Privacy data protection method and system for industrial internet of things

Abstract

This invention discloses a method and system for protecting privacy data in the Industrial Internet of Things (IIoT). It addresses the technical problem of data leakage due to the susceptibility of single-node data storage to failure or attack. The invention uses a threshold secret sharing algorithm to split data into invalid fragments and employs a distributed storage system with single-device storage of individual fragments and targeted storage of critical fragments. This constructs an active protection barrier. Since a single fragment contains no valid information, attackers must simultaneously acquire multiple fragments to reconstruct the data, significantly increasing the difficulty of attacks. This fundamentally changes the centralized storage model and completely eliminates the security risks of single-point breaches and total data leakage.

Description

Technical Field

[0001] This invention relates to the field of Internet of Things (IoT) information technology, specifically to a method and system for protecting privacy data in the industrial IoT. Background Technology

[0002] In the Industrial Internet of Things (IIoT, which connects industrial equipment, sensors, gateways, etc., through a network to achieve data collection and interaction), edge devices (terminal devices deployed in industrial production sites that directly connect to equipment or sensors, such as temperature sensors, pressure sensors, edge gateways, etc.) generally suffer from limited storage capacity (mostly MB to GB level, unable to store large amounts of sensitive data) and low computing power (weak CPU performance, making it difficult to run complex encryption or storage algorithms). However, edge devices need to collect and process large amounts of sensitive data (industrial data that, if leaked, tampered with, or lost, could lead to economic losses for enterprises, production safety accidents, or violations of laws and regulations, such as process parameters, equipment safety thresholds, etc.). Traditional centralized storage (storing all data in a central server) is prone to data leakage due to central node failure or attacks. Summary of the Invention

[0003] To address the shortcomings of existing technologies, this invention provides a method and system for protecting privacy data in the Industrial Internet of Things (IIoT), which solves the problem of data leakage caused by single data storage due to central node failure or attack.

[0004] To achieve the above objectives, the present invention provides a privacy data protection method for the Industrial Internet of Things, comprising the following steps:

[0005] S1: Edge device trust assessment and high-trust cluster screening. The trust of edge devices in the industrial Internet of Things is assessed from the dimensions of solid-state version, historical behavior and physical location. Devices that meet the trust standards are selected to form a storage cluster.

[0006] S2: Sensitive data quantification and classification and threshold secret sharing fragment execution. The collected sensitive data is preprocessed to remove invalid data. The sensitive data is matched with the classification rule base type to determine the total number of fragments of the threshold secret sharing algorithm and the minimum number of fragments required for restoration. The data is split into multiple fragments containing key fragments and ordinary fragments. The fragments are distributed according to the rule that each device in the storage cluster stores one fragment and key fragments are distributed to the edge gateway.

[0007] S3: Fragment access collaborative verification. After the authorized node initiates an access request, the cluster management node reviews and generates a temporary access token. The authorized node carries the token to send a request to the device in the storage cluster that has at least the minimum number of fragments required to restore the data. After the devices collaboratively verify the token, they return the fragments. The authorized node then collects the fragments to restore the data.

[0008] As a further aspect of the present invention: the edge devices in S1 include sensors and edge gateways. When assessing trust level, the solid-state version dimension is scored based on the matching of the device's operating system or core program version with the security version; the historical behavior dimension is scored based on the number of abnormal connections and data transmission errors in the device's operation logs over the past 3 months; and the physical location dimension is scored based on the protection status of the area where the device is located. Devices with a trust level of not less than 80 points are selected to form a high-trust storage cluster, and the ID, type, and physical location of each device in the cluster are recorded to ensure that the number of devices in the cluster is not less than the total number of fragments for subsequent data splitting.

[0009] As a further aspect of the present invention: In step S2, the classification rule base includes at least the categories of personal privacy data, trade secret data, and government sensitive data. The collected sensitive data is matched with the three categories in the preset classification rule base using a similarity matching algorithm to obtain the highest matching degree and automatically classified into the corresponding category.

[0010] As a further aspect of the present invention: Based on the aforementioned category, corresponding parameters are set for the threshold secret sharing algorithm, wherein the total number of fragments N and the minimum number of fragments M required for restoration satisfy M... <N;

[0011] The parameters include at least the following: N=4 and M=2 for medium-risk data, N=5 and M=3 for high-risk data, and N=6 and M=4 for extremely high-risk data.

