matching).

[0119] 2) Quantum fingerprint generation and blockchain anchoring The original data (waveform + coordinates + timestamp) is hashed using SHA-256 to generate a 48-bit binary string 001011010101.

[0120] The quantum circuit performs X-gate flipping on bits 0, 2, 3, and 5, applies a Hadamard gate to create a superposition state, and uses a CX gate to entangle bits 0-1, 2-3, and 4-5.

[0121] After 1024 measurements, the highest probability state 110100 (probability 62%) is output and written to the physical layer of the container-specific blockchain (block ID: BLK-CONT_001-20231001).

[0122] 3) Spatiotemporal decoupling analysis and rule triggering A search of the bill of lading system revealed that the title to goods in bill of lading B / L-2023-0876 was transferred to "Shenzhen First Health" on 2023-10-01T08:25:30Z.

[0123] Calculate the time difference between the transfer of ownership and unlocking: Δt = 295 seconds (satisfying the time-space conflict rule of 0 < Δt < 1800 seconds).

[0124] The Hurwitz engine calculates the continuity score: Time factor: min(1, 295 / 1800) = 0.164 Distance factor: 0 (same location) Velocity factor: 0 Score = 1.0 - (0.5 × 0.164 + 0.3 × 0 + 0.2 × 0) = 0.92 4) Risk Response Execution Start LEVEL 2 response (score > 0.8): Bill of Lading B / L-2023-0876 has been downgraded to "read-only mode", prohibiting pledging or transfer operations.

[0125] Automatically initiate three-party video verification: the cargo owner (Shenzhen First Health), the shipowner (Wang Hai), and the bank (China Merchants Bank Shenzhen Branch) check the real-time status of the container through an encrypted channel.

[0126] The verification video digital fingerprint 9a97b13221 is stored on the blockchain and confirmed to the business layer block BL-BILL-20231001.

[0127] Example 3: Vessel Deviation Prevention on Shanghai-Los Angeles Route (1) Case Background The vessel MV OCEANIC (AIS ID: SHIP_002) declared its route as Shanghai Port (121.47°E, 31.23°N) to Los Angeles Port (118.24°W, 33.97°N). On October 3, 2023, the system detected that its actual position (132.654°E, 28.542°N) had continuously deviated from its route for 1 hour and 2 minutes, with a maximum distance of 223 nautical miles from Haversington.

[0128] (2) Processing flow 1) Trajectory Anomaly Detection and Data Fusion AIS trajectory analysis: Five consecutive points have distances exceeding the threshold (200 nautical miles) at Haversin, such as: Point 1 (2023-10-03T12:00:00Z): Δd = 223 nautical miles Point 2 (2023-10-03T12:20:00Z): Δd=229 nautical miles The velocity model detected an average speed of 111 m / s (exceeding the 100 m / s threshold), triggering relativistic time dilation correction (correction factor γ = 1.00000007).

[0129] 2) Continuity score calculation Distance factor: min(1, 412km / 50km) = 1.0 (upper limit) Velocity factor: min(1, 111 / 100) = 1.0 (taking the upper limit) Score = 1.0 - (0.5 × 0 + 0.3 × 1.0 + 0.2 × 1.0) = 0.53 3) Level 3 legal-grade blocking Funds Freeze: Freeze cross-border fund flows of the associated account BANK-USD-XXXX-5678 in real time via SWIFT interface.

[0130] Bill of lading lock: The bill of lading status is marked as "judicially frozen" on the business layer blockchain, prohibiting any operation.

[0131] Generate a judicial evidence package (ID: EVID-20231003-002): Blockchain traceability path: Physical layer block BLK-SHIP_002-20231003 → Business layer block BL-BILL-SH20231003 Timestamped event chains: 2023-10-03T12:00:00Z | Deviation Δd = 223 nautical miles 2023-10-03T13:02:00Z | Deviation lasting over 1 hour Digital signature tree: Ship AIS equipment signature: 3325189440 Satellite remote sensing verification signature: 6166439feb Customs Spatiotemporal Consistency Signature: 07568c58f1 Example 4: Identification of Fraudulent Short Selling (1) Case Background On October 1, 2023, the RFID system at the Singapore warehouse detected container CONT_002 (Event ID: EVENT_0002CONT_002) entering the warehouse, but there was no corresponding bill of lading record on the blockchain. At the same time, the bill of lading system showed that the ownership of the container was transferred to two different companies on the same day.

[0132] (2) Processing flow 1) Physical - Document Breakage Detection Warehouse RFID record: Box number CONT_002, entry time 2023-10-01T14:30:00Z, geofence coordinates 103.85°E, 1.28°N.

