Multi-scene adaptive internet of things device data trusted on-chain computing power adaptive adjustment system

Abstract

The application relates to a power self-adaptive adjustment system for multi-scene adaptive Internet of Things equipment data trusted on-chain, comprising a security quantization module, which is used for collecting total power data and potential power data of an attacker of a blockchain Internet of Things network in real time, establishing real-time mapping of total power and attack success probability, and generating a security evaluation report; a dynamic incentive coordination module, which is used for obtaining a workshop server blockchain security reward budget based on the evaluation result of the security quantization module; and an equipment power adaptation module, which is used for calling a complete information power optimizer or a non-complete information reinforcement learning adapter according to whether the current is a complete information scene after receiving the workshop server blockchain security reward budget, and determining equipment power consumption decisions in this game iteration. The application proposes a safe and quantifiable Internet of Things blockchain architecture, aiming to solve the problem that heterogeneous Internet of Things devices with limited power resources participate in blockchain consensus, and optimize the balance between power distribution and network security.

Description

Technical Field

[0001] This invention relates to the field of Internet of Things (IoT) technology, and in particular to a computing power adaptive adjustment system for the trusted on-chain data storage of IoT devices that is adaptable to multiple scenarios. Background Technology

[0002] The Internet of Things (IoT) technology, through embedded sensors, communication modules, and data processing units, deeply integrates the physical and digital worlds, and has already achieved large-scale applications in several key fields. In Industry 4.0 scenarios, tens of thousands of vibration sensors monitor the status of factory equipment in real time, reducing downtime through predictive maintenance; in smart agriculture, soil moisture sensors work in conjunction with weather stations to optimize irrigation systems, significantly improving water resource utilization; and for cold chain logistics, temperature and humidity tracking devices ensure that medicines and fresh produce meet safety standards throughout the transportation process. The reliability of these applications directly depends on the authenticity, completeness, and timeliness of the data generated by terminal devices. However, current IoT architectures lack effective solutions to the centralized trust crisis. Most IoT systems rely on a central cloud platform or local gateway to aggregate data. For example, in a factory, equipment sensors need to aggregate data to a single central server. Once this server is attacked by an APT (such as by injecting malicious firmware through vulnerabilities), attackers can tamper with equipment alarm thresholds, mask bearing overheating faults, and cause damage to the entire production line. Similarly, in the medical cold chain scenario, if the temperature records in the central database are tampered with by internal personnel, problematic vaccines could enter the market and become untraceable. The central node becomes the source of systemic risk.

[0003] To address the trust impasse, blockchain technology has been introduced into the Internet of Things (IoT) field. Its core advantage lies in its decentralized trust infrastructure. (1) Distributed ledger mechanism: After data consensus verification, it is stored on all network nodes. Any node must control more than 51% of the computing power to tamper with the historical records; (2) Hash chain structure: Each block contains the hash value of the previous block, forming an unbreakable verification chain; (3) Smart contract automation: Automatically execute operations when preset conditions are met.

[0004] A typical application is cold chain pharmaceutical traceability. From temperature control records in the production workshop and GPS tracks of logistics vehicles to the entry time of hospital cold storage, all data is hashed and written to the blockchain. Any attempt to tamper with the data at any stage will cause subsequent block hash verification to fail, achieving trusted traceability throughout the entire chain.

[0005] While blockchain and the Internet of Things (IoT) are theoretically highly compatible, their mainstream consensus mechanisms and the characteristics of terminal hardware are somewhat incompatible. Mainstream blockchains rely on consensus mechanisms such as Proof-of-Work (PoW) to ensure security, but PoW requires nodes to continuously perform high-intensity hash calculations, while the computing power of typical industrial IoT devices is far lower than that of devices dedicated to blockchain. Forcing IoT devices to participate in PoW would rapidly deplete local resources, such as electricity. On the other hand, full blockchain nodes need to store hundreds of gigabytes of ledger data, while IoT devices have limited storage capacity and cannot handle gigabyte-level data. Furthermore, blockchain synchronization requires continuous bandwidth, but IoT devices have low peak bandwidth, which cannot support real-time block propagation.