[0012] As a further aspect of the present invention: In S2, the threshold secret sharing algorithm is invoked to split the preprocessed sensitive data into N data fragments, and a unique identifier containing data ID, fragment number and generation time is generated for each fragment. At the same time, 1 to 2 fragments are marked as key fragments from the N fragments, and the rest are ordinary fragments.

[0013] The key fragments are used for subsequent fragment splicing and verification. If the key fragments are missing, the data cannot be restored even if M fragments are collected.

[0014] As a further aspect of the present invention: In step S2, the distribution of fragments through the cluster management node specifically includes: counting the number of edge gateways and ordinary sensors within the storage cluster; sending critical fragments to the edge gateways and ordinary fragments to the ordinary sensors through a secure communication channel; each device sending a storage success signal to the cluster management node after receiving the fragment; the cluster management node verifying the number of fragments and their storage locations; and triggering an alarm in case of an anomaly.

[0015] As a further aspect of the present invention: in S3, the access request initiated by the authorized node must include the identity of the accessing node, the target data identifier, the purpose of access, and the required number of fragments, and the required number of fragments shall not be less than the minimum number of fragments M required for restoration determined in S2.

[0016] As a further aspect of the present invention: the temporary access token in S3 uses the AES-128 encryption algorithm and includes the access node ID, data ID, validity period and encryption verification code, with a maximum validity period of 2 hours;

[0017] The inter-device collaborative verification includes: receiving devices verifying the consistency and validity of tokens with each other, reaching a consensus, and synchronously returning fragments to the authorized node; if there is a disagreement, refusing to return and issuing an alarm.

[0018] As a further aspect of the present invention: In step S2, if a single storage node issues an attack alert or its stored fragments are lost, after authorization verification is passed, complete data is restored by collecting no less than M fragments from the remaining healthy devices, and the threshold secret sharing fragmentation and distribution steps are re-executed to replace the lost fragments.

[0019] Privacy data protection systems for the Industrial Internet of Things include:

[0020] The device evaluation cluster module is used to evaluate the trustworthiness of edge devices from the dimensions of firmware version, historical behavior, and physical location, and to select and build high-trust storage clusters.

[0021] The data fragmentation processing module is used to clean and classify sensitive data, and based on the classification results, call the threshold secret sharing algorithm to split the data into multiple fragments and mark the key fragments;

[0022] The fragment distribution management module is used to distribute data fragments to each edge device in the storage cluster according to rules, and manage the storage status of fragments;

[0023] The access collaboration verification module is used to receive access requests from authorized nodes, review and generate temporary access tokens, coordinate devices in the storage cluster to perform token collaboration verification, and control the return of fragments to complete data restoration after successful verification.

[0024] This invention provides a method and system for protecting privacy data in the Industrial Internet of Things (IIoT). Compared with existing technologies, it has the following advantages:

[0025] (1) This invention splits data into invalid fragments through a threshold secret sharing algorithm and uses a distributed storage rule of storing single fragments on a single device and storing key fragments in a targeted manner to build an active protection barrier. A single fragment does not contain valid information, and attackers must obtain multiple fragments at the same time to restore the data, which greatly increases the difficulty of attack, fundamentally changes the centralized storage mode, and completely solves the security risks of single point of attack and total leakage.

[0026] (2) This invention ensures that only legally authorized nodes can collect enough fragments to restore data through a dual access mechanism of management node audit and device collaborative verification, thus eliminating the possibility of attackers illegally obtaining distributed fragment splicing data and further protecting data security under non-single storage architecture, which is in line with the core goal of cracking the leakage risk of single storage.

[0027] (3) By splitting data into small fragments for storage, the present invention significantly reduces the storage capacity requirement of a single edge device. Each device only needs to store one fragment instead of a complete data packet, making it possible to securely store sensitive data on devices such as sensors and edge gateways with limited storage resources. This solves the problem of limited data storage caused by insufficient storage capacity on the edge side and improves the practicality and deployment flexibility of the solution. Attached Figure Description

[0028] Figure 1 This is a flowchart of the method of the present invention;

[0029] Figure 2 This is a system flowchart of the present invention. Detailed Implementation

[0030] 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.

