A data integrity verification method and a data integrity batch verification method in an edge computing environment

Abstract

The application relates to a data integrity verification method and a data integrity batch verification method in an edge computing environment, and belongs to the technical field of data security.The integrity verification method comprises an initialization process, a label generation process, an uploading process, a request process and a response process.The batch verification method comprises an initialization process, a label generation process, an uploading process, a batch request process, a batch response process and a batch verification process.The application realizes batch auditing of single-edge nodes and multiple-edge nodes.The application has lower calculation overhead, communication overhead and time consumption than existing schemes.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data security such as data outsourcing storage and edge data storage, and particularly relates to a data integrity verification method in an edge computing environment and a batch data integrity verification method. BACKGROUND

[0002] With the rapid development of Internet of Things technology and 5G network architecture, new service models and businesses such as intelligent transportation, smart city, location service, mobile payment, etc. are emerging. The number of smart phones, wearable devices, networked televisions and other sensing devices will show an explosive growth trend, and the "massive" data generated by Internet of Things terminals will follow. A joint study by Seagate and IDC shows that the global data volume will reach 163 ZB by 2025, and the era of edge big data processing of Internet of Things has arrived. Among them, the device nodes at the network edge are no longer just data users, but are transforming into computing nodes that take into account data collection, pattern recognition, data mining and other big data processing capabilities. At the same time, these edge device nodes provide rich service interfaces and work together with cloud computing centers to provide collaborative computing services for users.

[0003] The traditional cloud computing model cannot meet the application requirements of Internet of Things, of which the most notable is the security problem of data. Network edge data involves personal privacy, and the traditional cloud computing mode needs to upload these private data to the cloud computing center, which will increase the risk of leaking user privacy data, so edge computing emerges as the times require. Edge computing refers to a new service model in which data or tasks can be computed and executed at the network edge close to the data source. Here, the network edge side can be any functional entity between the data source and the cloud computing center, and these entities carry edge computing platforms that integrate network, computing, storage and application core capabilities to provide real-time, dynamic and intelligent service computing for end users. At the same time, the concept of data processing near the source also provides better structured support for data security.

[0004] The most important part of the edge computing architecture is the edge data center. Compared with the cloud environment, the edge computing environment is more vulnerable to external attacks, and the cached data is more vulnerable to damage. The data security of the edge data center has always been the concern of the end user. The distributed parallel data processing mode under the edge computing mode makes the edge computing platform face security challenges such as malicious tampering and data damage. Data security is the basis for creating a secure edge computing environment, and its fundamental purpose is to protect data integrity. Due to the dynamic and variable nature of the edge environment, the cached copies on the edge server are vulnerable to various intentional and accidental damage, and sudden hardware or software abnormalities can also cause data damage on the edge server. Data damage affects the quality of services provided by the edge server and is a major network security threat in the distributed edge computing environment. Therefore, the integrity of the edge data copies cached on the edge server must be ensured, and data modification or data damage must be detected in a timely manner. SUMMARY

[0005] To achieve the above purpose, the application provides a data integrity verification method and a data integrity batch verification method in an edge computing environment.

[0006] A fast data integrity verification method, the system applied by the method includes a plurality of user terminals, a third-party auditor, and an edge node, which are communicatively connected, and each of the three has its own public key and private key.

[0007] The method includes an initialization process, a label generation process, an upload process, a request process, and a response process.

[0008] In the initialization process, the initial parameters ec, G, M, H, Φ, f are given; ec is an elliptic curve, G is a generator on the elliptic curve, M is another point on the elliptic curve; H is a collision-resistant hash, H: (0, 1) * ∈ Zq, is a pseudo-random sequence, Φ is a pseudo-random function, and f is a pseudo-random function used by the user terminal to blind the data; in the label generation process, the user terminal divides the data file to be verified into n blocks, each data block corresponds to an index position, the user terminal blinds the data blocks, and generates corresponding labels for each blinded data block.

[0009] In the upload process, the user terminal uploads the data file to the edge node, and the label is uploaded to the third-party auditor.

[0010] In the request process, the third-party auditor generates a challenge index I, a random number corresponding to each index, a random number r1, r2, and encrypts Req with the edge node public key, and sends a request to the edge node, wherein the Req = {I, V, r1};

[0011] In the response process, after receiving the request, the edge node first decrypts it with the private key, generates signature R, P according to the challenge index and the random number, and calculates RES, and then uses the public key pk of the third-party auditor TPA The encrypted RES is sent to the third-party auditor;

[0012] In the verification process, the third-party auditor calculates the following formula:

[0013]

[0014] The third-party auditor compares the result returned by the edge node with T, and returns the comparison result to the user end.