[0006] To address the challenges of insufficient hardware performance when combining blockchain with the Internet of Things (IoT), existing solutions typically establish proxy nodes (such as workshop servers) in each factory workshop to handle storage and communication functions. IoT devices within the workshop access the blockchain network through these proxy nodes, and the computing power used for proof-of-work is usually provided by edge servers leased by the factory. However, existing blockchain-IoT solutions lack effective mechanisms to regulate the computing power consumption of IoT devices, resulting in low system efficiency. Furthermore, establishing a quantitative mapping between computing power investment and blockchain network security is difficult, and there is a lack of precise methods for assessing network security status, making it impossible to accurately measure the effectiveness of different computing power allocation strategies in resisting attacks (such as 51% attacks or double-spending attacks). Due to the heterogeneity of IoT devices (such as differences in sensor type and data acquisition intervals), the importance of the data collected by different IoT devices varies, and the data security improvement brought by consuming unit resources differs. Failure to adaptively and individually allocate computing power to each IoT device will further reduce system efficiency. Finally, when designing an adaptive IoT device computing power consumption regulation mechanism, it is necessary to consider two information scenarios: complete information scenarios and incomplete information scenarios. In a fully informed scenario, each IoT device is aware of the other's computing power consumption. However, in an incompletely informed scenario, communication between IoT devices is difficult, and for a single IoT device, various information from other IoT devices is unavailable. Existing computing power consumption adjustment mechanisms typically assume a uniform data importance or computing power allocation strategy for IoT devices and generally handle fully informed scenarios. They are incompatible with incompletely informed scenarios, struggle to adapt to diverse device needs, and limit the system's practicality and scalability.

[0007] Currently, existing mechanisms for adjusting edge computing resource consumption by devices in blockchain-based IoT can be broadly categorized into centralized computing power adjustment and passive computing power adjustment. A brief introduction to each of these technologies follows.

[0008] (1) Centralized computing power adjustment: This mechanism relies on a centralized decision-making node (such as a blockchain platform or a specific edge server) to allocate global computing power. The central node first collects static information of all IoT devices in the workshop / area (such as device type and preset data importance tags), and then allocates a fixed computing power quota to each device according to preset, fixed rules (such as static weight tables and priority lists). Finally, all devices are forced to consume edge server resources to participate in blockchain consensus (such as proof-of-work) according to the allocated quota. However, the "one-size-fits-all" static quota allocation cannot perceive and respond to real-time changes in the importance of device data. This results in computing power resources not being able to be dynamically adjusted as needed. Critical devices may face data uploading delays or security degradation risks due to insufficient quotas, while non-critical devices may occupy redundant resources for a long time. In addition, the adjustment rules are fixed and lack dynamic optimization capabilities. When the network scale expands (a large number of new devices are added) or the topology changes (devices are taken offline), the weight table needs to be manually reconfigured, resulting in poor scalability. The existence of the central node introduces the risk of single point of failure. Once it fails, the entire adjustment mechanism will be paralyzed. Finally, this mechanism itself does not rely on information exchange between devices, but due to its centralized nature, it does not consider and cannot adapt to the incomplete information needs or device autonomy that may exist in a distributed environment.

[0009] (2) Passive Adjustment Mechanism: Under this mechanism, the blockchain platform broadcasts a unified reward signal (representing the reward for participating in block generation) to all IoT devices. Each IoT device, based on this unified reward and its own perception of the importance of the data (usually subjective or static), independently determines its computing power consumption without considering the behavior of other devices. In a fully informed scenario, devices tend to overinvest in computing power to compete for the limited unified reward. This competitive behavior leads to a surge in overall computing power consumption while the network security gain (or data security improvement) per unit of computing power decreases, damaging the overall utility and long-term security of the blockchain platform. In the more common incomplete information scenario, devices can only make isolated decisions based on their own information and the rewards broadcast by the platform. This lack of a global perspective makes decisions prone to deviating from the optimal state of the system, causing resource misallocation, resulting in insufficient investment in high-importance devices and excessive investment in low-importance devices. Regardless of the information scenario, using a unified reward cannot reflect the true differences in data importance between devices. This leads to unfair and inefficient resource allocation.

[0010] In summary, designing a computing power consumption adjustment mechanism for IoT devices within a blockchain-based IoT framework still faces three main challenges: First, the lack of quantitative modeling for computing power versus security. Existing mechanisms generally fail to establish a precise and dynamic mapping relationship between the edge computing power resources consumed by devices and the overall security of the blockchain network. The lack of a scientific evaluation model for the probability of successful attacks makes it impossible to quantify the actual security gains of different computing power allocation strategies, resulting in a lack of clear security objectives for computing power adjustment. Second, the conflict between device heterogeneity and personalized needs. IoT devices exhibit significant differences in data importance. Existing solutions typically ignore this heterogeneity, making it difficult to achieve fair and efficient personalized computing power resource allocation and maximizing the overall security benefits per unit of computing power consumption. Third, the lack of adaptability across information scenarios. Real-world IoT networks are complex and dynamic, and the communication conditions between devices determine information accessibility. Existing mechanisms often presuppose fully information scenarios, relying on information sharing between devices to achieve effective strategic interaction. However, in scenarios where devices are widely distributed, communication is limited, or privacy concerns exist, incomplete information is the norm. Existing technologies lack effective mechanisms to adapt to this information asymmetric environment, resulting in the failure or inefficiency of adjustment mechanisms in actual deployment. Summary of the Invention