[0031] Example 1

[0032] Please see Figure 1 This application provides a privacy data protection method for industrial IoT, including the following steps: S1: Evaluate the trust level of edge devices in industrial IoT from the dimensions of solid-state version, historical behavior and physical location, and select devices with qualified trust levels to form a storage cluster;

[0033] S2: Preprocess the collected sensitive data to remove invalid data, match the sensitive data with the classification rule base type, determine the total number of fragments of the threshold secret sharing algorithm and the minimum number of fragments required for restoration, split the data into multiple fragments containing key fragments and ordinary fragments, and distribute the fragments according to the rule that each device in the storage cluster stores one fragment and key fragments are distributed to the edge gateway.

[0034] S3: After the authorized node initiates an access request, the cluster management node reviews and generates a temporary access token. The authorized node carries the token and sends a request to the device in the storage cluster that has at least the minimum number of fragments required for restoration. After the devices jointly verify the token, they return the fragments. The authorized node then collects the fragments to restore the data.

[0035] Example 2

[0036] This illustrates another embodiment of the present invention, further detailing the content of Embodiment 1 based on Embodiment 1. This application provides a privacy data protection method for industrial Internet of Things, specifically including the following steps:

[0037] S1: For all edge devices in the Industrial Internet of Things (IIoT), including sensors and edge gateways, trust levels are assessed through data extraction and automatic calculation across three dimensions. Devices scoring 80 or above out of 100 are considered high-trust devices.

[0038] The total device trust score is T, the solid-state version score is T1, the historical behavior score is T2, and the physical score is T3. The calculation formula is T=T1+T2+T3.

[0039] The system reads the version number of the device's operating system or core program, and the security version: if it is the latest security version and there are no known vulnerabilities, then T1=30; if the version is one cycle behind and there are low-risk vulnerabilities, then T1=20; if the version is two or more cycles behind and there are high-risk vulnerabilities, then T1=0.

[0040] The system automatically extracts the device's operation logs from the past 3 months and counts the number of abnormal connection and data transmission errors P: if P=0, then T2=40; if P=1, then T2=20; if P≥2, then T2=0.

[0041] The system reads the device location information and area protection status: If the device is in a closed workshop, such as with access control and monitoring, and only authorized personnel can access it, then T3=30; if the device is in a semi-open area, such as without access control but with monitoring, then T3=15; if the device is in an open area without protection, and anyone can access it, then T3=0.

[0042] Based on the above scoring criteria, the firmware version, historical behavior logs, and physical location information of all edge devices are extracted from the device management database. The trust score T of each device is calculated. Then, all devices with T≥80 are selected to form a high-trust device cluster, referred to as a storage cluster. This refers to a set of edge devices specifically used to store data fragments. The ID, type (including sensor / edge gateway), and physical location of each device in the storage cluster are recorded to ensure that the number of devices in the cluster is ≥ the total number of fragments N of the subsequent data splitting.

[0043] Specifically, based on three quantitative dimensions—solid-state version security, historical behavior stability, and physical location protection level—all edge devices are automatically trusted. High-trust devices are selected according to preset standards, their detailed information is recorded, and they are assembled into a storage cluster dedicated to storing data fragments. At the same time, it is ensured that the cluster size can meet the total number of data shards required in the future.

[0044] S2: Collect information in real time through sensors, device interfaces and other acquisition nodes to obtain sensitive data. The collected information includes various sensitive data such as energy consumption and user privacy information. Scan the collected sensitive data to remove duplicate invalid parameters.

[0045] Using a pre-defined classification rule base, sensitive data is categorized into personal privacy data such as ID card numbers and biometrics, commercial secret data such as financial statements and core technology solutions, and government sensitive data such as policy documents and military deployment information. The collected sensitive data is then matched with the three categories in the pre-defined classification rule base using a similarity matching algorithm. The highest matching degree among the three categories is obtained and the sensitive data is automatically assigned to that category.