[0133] Blockchain Bill of Lading Search: Two valid bills of lading were found. B / L-2023-1001A (Transfer time 14:25:00Z, consignee Company A) B / L-2023-1001B (Transfer time 14:28:00Z, consignee Company B) 2) Conflict Analysis of Multiple Statements The dynamic rule engine triggers a composite rule of "selling one commodity multiple times + short selling": Rule 1: Multiple valid bills of lading exist for the same container (time difference 180 seconds). Rule 2: Warehouse RFID records lack bill of lading anchoring (timestamp matching but no container number binding). 3) Continuity scoring and response Herwitz score: 0.42 (due to bill of lading time conflict, Δd=0 but Δt=180 seconds) Level 3 response execution: Frozen the bank accounts of two recipients The generated judicial evidence package includes: Dual Bill of Lading Transfer Records (Timestamp Sequence); RFID entry evidence quantum fingerprint e9cdc0388c; Cargo entropy analysis report (entropy value 0.78 > 0.7 threshold).

[0134] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A real-time blocking system for maritime cargo title fraud based on physical event anchoring and spatiotemporal decoupling, characterized in that, Comprising: Physical Event Four-Dimensional Tensor Construction Module: Used to construct a physical event characterized by a four-dimensional data tensor containing spatial coordinates, time series, device signatures, and event types based on the three-axis acceleration waveform of the container intelligent lock, the longitude and latitude coordinate sequence of the ship AIS, the digital signature image of the customs inspection equipment, and the geofence timestamp of the warehouse RFID; Quantum Fingerprint Anchoring Module: Used to generate a non-replicable unique identifier for the physical event data generated by the Physical Event Four-Dimensional Tensor Construction Module through a 12-bit quantum circuit and anchor it to the blockchain; Hurwitz Spacetime Topology Engine: Used to perform a continuity scoring calculation for generating a non-replicable unique identifier for the physical event generated by the Physical Event Four-Dimensional Tensor Construction Module and the anti-counterfeiting identifier provided by the Quantum Fingerprint Anchoring Module; Zero-Knowledge Cross-Verification Module: Used to implement trusted verification of physical events under privacy protection using the zero-knowledge proof method based on the physical events generated by the Physical Event Four-Dimensional Tensor Construction Module; Dynamic Decoupling Rule Engine: Based on the physical events generated by the Physical Event Four-Dimensional Tensor Construction Module and the obtained bill of lading flow, as well as the trusted verification transmitted by the Zero-Knowledge Cross-Verification Module, it compares the bill of lading flow with the logistics events in real time through the built-in fraud rule library and identifies fraud patterns based on preset rules; Three-Level Risk Response Execution Module: Based on the continuous score transmitted by the Hurwitz Spacetime Topology Engine and the fraud pattern transmitted by the Dynamic Decoupling Rule Engine, it automatically triggers an on-chain blocking operation to make a judgment on the three-level risk response; Self-Evolving Risk Control System: Using the continuous score transmitted by the Hurwitz Spacetime Topology Engine, it dynamically optimizes the rule weights and the cargo right entropy value model through fraud pattern clustering to generate a defense mechanism for triggering an on-chain blocking operation.

2. The real-time blocking system for maritime cargo ownership fraud based on physical event anchoring and spatiotemporal decoupling as described in claim 1, characterized in that, The Dynamic Decoupling Rule Engine includes four core rules: spacetime conflict detection, path deviation detection, multi-party statement conflict detection, and physical-document disconnection detection, which are specifically as follows: Spacetime Conflict Detection: When the bill of lading transfer timestamp tb and the container opening timestamp to satisfy 0 < to - tb < 1800 seconds, a warning of cargo right theft is triggered; Path deviation detection: Real-time calculation of the Havelsing distance between the vessel's position and the declared route. ,when Ghost ship blocking is triggered when the distance exceeds 200 nautical miles for more than 1 hour. Multi-Party Statement Conflict Detection: Retrieve the ownership statement records of the same container on the blockchain. When there are multiple valid bills of lading, a judicial evidence package for selling the same cargo to multiple parties is generated; Physical-Document Disconnection Detection: When the warehouse RFID entry timestamp has no corresponding bill of lading data on the blockchain, it is marked as a high-risk event of empty bill of lading pledge; The determination of the dynamic optimization algorithm for the weights of the four rules of spacetime conflict detection, path deviation detection, multi-party statement conflict detection, and physical-document disconnection detection is as follows: Where: FP is the false alarm rate of the rule. For learning rate, This represents the current weight value of the i-th rule. This represents the updated weight value of the i-th rule, which is the result of dynamic optimization.