[0011] This invention addresses the aforementioned problems by proposing a secure and quantifiable IoT blockchain architecture. It aims to solve the challenge of enabling heterogeneous IoT devices with limited computing resources to participate in blockchain consensus and optimize the balance between computing power allocation and network security. By establishing a direct mapping relationship between computing power input and blockchain network security, the network security status becomes measurable and assessable. This invention designs a multi-scenario adaptable computing power adaptive adjustment system for the trusted on-chain storage of IoT device data. This system allows IoT devices to proactively adjust the edge server computing resources they consume, fully considering the heterogeneity of devices and ensuring fairer and more efficient consumption of computing resources. Furthermore, this invention provides a precise method for calculating the probability of successful attacks, providing a scientific basis for assessing the network security status under different computing power levels. These designs not only solve the bottleneck of insufficient computing power in IoT devices but also improve the security and practicality of blockchain IoT systems.

[0012] To achieve the above objectives, the present invention provides the following solution: A computing power adaptive adjustment system for trusted on-chain data transmission from IoT devices, adaptable to multiple scenarios, including: The security quantification module is used to collect the total computing power data of the blockchain IoT network and the potential computing power data of attackers in real time, establish a real-time mapping between total computing power and the probability of successful attack, and generate a security assessment report. The dynamic incentive coordination module is used to obtain the workshop server blockchain security reward budget based on the evaluation results of the security quantification module. The device computing power adaptation module is used to determine the device computing power consumption decision in this game iteration by calling the full information computing power optimizer or the incomplete information reinforcement learning adapter based on whether the current scenario is fully informational.

[0013] Optionally, the security quantification module includes: The data acquisition unit is used to collect the total computing power contributed by all IoT devices through the edge server in real time. ; The monitoring unit is used to monitor and estimate the computing power of potential attackers. ; The first calculation unit is used to calculate based on the total computing power value. and the computing power of the potential attacker Calculate the probability that an IoT device will find the next block. And the probability of an attacker finding the next block ; The second calculation unit is used to determine the probability of finding the next block based on the IoT device. And the probability of an attacker finding the next block Calculate the attacker's The probability of successfully finding the next block after a certain number of blocks. ; Integration unit, used to integrate attackers'... The probability of successfully finding the next block after a certain number of blocks. Total computing power potential attacker computing power The system generates a structured security posture report, including timestamps.

[0014] Optionally, the attacker in The probability of successfully finding the next block after a certain number of blocks. for: ; ; in, Incomplete regularization function.

[0015] Optionally, the dynamic incentive coordination module is used to obtain the workshop server blockchain security reward budget based on the evaluation results of the security quantification module, including: Step 2.1: Based on the evaluation results of the security quantification module, extract the attacker's... The probability of successfully finding the next block after a certain number of blocks. ; Step 2.2: If this is the first game iteration, then initialize randomly. If the condition is met, proceed to step 2.6; otherwise, proceed to step 2.3. Rewards for blockchain security; Step 2.3: Receive computing power consumption decisions sent by the device computing power adaptation module of each IoT device; Step 2.4: Count the number of IoT devices with a non-zero computing power contribution. The workshop server blockchain security reward budget is calculated using the first preset formula; Step 2.5: If the workshop server blockchain security reward budget calculated in this game iteration is consistent with that in the previous game iteration, then stop the iteration; Step 2.6: Broadcast the calculated workshop server blockchain security reward budget to all IoT devices.

[0016] Optionally, the first preset formula is: in, Budget for blockchain security rewards for workshop servers. To control the constant parameters of the total computing power, To control the constant parameter of the blockchain security evaluation function of the workshop server, To incentivize the budget, For IoT device data, To assist in calculating variables, i For the i-th IoT device, The importance of the i-th IoT device.

[0017] Optionally, the device computing power adaptation module, after receiving the blockchain security reward budget from the workshop server, determines the device computing power consumption decision in this game iteration by calling the full-information computing power optimizer or the incomplete-information reinforcement learning adapter based on whether the current scenario is one of complete information. This includes: Step 3.1: Receive the workshop server blockchain security reward budget sent by the dynamic incentive coordination module; Step 3.2: Determine if the current scenario is one with complete information. If yes, proceed to step 3.3; otherwise, proceed to step 3.7. Step 3.3: Devices exchange their computing power consumption decisions with each other via the local network; Step 3.4: Sort all IoT devices in ascending order of data importance, and select the top two sorted IoT devices into the set of devices with non-zero computing power contribution; initialize the iteration parameters. ; Step 3.5: When Less than the total number of IoT devices and At that time, the first Add an IoT device to the set of devices whose computing power contribution is not zero, and make... ;in, The importance of the (m+1)th IoT device The importance of the i-th IoT device; Step 3.6: Calculate the optimal computing power consumption decision for each IoT device in the set of devices with a non-zero computing power contribution according to the second preset formula. Then proceed to step 3.9; Step 3.7: Collect environmental information that the device can sense; Step 3.8: Construct a Markov process for solving the optimal computing power consumption decision. Use the Q-learning algorithm to learn the optimal computing power consumption strategy through interaction with the environment, and approximate the optimal computing power consumption strategy. Then proceed to step 3.9; Step 3.9: If the optimal computing power consumption decision for IoT devices calculated in this game iteration is consistent with that in the previous game iteration, then stop the iteration; Step 3.10: Calculate the first... The optimal computing power consumption decision for each IoT device is sent to the dynamic incentive coordination module.