[0046] The categorized sensitive data is input into the threshold secret sharing algorithm. The data is then divided into fragments and the number of fragments required to restore the complete data, according to their respective types. Specifically, medium-risk data corresponds to N=4 fragments and M=2 fragments, high-risk data corresponds to N=5 fragments and M=3 fragments, and extremely high-risk data corresponds to N=6 fragments and M=4 fragments. Here, N is the total number of fragments, and M is the minimum number of fragments required to restore the complete data. <N;

[0047] The algorithm automatically splits the data into N data fragments and generates a unique identifier for each fragment, including data ID, fragment number, and generation time. At the same time, it marks 1 to 2 "key fragments" from the N fragments for verification during subsequent fragment splicing. If the key fragments are missing, the data cannot be restored even if M fragments are collected. The rest are "ordinary fragments".

[0048] For example: The energy consumption core data of a certain automotive parts production line is commercially sensitive and high-risk. It is set to N=5 and M=3. It is split into 5 fragments through an algorithm, such as numbered P1-P5. P1 and P2 are marked as critical fragments, and P3-P5 are ordinary fragments. Only by obtaining 3 fragments and substituting them into the threshold secret sharing algorithm for combination calculation can the original data be combined. Among the 3 fragments, there are 1 to 2 critical fragments.

[0049] It should be noted that Threshold Secret Sharing is a cryptographically secure data segmentation and recovery technique. Its core idea is to divide a complete piece of sensitive data, called a "secret," into multiple data fragments using mathematical algorithms. These fragments themselves do not contain any intelligible original information. Only when a sufficient number of fragments are collected to reach a "threshold value" can the original data be restored through mathematical calculations. If the number of fragments collected is insufficient, no information about the original data can be obtained.

[0050] The fragmented data is distributed through the cluster management node: a core node deployed on an industrial gateway or local server, responsible for fragment distribution and device management, following these steps:

[0051] Step 1: Count the number of edge gateways and ordinary sensors within the storage cluster to identify critical fragments and ordinary fragments;

[0052] Step 2: Send critical fragments to edge gateways and ordinary fragments to ordinary sensors through secure communication channels such as encrypted Ethernet. The edge gateways and ordinary sensors are devices in a storage cluster with T≥80 points, and each edge gateway and ordinary sensor stores only one critical fragment and one ordinary fragment.

[0053] Step 3: After each device receives the fragment, it sends a storage success signal to the cluster management node through the system's built-in message feedback mechanism. The cluster management node calls the system database to verify the number of fragments and their storage locations. For example, if 5 fragments need to be successfully distributed to 5 devices, an alarm will be triggered if an abnormality occurs, and error details will be displayed on the system operation interface.

[0054] If a fragment of a single node is maliciously attacked, an alarm is issued and all fragment information is blocked, waiting for an authorized engineer to verify and unlock it. If the fragment of this node is lost, after the authorized engineer verifies and unlocks it, the fragments required for restoration are input into the threshold secret sharing algorithm. After obtaining complete data, the threshold secret sharing algorithm is used to split the new N fragments, including one new fragment used to replace the lost fragment Px.

[0055] It should be noted that since M < N, as long as the number of lost fragments does not exceed N - M and there is at least one critical fragment, it is still possible to collect at least M fragments from the remaining healthy devices to restore the original data;

[0056] Specifically, after cleaning the collected original sensitive data, it is automatically identified and classified using a rule base and a similarity matching algorithm. According to the risk level corresponding to the classification result, the parameters of the threshold secret sharing algorithm are automatically determined. The complete data is encrypted and split into N invalidated fragments, and a small number of critical fragments are marked from them. Subsequently, strictly in accordance with the rule that each device stores a single fragment and the critical fragments are stored directionally in a high-security edge gateway, the fragments are distributed to the storage cluster established in S1 through a secure channel.

[0057] S3: When an authorized node, which refers to a device or terminal that has been enterprise-certified and has the right to access sensitive data, such as the operation computer of an operation and maintenance engineer or a production management platform, needs to view the complete sensitive data, it submits an access request to the cluster management node. The content of the request should include: the identity of the access node, such as the operation and maintenance terminal ID: TERM001, the target data identifier, such as the production line energy consumption data ID: DATA005, the purpose of the access, such as production parameter optimization, and the number of required fragments ≥ M. For example, if M = 3, then 3 or more fragments are requested;