[0015] The label generation process is as follows:

[0016] Suppose the file uploaded by the user end is F, and the user end divides it into n blocks F=(m1, m2,...m n ), where i is the index position corresponding to the data block, and i takes values from 1 to n; the user first selects a random number k∈Z q as the key of the pseudo-random function f, and generates a corresponding blinding factor b i for each block, where b k (j, name) is the blinding factor of the jth block, and the blinded data block is F i = m i +b i , where F i represents the ith block of the blinded file;

[0017] For the blinded data, generate a label T=(T1, T2,...T n );

[0018] T i = H(F id ||F i )+sk user

[0019] where F id is the unique identity of the file F, and T i represents the label of the ith data block.

[0020] The request process is as follows: the third-party auditor first randomly generates a data block number c that needs to be verified for the data file that needs to be verified, selects the index sequence block that meets the requirements according to the stored time and the time verified in the last round, generates challenge index I={s1, s2,...s c}, c∈{1,2,...,n}; generates a random number according to the calculation formula V i = φ(k, i), where one index corresponds to one random number Zq* is a positive integer less than q, each data block corresponds to a set of random numbers, and a number r1 is randomly generated for the edge node, and a random number r2 is saved by itself, and the public key pk of the edge node is used edge Encrypt Req = {I, V, r1}, send the request to the corresponding edge node, where V is the set of V i .

[0021] The response process is as follows: after the edge node receives the request, the private key is used for decryption, the signature R, P is generated according to the challenge index and the random number, and then the calculation result RES is encrypted and sent by using the public key pk of the third-party auditor TPA The calculation formula is as follows:

[0022]

[0023] In the verification process, the third-party auditor verifies whether the following equation is equal:

[0024] T-RES≡(r2-r1)M

[0025] The verification result is returned to the user end, and if the equation is established, True is returned, and if the equation is not established, False is returned.

[0026] A fast data integrity batch verification method, the system applied by the method includes a plurality of user ends, a third-party auditor, a plurality of edge nodes, and communication connections among the three parties, each device has its own public key and private key; the method includes initialization process, label generation process, uploading process, batch request process, batch response process, batch verification process;

[0027] In the initialization process, the initial parameters ec, G, M, H, Φ, f are given; ec is an elliptic curve, G is a generator on the elliptic curve, M is another point on the elliptic curve; H is a collision-resistant hash, H: (0, 1) * ∈Zq, is a pseudo-random sequence, Φ is a pseudo-random function, and f is a pseudo-random function for user end to blind the data;

[0028] In the label generation process, the user end divides the data file to be verified into n blocks, and performs blind processing, and generates a label for the blind data block. The label can be a text label or a data label;

[0029] In the uploading process, the user end uploads the data file to the edge node, and the label is uploaded to the third-party auditor;

[0030] In the batch request procedure, the third-party auditor generates request information for each edge node, and randomly generates a number of data blocks to be verified for a data file to be verified, generates a challenge index I, a random number corresponding to each index, random numbers r1 and r2, encrypts the request Req = {I, V, r1} with the public key, and sends the request to the corresponding edge node;

[0031] In the batch response procedure, after receiving the request, the edge node first decrypts with the private key, generates a signature R, P according to the challenge index and the random number, calculates RES, and then encrypts and sends to the third-party auditor with the public key of the third-party auditor;

[0032] In the batch verification procedure, the third-party auditor calculates the following formula:

[0033]

[0034] The third-party auditor compares the sum of the results returned by each edge node with T, and returns the comparison result to the user end.

[0035] In the batch response procedure, the edge node receiving the request needs to first decrypt with the private key, each edge node generates a signature R, P according to the challenge index and the random number, then hides the generated information, calculates RES, and encrypts and sends to the third-party auditor with the public key pk TPA The calculation formula is as follows:

[0036]

[0037] RES = r1 * M + |R + P| * G.

[0038] In the batch response procedure, the third-party auditor compares the sum of the results returned by each edge node with T, and the specific method is to verify whether the following equation is equal:

[0039] T-res ≡ (r2-j*r1)M

[0040] res is the sum of RES generated by each edge node;

[0041] If the equation is established, True is returned to the user end, and False is returned if the equation is not established.

[0042] Both the above two methods can use index-chain table storage form, each index corresponds to each part of the chain table, and each part contains a sequence number, an index number of a data block, a last modification time, a last verification time, and a pointer to a physical storage location of the data block in the single chain table.

[0043] The beneficial effects of the present application are:

[0044] 1. In the process of computing the verification request received by the edge server in the scheme, compared with other schemes, the calculation only involves simple addition, subtraction, multiplication and hash, and does not involve complex operation, which significantly reduces the calculation overhead.