3. The real-time blocking system for maritime cargo title fraud based on physical event anchoring and spatiotemporal decoupling as described in claim 1, characterized in that, The process of constructing a physical event characterized by a four-dimensional data tensor containing spatial coordinates, time series, device signatures, and event types based on the three-axis acceleration waveform of the container intelligent lock, the longitude and latitude coordinate sequence of the ship AIS, the digital signature image of the customs inspection equipment, and the geofence timestamp of the warehouse RFID is as follows: S11: Perform a joint time-frequency analysis on the container intelligent lock waveform: In the time domain, acceleration amplitude mutation detection is used. When the waveform peak value > 1.5g, a violent unlocking event is marked; In the frequency domain, the fast Fourier transform is used to extract the main frequency component. When the main frequency offset rate > 15%, illegal interference is marked. S12: The Haversine distance model is used for ship trajectory anomaly detection: 1、 in: Latitude R is the longitude, and R is the Earth's radius. When three consecutive points are... A course deviation warning is triggered when the distance exceeds 200 nautical miles. R is the spherical distance between two geographical locations, where R is the Earth's radius. The difference in latitude between two points. The difference in longitude between the two points. The latitude value representing the starting point Indicates the latitude value of the endpoint. S13: Establish a space-time consistency verification mechanism for the customs inspection timestamp tc and the smart lock operation timestamp tl: When |tc - tl| < n seconds and the container numbers are the same, it is determined that the verification is passed.

4. The real-time blocking system for maritime cargo ownership fraud based on physical event anchoring and spatiotemporal decoupling as described in claim 1, characterized in that, The quantum fingerprint anchoring module includes: Quantum fingerprint generation unit: Used to calculate a unique quantum fingerprint identifier that cannot be replicated for each original physical event; Multi-source space-time alignment unit: Used to receive the event data with fingerprint identifiers transmitted by the quantum fingerprint generation unit, verify whether they belong to the same coherent physical process by checking whether the timestamp deviations of multiple related events are within the threshold, and output a group of associated events that pass the space-time consistency verification; Blockchain evidence storage unit: Used to construct a causal chain with a Merkle root based on the group of associated events transmitted by the multi-source space-time alignment unit in chronological order, and package it together with the previous block hash, quantum fingerprint, and digital signature to form an immutable permanent blockchain evidence storage; The quantum fingerprint generation unit: Performs the following operations through a quantum circuit: S21: After hashing the original event data, it is intercepted as a 48-bit classical binary string and output as an initialization parameter; S22: Based on the binary string, apply the X gate to the corresponding quantum bits to encode the classical information into a definite initial quantum state; S23: Apply the Hadamard gate to the initial quantum state to convert the initial quantum state after applying the Hadamard gate into a uniform superposition state, and output a quantum superposition state containing all possibilities; S24: Apply the CX entanglement gate to the quantum superposition state to cause quantum entanglement between the bits, and output a highly correlated entangled state; S25: Finally, perform multiple measurements on this entangled state to collapse the quantum information into a classical probability distribution, and output the classical bit sequence with the highest probability as the final fingerprint identifier; The execution process of the blockchain evidence storage unit is as follows: The complete block data to be signed is jointly composed of the event causal chain, the previous hash, and the quantum fingerprint, and then a digital signature covering these data is generated using the private key.

5. A real-time blocking system for maritime cargo title fraud based on physical event anchoring and spatiotemporal decoupling as described in claim 1, characterized in that... : Used for the physical events generated by the physical event four-dimensional tensor construction module, and the zero-knowledge proof method is used to achieve trusted verification of physical events under privacy protection; S51: Generate a customs inspection certificate. The specific process is as follows: S511: Obtain the container inspection photos; S512: Based on the container inspection photos, extract the box number hash value and the geographical fence location stamp; S513: Based on the hash value and the geographical fence location stamp, construct a zk-SNARK circuit verification statement; S52: Bill of lading advance trading detection: When the cargo right transfer timestamp to and the container operation timestamp tc satisfy |to - tc| > 1800 seconds, a fraud evidence vector is generated.

6. A real-time blocking system for maritime cargo title fraud based on physical event anchoring and spatiotemporal decoupling as described in claim 1, characterized in that, Hurwitz space-time topology engine: Used to perform the process of calculating the continuity score of generating a unique anti-copying identifier for the physical events generated by the physical event four-dimensional tensor construction module and the anti-counterfeiting identifier provided by the quantum fingerprint anchoring module as follows: 5.1 Relativistic Time Dilation Correction: Introducing the Speed ​​of Light Calculate the time dilation factor corresponding to the velocity v: Corrected time difference ; 5.2 Continuity break determination: The score is calculated by the mathematical formula of the Hurwitz spatiotemporal topology model. When the score value is <0.8, a three-dimensional alarm vector containing the break location, time deviation and velocity factor is generated.

7. A real-time blocking system for maritime cargo ownership fraud based on physical event anchoring and spatiotemporal decoupling as described in claim 1, characterized in that, The three-tier risk classification includes a three-tier response, a two-tier response, and a one-tier response, as detailed below: Level 3 response: Continuity score < 0.