[0018] Optionally, the second preset formula is: in, For optimal computing power consumption strategy, The server's current reward strategy, There are a total of N IoT devices. For the j-th IoT device, This is the decision on computing power consumption for the i-th IoT device.

[0019] Optionally, the environmental information that the device can sense includes: historical block generation success rate, macro-level security indicators broadcast by the blockchain platform, and edge server computing load.

[0020] The beneficial effects of this invention are as follows: This invention achieves the following breakthrough effects in blockchain IoT through a computing power adaptive adjustment system for the reliable on-chain data storage of IoT devices that adapts to multiple scenarios: (1) Quantitative security-driven decision-making: Dynamic security assessment based on the attack success probability model, which strongly links computing power allocation with network security.

[0021] (2) Precisely counter the threat of resource heterogeneity: The differentiated computing power allocation mechanism obtains redundant security barriers, avoiding the weak points of defense caused by the "one-size-fits-all" approach in traditional solutions.

[0022] (3) Efficient collaboration in fully information scenarios: By balancing the computing power strategies among devices, the efficiency of sensor computing power allocation in the workshop is improved.

[0023] (4) Intelligent decision-making in scenarios with incomplete information: The reinforcement learning module enables mobile devices to approach the global optimal solution even when information is isolated, solving the problem of cross-regional device collaboration.

[0024] (5) Closed-loop dynamic optimization: The closed-loop process of security monitoring → incentive adjustment → computing power response enables the system to maintain stability in dynamic environments such as attack fluctuations and equipment additions and subtractions.

[0025] (6) Maximize edge computing power utilization: By differentiating the importance of IoT device data, guide computing power to tilt towards high-value data and improve the security gain per unit of computing power.

[0026] (7) Low threshold to empower massive IoT devices: Resource-constrained devices can apply for edge computing power to participate in proof-of-work consensus, breaking through computing power bottlenecks and reducing data upload delay. Attached Figure Description

[0027] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0028] Figure 1 A schematic diagram illustrating the logical relationships between modules in the computing power adaptive adjustment system architecture for trusted on-chain data upload of IoT devices in multiple scenarios, as shown in this embodiment of the invention: Figure 2 This is a flowchart illustrating the implementation process of a computing power adaptive adjustment system for trusted on-chain data storage of IoT devices that adapts to multiple scenarios, as described in this invention. Detailed Implementation

[0029] 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. 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.

[0030] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0031] This embodiment proposes a computing power adaptive adjustment system for trusted on-chain data from IoT devices that adapts to multiple scenarios, including: The security quantification module is used to collect the total computing power data of the blockchain IoT network and the potential computing power data of attackers in real time, establish a real-time mapping between total computing power and the probability of successful attack, and generate a security assessment report. The dynamic incentive coordination module is used to obtain the workshop server blockchain security reward budget based on the evaluation results of the security quantification module. The device computing power adaptation module is used to determine the device computing power consumption decision in this game iteration by calling the full information computing power optimizer or the incomplete information reinforcement learning adapter based on whether the current scenario is fully informational.

[0032] Specifically, in this embodiment, to address the core technical bottleneck of uneven computing power consumption of blockchain IoT devices across multiple scenarios, this invention proposes a multi-scenario adaptable computing power adaptive adjustment system for trusted on-chain data upload of IoT devices. This system mainly comprises the following three modules, with the logical relationships between the modules as follows: Figure 1 As shown: (1) Security Quantification Module: This module serves as the system's security situation awareness center, deployed at the blockchain network platform layer. Its core function is to monitor the overall computing power (total computing power) of the entire blockchain network in real time, as well as the computing power that potential attackers may possess. Based on this real-time data, this module dynamically assesses the current level of security risks faced by the network, particularly quantifying the probability of an attacker successfully launching an attack (such as attempting to tamper with blockchain data). Ultimately, it generates a comprehensive assessment report reflecting the real-time security status of the network, providing crucial security basis for subsequent incentive decisions.