[0058] The management node reviews the request: confirms whether the identity of the access node is an authorized object and whether the purpose of the access is compliant. After passing the review, it generates a temporary access token using the AES-128 encryption algorithm, which includes the access node ID, data ID, valid time, up to 2 hours at most, an encrypted verification code, and sends the token to the authorized node;

[0059] It should be noted that the core feature of the temporary access token is that it has a short validity period, up to 2 hours at most, and needs to be quickly generated, transmitted, and verified between the management node and the authorized node. As a symmetric encryption algorithm, AES-128 uses the same key for encryption and decryption, and its operation logic is simple. Based on lightweight operations such as byte substitution and shifting, its computational overhead is extremely low compared to asymmetric encryption;

[0060] Authorized nodes, carrying temporary access tokens, initiate fragment acquisition requests to at least M devices within the storage cluster via the industrial intranet. For example, if M=3, requests are sent to 3 devices, including critical fragment storage devices. Each device receiving a request does not directly return fragments but first performs token collaborative verification with other devices that received requests: they send their respective received token information to each other, verifying the consistency of the tokens, whether they are the same encrypted token, and their validity; they check whether the accessing node is legitimate and whether the token is valid. If all participating devices reach a consensus on the validity of the token, such as all 3 devices confirming the token is valid, they synchronously return their respective stored fragments to the authorized node. If there is a disagreement, such as one device receiving a different token than the others, the fragments are refused to be returned, and an abnormal request alarm is sent to the management node.

[0061] Once an authorized node has collected ≥M fragments, it invokes the restoration function of the threshold secret sharing algorithm, inputs all fragments, and the algorithm automatically splices the fragments to generate complete sensitive data for use by the authorized node, such as for operations engineers to view energy consumption data and optimize production parameters.

[0062] When a legitimate authorized node needs to access data, it must submit a detailed application to the cluster management node and obtain a short-term encrypted token after approval. The authorized node must use this token to initiate fragment requests to multiple devices in the storage cluster, no less than M devices. The devices receiving the requests do not respond individually, but first communicate with each other and collaboratively verify the consistency and validity of the tokens. Only after reaching a consensus will they synchronously return the fragments they hold. Only after the authorized node has finally collected all the fragments can the complete data be restored.

[0063] Example 3

[0064] Furthermore, refer to Figure 2 As shown, a privacy data protection system for the Industrial Internet of Things (IIoT) is proposed to implement the privacy data protection method for the IIoT as described above, including:

[0065] The device evaluation cluster module is used to evaluate the trustworthiness of edge devices from the dimensions of firmware version, historical behavior, and physical location, and to select and build high-trust storage clusters.

[0066] The data fragmentation processing module is used to clean and classify sensitive data, and based on the classification results, call the threshold secret sharing algorithm to split the data into multiple fragments and mark the key fragments;

[0067] The fragment distribution management module is used to distribute data fragments to each edge device in the storage cluster according to rules, and manage the storage status of fragments;

[0068] The access collaboration verification module is used to receive access requests from authorized nodes, review and generate temporary access tokens, coordinate devices in the storage cluster to perform token collaboration verification, and control the return of fragments to complete data restoration after successful verification.

[0069] Some of the data in the above formulas are numerical calculations with dimensions removed, and the contents not described in detail in this specification are all prior art known to those skilled in the art.

[0070] The above embodiments are only used to illustrate the technical methods of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical methods of the present invention without departing from the spirit and scope of the technical methods of the present invention.

Claims

1. A privacy data protection method for industrial IoT, characterized in that, Includes the following steps: S1: Evaluate the trustworthiness of edge devices in the Industrial Internet of Things from the dimensions of solid-state version, historical behavior, and physical location, and select devices that meet the trustworthiness standards to form a storage cluster; S2: Preprocess the collected sensitive data to remove invalid data, match the sensitive data with the classification rule base type, determine the total number of fragments of the threshold secret sharing algorithm and the minimum number of fragments required for restoration, split the data into multiple fragments containing key fragments and ordinary fragments, and distribute the fragments according to the rule that each device in the storage cluster stores one fragment and key fragments are distributed to the edge gateway. The total number of fragments N and the minimum number of fragments M required for restoration satisfy M. <N; In S2, the threshold secret sharing algorithm is invoked to split the preprocessed sensitive data into N data fragments, and a unique identifier containing data ID, fragment number and generation time is generated for each fragment. At the same time, 1 to 2 fragments are marked as key fragments from the N fragments, and the rest are ordinary fragments. The key fragments are used for subsequent fragment splicing and verification. If the key fragments are missing, the data cannot be restored even if M fragments are collected. S3: After the authorized node initiates an access request, the cluster management node reviews and generates a temporary access token. The authorized node carries the token and sends a request to the device in the storage cluster that has at least the minimum number of fragments required to restore the data. After the devices jointly verify the token, they return the fragments. The authorized node then collects the fragments to restore the data. The inter-device collaborative verification includes: receiving devices verifying the consistency and validity of tokens with each other, reaching a consensus, and synchronously returning fragments to the authorized node; if there is a disagreement, refusing to return and issuing an alarm.