[0045] 2. The elliptic curve and the Pedersen commitment based on the elliptic curve are used to map the data to the elliptic curve, and the data hiding in the verification process is realized. Even if the enemy obtains the related data, the useful data cannot be recovered.

[0046] 3. Before the user uploads the file, the data is blinded, which makes the edge server unable to obtain the real data and causes data leakage.

[0047] 4. In the process of modifying the data, block-level modification is realized, and the influence on other data blocks is significantly reduced. BRIEF DESCRIPTION OF DRAWINGS

[0048] Figure 1 The system model diagram is shown.

[0049] Figure 2 The invention index-single chain represents the intention. DETAILED DESCRIPTION

[0050] The present application proposes a lightweight data integrity batch verification method (MEC-P) for edge computing environment, which adopts Pedersen commitment based on elliptic curve and improved I-LST, realizes the batch audit of single edge node and multiple edge nodes. The present application has lower calculation overhead, communication overhead and time consumption than the existing scheme.

[0051] The present application also proposes a system model for data integrity verification in mobile edge computing environment. In the model, the user needs a third-party verifier to verify the integrity of the required data block, and the data block will be protected, and the privacy of the user will not be leaked; the use of bilinear pair is avoided, thereby significantly reducing the calculation cost.

[0052] The present application adopts the blinding technology, avoids the privacy leakage of outsourcing Internet of Things data, and further protects the privacy of Internet of Things users. At the same time, the Pedersen commitment based on the elliptic curve is adopted to realize the data hiding in the verification process.

[0053] A fast data integrity verification method is described below.

[0054] The application scenario of the method includes several user terminals, a third party auditor (TPA), and an edge node. The user terminals, the third party auditor, and the edge node are communicatively connected. The framework of the system model and the relationship between the three entities are as shown in Figure 1 .

[0055] Edge node: The edge node generally refers to a computing device close to a data source or a data user (a user equipment, etc.) in an edge computing architecture. It is assumed in the embodiment that the edge node is untrusted. One of the benefits of edge computing is that the edge node is often deployed close to the user and even owned by the user (for example, a PC or a router). However, the ordinary user and the small organization can only have limited IT operation technology to maintain the machine. Therefore, the edge node can suffer from internal failures and is vulnerable to attacks by external enemies, and data damage often occurs.

[0056] User terminal: The initiator of data integrity verification. Different user terminals can be located in different geographical locations. They communicate with surrounding edge nodes through wireless communication technology. To ensure that the data is not lost before being downloaded, the user usually sends a data integrity verification request to the edge node and finally receives the verification result returned by the TPA.

[0057] Third party auditor: The executor of data integrity verification. In order to reduce the computing overhead of the user and the communication overhead between the user and the edge node, the TPA is introduced to replace the user to perform verification. It is assumed in the embodiment that the TPA is semi-honest and interested in obtaining the user's data, and can spy on the user's sensitive information by analyzing the collected information. The TPA is responsible for verifying the proof sent by the edge node and returning the verification result to the user. It has strong computing and storage capabilities and can efficiently perform verification operations. The third party auditor can be a server or a dedicated auditing device.

[0058] The above method includes the following flow:

[0059] I. Initialization flow

[0060] In the initial flow, the initial parameters are given

[0061] where ec is an elliptic curve, G is a generator on the elliptic curve, and M is another point on the elliptic curve. H is a collision-resistant hash, H: (0, 1) * ∈ Zq (Zq refers to an integer less than q), is a pseudo-random sequence, Φ is a pseudo-random function, and f is a pseudo-random function for the user terminal to blind the data. The user terminal has its own public key pk user and private key skuser , wherein where pk user = sk user * G.

[0062] II. Tag generation process

[0063] Suppose the file uploaded by the user is F, the user divides it into n blocks F = (m1, m2,... m n n), where i is the index of the data block, i is in the range of 1 to n. The user first selects a random number k e Z q as the key of the pseudo-random function f, and generates the corresponding blind factor b i = f k (i, name) for each block. name is the file name, and for the blinded data block, F i = m i + b i . F i i represents the i-th block of the blinded file. It is very difficult to recover the original data without the blind factor. Blindness also ensures that no entity other than the user can obtain the real data.

[0064] For the blinded data, generate the tag T = (T1, T2,... T n n).

[0065] T i i = H (F id || F i ) + sk user (1)

[0066] where F id is the unique identity of the file F, and T i represents the tag for the i-th data block. f k is a pseudo-random function with a random number k as the key.