[0033] (2) Dynamic Incentive Coordination Module: Also deployed on the workshop server, this module serves as the decision-making center for resource optimization and allocation. It receives security assessment reports from the security quantification module. Based on the level of network security vulnerability reflected in the report (e.g., higher attack risk), this module dynamically calculates and sets the total reward that the blockchain platform is willing to pay to IoT devices participating in maintaining network security. Its main goal is to improve blockchain security as much as possible while meeting the platform's security requirements through adaptive reward adjustments. The calculated optimal reward budget will be broadcast to all connected IoT devices.

[0034] (3) Device Computing Power Adaptation Module: This module is deployed on the device side and is the execution unit for IoT devices to respond to platform incentives and make personalized resource consumption decisions. It includes a fully information computing power optimizer and a partially information reinforcement learning adapter. After receiving the reward broadcast by the platform, this module selects the optimal strategy to determine how much edge computing resources to consume to participate in blockchain maintenance work based on the device's own characteristics (such as the criticality of the task undertaken and the cost of acquiring edge computing resources) and the current network communication conditions (whether information is reachable). It includes two core decision-making modes: one is suitable for scenarios where devices can fully share information and perform collaborative optimization; the other is designed for scenarios where information is isolated between devices and communication is limited, and uses intelligent learning capabilities to make independent decisions. The device finally applies for and consumes computing resources from the edge server based on the decision result.

[0035] This invention proposes a computing power adaptive adjustment system for trusted on-chain data upload of IoT devices in multiple scenarios. To make the technical implementation scheme and effects of this invention clearer and more explicit, refer to... Figure 2 The invention will be further described in detail with reference to specific examples. Figure 2 It includes a security quantification module, a dynamic incentive coordination module, and a device computing power adaptation module. The following sections provide detailed explanations of each of these modules. To make the technical implementation scheme and effects of the present invention clearer and more explicit, this embodiment uses the scenario of bearing fault prediction and anti-tampering in a smart factory as an example for illustration.

[0036] A large manufacturing company has deployed hundreds of vibration sensors (such as the EM-300 model) on its critical production lines to monitor the operating status of high-value bearings in real time. The high-frequency vibration data generated by these sensors is used in predictive maintenance models to provide early warnings of potential faults (such as cracks and imbalances). Simultaneously, environmental temperature and humidity sensors (such as the TH-100 model) collect environmental parameters to aid in analysis. In this scenario, the following core issues are addressed: (1) Data credibility: Sensor data must be tamper-proof and uploaded to the blockchain to prevent hackers or insiders from tampering with alarm thresholds or original data to cover up the real fault.

[0037] (2) Resource constraints: Both EM-300 and TH-100 are resource-constrained devices (low computing power, limited battery), and cannot directly run proof-of-work block generation.

[0038] (3) Difference in data value: The value of bearing vibration data (EM-300) is much higher than that of environmental data (TH-100), requiring more safety resources to ensure its safety.

[0039] (4) Information scenario mixing: Sensors within the same workshop can share information through a local area network (complete information scenario); however, AGVs (carrying mobile sensors) that cross workshops / factory areas have difficulty communicating in real time (incomplete information scenario).

[0040] In this embodiment, a security quantification module and a dynamic incentive coordination module are deployed on the workshop server. All sensors and AGVs access the edge server to request computing power. Each sensor / AGV has a built-in device computing power adaptation module. The data importance coefficient of EM-300 is set to be significantly higher than that of TH-100. When an attacker attempts to penetrate the network, the security quantification module detects a decrease in total computing power and an increase in the probability of attack, generating a structured security posture report. The dynamic incentive coordination module receives the structured security posture report, calculates and broadcasts a significantly improved blockchain security reward, aiming to enable IoT devices to request more computing power to enhance security. When a moving AGV becomes aware of the real-time computing power of other AGVs in the workshop, EM-300 and TH-100 exchange information via the local area network. The device computing power adaptation module calls the full-information computing power optimizer, enabling the highly important EM-300 to obtain a higher computing power quota than TH-100. When a moving AGV is unaware of the real-time computing power of other AGVs within the workshop, the device computing power adaptation module activates an incomplete information reinforcement learning adapter. This adapter collects information such as a recent decrease in its own block generation success rate, the probability of being attacked broadcast by the platform, and the computing power load of the edge server. Based on a pre-trained Q-learning model, it outputs a computing power consumption decision according to the current state. All devices apply for and execute computing power through the edge server. Newly added computing power is incorporated into the total computing power. Once the security quantification module detects a significant increase in total computing power, it recalculates the probability of being attacked and generates a new structured security posture report. The dynamic incentive coordination module accordingly adjusts the blockchain security reward appropriately, initiating the next round of optimization.