5. When a response is classified as Level 3, perform the following actions: Freezing the flow of funds in relevant bank accounts or suspending the transfer of bills of lading on the blockchain network; or generating a judicial evidence package that complies with the Hague-Visby Rules. The generated judicial evidence package conforming to the Hague-Visby Rules includes: a blockchain transaction ID Merkle path, a timestamped event sequence causal chain, and an ECDSA signature verification tree; Level 2 response: 0.5 ≤ continuous score < 0.8; When judged as a Level 2 response, perform the following operations: restrict bill of lading transfer operation permissions or initiate a multi-party video verification mechanism or generate a risk warning evidence chain; Level 1 Response: Consecutive scores ≥ 0.8: When a response is determined to be Level 1, perform the following actions: Send AES-256 encryption risk warnings to regulators and update historical pattern records in the risk knowledge base.

8. A real-time blocking system for maritime cargo title fraud based on physical event anchoring and spatiotemporal decoupling as described in claim 1, characterized in that, The self-evolving risk control system uses continuous scores transmitted by the Hurwitz spatiotemporal topology engine to dynamically optimize rule weights and asset entropy models through fraud pattern clustering, generating a defense mechanism that triggers on-chain blocking operations. The process is as follows: S71: Calculate the entropy value of cargo ownership based on the dynamic model of cargo ownership entropy value; S72: Based on the entropy value of goods ownership, a fraud pattern clustering engine is adopted, and the historical event feature vector is used for cluster analysis to discover fraud patterns and generate defense rules. The feature vector includes continuous scores, entropy values ​​of goods ownership, and rule weights. S73: Utilize the rule base evolution mechanism to optimize the weights and expand the rules of the rule base, thereby completing the system evolution. The fraud pattern clustering engine is implemented as follows: (1) Extract the historical event feature vector X = [continuity score, cargo entropy value, rule weight]; (2) Identify new fraud patterns using the DBSCAN clustering algorithm; (3) When ghost container fraud is detected, generate defense rules; The expression for the dynamic model of cargo ownership entropy is as follows: Where: pk is the probability of the cargo's safe state. is the standard deviation of the continuous score; wi is the event type weight; ft is the time decay adjustment coefficient for spatial dispersion. The system automatically performs weight optimization and rule expansion after processing every 100 events.

9. A real-time blocking system for maritime cargo title fraud based on physical event anchoring and spatiotemporal decoupling as described in claim 1, characterized in that, It also includes a thought-decision output module: Connect to the Hurwitz spatiotemporal topology engine: Receive continuous scores as core input data for building risk visualizations and assessing global ownership entropy values; Connect to the dynamic decoupled rule engine: Receive the specific triggered fraud rule type, rule ID, and judgment details, which are used to mark abnormal relationships in the knowledge graph and generate rule trigger decomposition paths in the auditable evidence chain; Connect to the self-evolving risk control system: Receive updated cargo ownership entropy value model parameters and rule weight changes, which are used to refresh the risk heat map and display the system's own evolution timeline; Connect with quantum fingerprint anchoring and blockchain evidence storage modules: call on transaction records, quantum fingerprints and Merkel paths on the blockchain to provide underlying tamper-proof data traceability for auditable evidence chains; Connect to the physical event four-dimensional tensor construction module: Obtain the original spatiotemporal event sequence as the basic data for constructing nodes and edges in the spatiotemporal knowledge graph.

10. A real-time blocking method for maritime cargo title fraud based on physical event anchoring and spatiotemporal decoupling, characterized in that, The following processing steps are included: Real-time acquisition phase: Container acceleration waveforms, ship trajectory coordinates, customs digital signatures, and RFID timestamps are synchronously acquired through an IoT device network to construct a four-dimensional data tensor containing spatial coordinates, time series, device signatures, and event types; Quantum anchoring phase: Generate an event fingerprint containing device signature, GPS coordinates and millisecond-level timestamp, generate a copy-resistant identifier through quantum circuits and write it into the blockchain; The spatiotemporal decoupling phase is carried out as follows: (1) The continuity score is calculated using the Hurwitz model; (2) Dynamic rule matching is performed: when a bill of lading transfer is detected, there is illegal unlocking, route deviation > 200 nautical miles or multiple bill of lading conflicts, a high-risk response is triggered. Zero-knowledge verification stage: Generate zk-SNARK proofs for customs inspection evidence and verify spatiotemporal consistency; Conduct risk prevention and control assessment and implementation: Level 3 Response: Freeze Account + Suspend Bill of Lading Processing + Generate Legal Evidence Package Level 2 Response: Initiate video verification + restrict bill of lading operations Level 1 Response: Send encrypted alert + update knowledge base; Evolutionary stage: Historical events are analyzed through DBSCAN clustering. When the false alarm rate of a rule is less than 5%, the weight is increased by 0.05.

Images