2. The privacy data protection method for industrial IoT according to claim 1, characterized in that, The edge devices in S1 include sensors and edge gateways. When assessing trust level, the solid-state version dimension is scored based on the matching of the device's operating system or core program version with the security version. The historical behavior dimension is scored based on the number of abnormal connections and data transmission errors in the device's operation logs over the past 3 months. The physical location dimension is scored based on the protection status of the area where the device is located. Devices that meet the trust level are selected to form a high-trust storage cluster, and the ID, type, and physical location of each device in the cluster are recorded to ensure that the number of devices in the cluster is not less than the total number of fragments for subsequent data splitting.

3. The privacy data protection method for industrial IoT according to claim 1, characterized in that, In step S2, the classification rule base includes at least three categories: personal privacy data, trade secret data, and government sensitive data. The collected sensitive data is matched with the three categories in the preset classification rule base using a similarity matching algorithm to obtain the highest matching degree and automatically classify it into the corresponding category.

4. The privacy data protection method for industrial IoT according to claim 3, characterized in that, Based on the categories described, set the corresponding parameters for the threshold secret sharing algorithm; The parameters include at least the following: N=4 and M=2 for medium-risk data, N=5 and M=3 for high-risk data, and N=6 and M=4 for extremely high-risk data.

5. The privacy data protection method for industrial IoT according to claim 1, characterized in that, In step S2, fragments are distributed through the cluster management node. Specifically, this includes: counting the number of edge gateways and ordinary sensors within the storage cluster; sending critical fragments to the edge gateways and ordinary fragments to the ordinary sensors through a secure communication channel; each device sending a storage success signal to the cluster management node after receiving the fragment; the cluster management node verifying the number and storage location of fragments; and triggering an alarm when an anomaly occurs.

6. The privacy data protection method for industrial IoT according to claim 1, characterized in that, In S3, the access request initiated by the authorized node must include the identity of the accessing node, the target data identifier, the purpose of access, and the required number of fragments. The required number of fragments shall not be less than the minimum number of fragments M required for restoration determined in S2.

7. The privacy data protection method for industrial IoT according to claim 1, characterized in that, The temporary access token in S3 uses the AES-128 encryption algorithm and includes the access node ID, data ID, validity period and encryption verification code. The validity period is up to 2 hours.

8. The privacy data protection method for industrial IoT according to claim 1, characterized in that, In step S2, if a single storage node issues an attack alert or its stored fragments are lost, after authorization verification is passed, the complete data is restored by collecting no less than M fragments from the remaining healthy devices, and the threshold secret sharing fragmentation and distribution steps are re-executed to replace the lost fragments.

9. A privacy data protection system for industrial IoT, used to execute the privacy data protection method for industrial IoT as described in any one of claims 1-8, characterized in that, include: The device evaluation cluster module is used to evaluate the trustworthiness of edge devices from the dimensions of firmware version, historical behavior, and physical location, and to select and build high-trust storage clusters. The data fragmentation processing module is used to clean and classify sensitive data, and based on the classification results, call the threshold secret sharing algorithm to split the data into multiple fragments and mark the key fragments; The fragment distribution management module is used to distribute data fragments to each edge device in the storage cluster according to rules, and manage the storage status of fragments; The access collaboration verification module is used to receive access requests from authorized nodes, review and generate temporary access tokens, coordinate devices in the storage cluster to perform token collaboration verification, and control the return of fragments after successful verification to complete data restoration.

Images