[0041] (1) Security Quantification Module: This module collects the total computing power data of the blockchain IoT network and the potential computing power of attackers in real time, and constructs a mathematical mapping relationship between computing power and the probability of successful attack, providing a basis for network security assessment. The specific execution process of this module is as follows: Step 1: Real-time collection of the total computing power contributed by all IoT devices through the edge server (Calculated by hash rate).

[0042] Step 2: Monitor and estimate the computing power of potential attackers (By analyzing abnormal computing power fluctuations or external threat intelligence).

[0043] Step 3: Calculate the probability that the IoT device will find the next block. And the probability of an attacker finding the next block The calculation formula is as follows: Step 4: Calculate the attacker's... The probability of successfully finding the next block after 1 block: ; Among them, incomplete regularization function The expression for accurately calculating the probability distribution in complex scenarios is as follows: ; in, This is a gamma function. The higher the value, the more fragile the network.

[0044] Step 5: Integration The system generates a structured security situation report based on the timestamp and pushes the report to the dynamic incentive coordination module in real time.

[0045] This security quantification module establishes a relationship between the total computing power of the blockchain IoT and the probability of an attacker successfully attacking the network—as the total computing power increases, the probability of an attacker's success gradually decreases. Through this model, a real-time mapping between total computing power and the probability of successful attacks is established, generating a security assessment report and providing a decision-making basis for the dynamic incentive coordination module.

[0046] (2) Dynamic Incentive Coordination Module: Based on the security identification module, this module interacts with the device computing power adaptation module to dynamically iterate and determine the optimal blockchain security reward. This enables IoT devices to adaptively adjust the computing power resources they request from the edge server, thereby optimizing the computing power consumption of the edge server and improving the security of the blockchain network. After the workshop server provides its reward strategy, each IoT device determines its computing power consumption strategy according to the reward strategy. This process is repeated until game equilibrium is reached.

[0047] The game equilibrium is defined as follows: a workshop server cannot improve blockchain security simply by changing its own reward decisions, because if a workshop server changes its reward decisions, IoT devices will also change their computing power consumption strategies. Similarly, IoT devices cannot improve their chances of obtaining blockchain security rewards simply by changing their own computing power consumption decisions.

[0048] With Taking a workshop with a single IoT device as an example, the specific blockchain security assessment function of the workshop server is as follows: ; in Rewards for blockchain security (reward budget is) ,Right now ), It is an important constant parameter controlling the overall computing power. This is a constant parameter that controls the convergence rate of the blockchain security assessment function on the workshop server. The workshop server adjusts its reward strategy... To influence the computing power consumption strategies of various IoT devices This maximizes the evaluation function value.

[0049] The specific execution process of this module is as follows: Step 1: Analyze the structured security posture report pushed by the security quantification module and extract key indicators. (Probability of successful attack).

[0050] Step 2: If this is the first game iteration, then initialize randomly. If the above steps are not executed, proceed to step six; otherwise, proceed to step three.

[0051] Step 3: Receive computing power consumption decisions sent by the computing power adaptation modules of each IoT device.

[0052] Step 4: Count the number of IoT devices whose computing power contribution is not zero. The platform calculates the optimal reward using the following formula: in, The workshop server's dynamic reward adjustment mechanism enables it to flexibly allocate resources based on security needs and equipment participation. (Formula) In order to make the formula The optimal solution that maximizes the value.

[0053] Step 5: If the optimal reward decision for the workshop server calculated in this game iteration is consistent with that of the previous game iteration, then stop the iteration.

[0054] Step Six: Calculate the... (Workshop server blockchain security reward budget) is broadcast to all IoT devices.

[0055] (3) Device Computing Power Adaptation Module: After receiving the optimal reward decision for the workshop server from the dynamic incentive coordination module, this module determines the device computing power consumption decision in this game iteration by calling the complete information computing power optimizer or the incomplete information reinforcement learning adapter, depending on whether the current scenario is one of complete information. The blockchain reward function for each IoT device is as follows: The first item represents the blockchain security reward that the device can obtain for contributing computing power, and the second item represents the computing power consumption of the device. Indicates the first The importance of data from each device. The specific execution flow of this module is as follows: Step 1: Receive the workshop server blockchain security reward decision sent by the dynamic incentive coordination module.

[0056] Step 2: Determine if the current scenario is one with complete information. If so, proceed to Step 3; otherwise, proceed to Step 7.

[0057] Step 3: Devices exchange computing power consumption decisions with each other through a local network (such as a workshop LAN).

[0058] Step 4: Sort all IoT devices in ascending order of data importance, and select the top two IoT devices from the sorted list to include the set of devices with a non-zero computing power contribution. Initialize iteration parameters. .

[0059] Step 5: When Less than the total number of IoT devices and At that time, the first Add an IoT device to the set of devices whose computing power contribution is not zero, and make... .

[0060] Step Six: Calculate the optimal computing power contribution for each IoT device in the set of devices with a non-zero computing power contribution according to the following formula. Then proceed to step nine.

[0061] ; Step 7: Collect environmental information that the device can perceive: historical block generation success rate, macro security indicators broadcast by the blockchain platform, and edge server computing load.

[0062] Step 8: Construct a Markov process to solve the optimal computing power consumption decision, using the Q-learning algorithm to learn the optimal computing power consumption strategy through interaction with the environment. Approximate the optimal computing power consumption strategy. (No need to know the policies of other devices), and proceed to step nine.

[0063] Step 9: If the optimal computing power consumption decision for IoT devices calculated in this game iteration is consistent with that in the previous game iteration, then stop the iteration.

[0064] Step 10: Calculate the... (No. The optimal computing power consumption decision for each IoT device is sent to the dynamic incentive coordination module.

[0065] This embodiment achieves the following breakthrough effects in blockchain IoT through a computing power adaptive adjustment system that enables trusted on-chain data storage for IoT devices adapted to multiple scenarios: (1) Quantitative security-driven decision-making: Dynamic security assessment based on the attack success probability model, which strongly links computing power allocation with network security.

[0066] (2) Precisely counter the threat of resource heterogeneity: The differentiated computing power allocation mechanism obtains redundant security barriers, avoiding the weak points of defense caused by the "one-size-fits-all" approach in traditional solutions.

[0067] (3) Efficient collaboration in fully information scenarios: By balancing the computing power strategies among devices, the efficiency of sensor computing power allocation in the workshop is improved.

[0068] (4) Intelligent decision-making in scenarios with incomplete information: The reinforcement learning module enables mobile devices to approach the global optimal solution even when information is isolated, solving the problem of cross-regional device collaboration.

[0069] (5) Closed-loop dynamic optimization: The closed-loop process of security monitoring → incentive adjustment → computing power response enables the system to maintain stability in dynamic environments such as attack fluctuations and equipment additions and subtractions.

[0070] (6) Maximize edge computing power utilization: By differentiating the importance of IoT device data, guide computing power to tilt towards high-value data and improve the security gain per unit of computing power.

[0071] (7) Low threshold to empower massive IoT devices: Resource-constrained devices can apply for edge computing power to participate in proof-of-work consensus, breaking through computing power bottlenecks and reducing data upload delay.

[0072] The adaptive computing power adjustment system for reliable on-chain data storage of IoT devices, adaptable to multiple scenarios, is an effective method to improve the security and resource efficiency of distributed IoT data. It can also be widely applied to other blockchain integration scenarios that require consideration of data reliability, device heterogeneity, and dynamic environments. For example, in smart grids, dynamically adjusting the edge computing power input of power distribution terminal equipment ensures the real-time nature of electricity metering data tamper-proofing and electricity theft detection; in unmanned mining areas, it adapts to the mixed communication scenarios of open-pit mining trucks and underground sensors, ensuring highly reliable on-chain storage of safety monitoring data; in cross-border cold chain logistics, it dynamically allocates edge computing power based on differences in cargo value to achieve reliable end-to-end traceability of critical temperature-sensitive pharmaceuticals. Related practitioners can utilize the security quantification driving mechanism, multi-scenario decision architecture, and closed-loop incentive model of this invention to understand the basic principles and general processes of achieving reliable on-chain computing power adjustment systems for reliable on-chain data storage of IoT devices, thereby better solving related problems in their respective fields. Therefore, this invention is not limited to industrial IoT scenarios and can be used for broader conceptualization.

[0073] The embodiments described above are merely preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Various modifications and improvements made to the technical solutions of the present invention by those skilled in the art without departing from the spirit of the present invention should fall within the protection scope defined by the claims of the present invention.

Claims

1. A computing power adaptive adjustment system for trusted on-chain data transmission of IoT devices in multiple scenarios, characterized in that: include: The security quantification module is used to collect the total computing power data of the blockchain IoT network and the potential computing power data of attackers in real time, establish a real-time mapping between total computing power and the probability of successful attack, and generate a security assessment report. The dynamic incentive coordination module is used to obtain the workshop server blockchain security reward budget based on the evaluation results of the security quantification module. The device computing power adaptation module is used to determine the device computing power consumption decision in this game iteration by calling the full information computing power optimizer or the incomplete information reinforcement learning adapter based on whether the current scenario is a complete information scenario. The device computing power adaptation module, after receiving the blockchain security reward budget from the workshop server, determines the device computing power consumption decision in this game iteration by calling the full-information computing power optimizer or the incomplete-information reinforcement learning adapter based on whether the current scenario involves complete information. This includes: Step 3.1: Receive the workshop server blockchain security reward budget sent by the dynamic incentive coordination module; Step 3.2: Determine if the current scenario is one with complete information. If yes, proceed to step 3.3; otherwise, proceed to step 3.

7. Step 3.3: Devices exchange their computing power consumption decisions with each other via the local network; Step 3.4: Sort all IoT devices in ascending order of data importance, and select the top two sorted IoT devices into the set of devices with non-zero computing power contribution; initialize the iteration parameters. ; Step 3.5: When Less than the total number of IoT devices and At that time, the first Add an IoT device to the set of devices with a non-zero computing power contribution, and make... ;in, The importance of the (m+1)th IoT device The importance of the i-th IoT device; Step 3.6: Calculate the optimal computing power consumption decision for each IoT device in the set of devices with a non-zero computing power contribution, based on the second preset formula. Then proceed to step 3.9; Step 3.7: Collect environmental information that the device can sense; Step 3.8: Construct a Markov process for solving the optimal computing power consumption decision. Use the Q-learning algorithm to learn the optimal computing power consumption strategy through interaction with the environment, and approximate the optimal computing power consumption strategy. Then proceed to step 3.9; Step 3.9: If the optimal computing power consumption decision for IoT devices calculated in this game iteration is consistent with that in the previous game iteration, then stop the iteration; Step 3.10: Calculate the first... The optimal computing power consumption decision for each IoT device is sent to the dynamic incentive coordination module.

2. The computing power adaptive adjustment system for trusted on-chain data of IoT devices adapted to multiple scenarios as described in claim 1, characterized in that, The security quantification module includes: The data acquisition unit is used to collect the total computing power contributed by all IoT devices through the edge server in real time. ; The monitoring unit is used to monitor and estimate the computing power of potential attackers. ; The first calculation unit is used to calculate based on the total computing power value. and the computing power of the potential attacker Calculate the probability that an IoT device will find the next block. And the probability of an attacker finding the next block ; The second calculation unit is used to determine the probability of finding the next block based on the IoT device. And the probability of an attacker finding the next block Calculate the attacker's The probability of successfully finding the next block after a certain number of blocks. ; Integration unit, used to integrate attackers'... The probability of successfully finding the next block after a certain number of blocks. Total computing power potential attacker computing power Combined with timestamps, a structured security posture report is generated.

3. The computing power adaptive adjustment system for trusted on-chain data of IoT devices adapted to multiple scenarios as described in claim 2, characterized in that, attackers The probability of successfully finding the next block after a certain number of blocks. for: ； ； in, Incomplete regularization function.

4. The computing power adaptive adjustment system for trusted on-chain data of IoT devices adapted to multiple scenarios as described in claim 1, characterized in that, The dynamic incentive coordination module is used to obtain the workshop server blockchain security reward budget based on the evaluation results of the security quantification module, including: Step 2.1: Based on the evaluation results of the security quantification module, extract the attacker's... The probability of successfully finding the next block after a certain number of blocks. ; Step 2.2: If this is the first game iteration, then initialize randomly. If the condition is met, proceed to step 2.6; otherwise, proceed to step 2.

3. Rewards for blockchain security; Step 2.3: Receive computing power consumption decisions sent by the device computing power adaptation module of each IoT device; Step 2.4: Count the number of IoT devices with a non-zero computing power contribution. The workshop server blockchain security reward budget is calculated using the first preset formula. Step 2.5: If the workshop server blockchain security reward budget calculated in this game iteration is consistent with that in the previous game iteration, then stop the iteration; Step 2.6: Broadcast the calculated workshop server blockchain security reward budget to all IoT devices.

5. The computing power adaptive adjustment system for trusted on-chain data of IoT devices adapted to multiple scenarios as described in claim 4, characterized in that, The first preset formula is: in, Budget for blockchain security rewards for workshop servers. To control the constant parameters of the total computing power, To control the constant parameter of the blockchain security evaluation function of the workshop server, To incentivize the budget, For the total One IoT device To assist in calculating variables, i For the i-th IoT device, The importance of the i-th IoT device.

6. The computing power adaptive adjustment system for trusted on-chain data of IoT devices adapted to multiple scenarios as described in claim 1, characterized in that, The second preset formula is: in, For optimal computing power consumption strategy, The server's current reward strategy, For the total One IoT device For the first One IoT device For the first Decisions regarding the computing power consumption of individual IoT devices.

7. The computing power adaptive adjustment system for trusted on-chain data of IoT devices adapted to multiple scenarios as described in claim 1, characterized in that, The environmental information that the device can sense includes: historical block generation success rate, macro-level security indicators broadcast by the blockchain platform, and edge server computing load.

Images