[0067] III. Upload process

[0068] The upload process is that the user uploads the file to the edge node, and the tag is uploaded to the third-party auditor. Preferably, the user deletes the local data after uploading, the purpose is to reduce the storage consumption of the local.

[0069] IV. Request process

[0070] The third-party auditor first randomly generates a data block number c for the data file to be verified, according to the size of c, the related index I is generated, according to the storage time and the time verified in the last round, the index sequence block meeting the requirements is selected to generate the challenge index I = {s1, s2,... sc}, c e {1, 2,..., n}. According to the calculation formula V i = φ(k, i) generates a random number, one index corresponds to one random number, wherein Zq * is a positive integer less than q, each data block corresponds to a set of random numbers. At the same time, a number r1 is randomly generated for the edge node, and a random number r2 is saved by itself. The public key pk edge of the edge node is used to encrypt Req = {I, V, r1}, and the request is sent to the corresponding edge node. V is the set of V i . There are c elements in the V set.

[0071] Five, response process

[0072] In the response process, after the edge node receives the request, it needs to be decrypted with the private key first. According to the challenge index and the random number, a signature R, P is generated, and then the calculation result RES is encrypted and sent with the public key pk TPA of the third-party auditor. RES is the result returned by the edge node.

[0073]

[0074] Six, verification process

[0075] The third-party auditor calculates T according to the label:

[0076]

[0077] The TPA compares the result returned by the edge node with T, and verifies whether the following equation is equal.

[0078] T-RES≡(r2-r1)M (4)

[0079] The result is returned to the user end. If the equation is true, True is returned, and if the equation is not true, False is returned. True proves that the verification result is correct and the data is complete; False proves that the data is inconsistent and is damaged.

[0080] The application also provides a fast batch data integrity verification method. The application scenarios of the method include a plurality of user ends, a third-party auditor, and a plurality of edge nodes, which are communicatively connected. One third-party auditor can correspond to multiple users and multiple edge nodes, and the edge nodes are not connected, and the user end is directly connected.

[0081] This method extends the aforementioned basic data integrity verification method by proposing an auditing scheme for multiple edge nodes. The initialization, tag generation, and download processes are identical to the batch verification method described above. The difference lies in that this method also supports batch auditing, improving auditing efficiency and meeting the real-world needs of massive IoT data. Furthermore, an improved index-linked list (I-SLT) data structure is used to support fully dynamic updates of IoT data.

[0082] This section covers the initialization process, tag generation process, upload process, batch request process, batch response process, and batch verification process. The batch request process, batch response process, and batch verification process are described below.

[0083] Batch Request Process: The third-party auditor generates request information for each edge node. Assume there are j edge nodes. First, a random number of data blocks C = {c1, c2, ..., c...} is generated for the data file to be verified. j}, any element c in this set j For each element ∈ {1, 2, ..., n}, select a suitable index sequence block based on its storage time and the verification time from the previous round to generate a challenge index I = {I1, I2, ..., I...}. j}, where I i ={s1,s2,...s} n According to formula v i =φ(k,i) generates random numbers V = {V1,V2,…V} for each index. j},in Simultaneously, a random number r1 is generated and assigned to the edge node, while the random number r2 is stored internally. The corresponding edge node's public key is used for PK (player killing). edge The encrypted request, Req = {I, V, r1}, is sent to the corresponding edge node. Each Req is unique, with I and V corresponding one-to-one.

[0084] Batch response process:

[0085] Upon receiving the request, it must first be decrypted using the private key. Each edge node generates signatures R and P based on the challenge index and a random number, and then hides the generated information (R and P information) through a Pedersen commitment. The public key of the TPA is then used for PK. TPA Send RES in encrypted form.

[0086]

[0087] Batch verification process:

[0088] TPA is calculated as follows:

[0089] TPA is to calculate T according to its own label

[0090]

[0091] TPA accumulates the results returned by the edge nodes to generate res, which is the sum of the RES generated by each edge node, and compares it with the T generated by itself to verify whether the following equation is equal.

[0092] T-res≡(r2-j*r1)M(7)

[0093] Return the result to the user. If the equation is true, return True, and if the equation is not true, return False. True proves that the verification result is correct and the data is complete; False proves that the data is inconsistent and the data is damaged.

[0094] The following describes dynamic verification. Dynamic verification refers to the user's addition, deletion, modification and query of the uploaded file, which can more accurately be called support for full dynamic data update.

[0095] At present, there are many structure types supporting full dynamic data update (including insertion, deletion and modification) in existing integrity audit schemes, such as linked list-based structure and index array-based structure. These two structures have advantages and disadvantages. The linked list-based structure can easily insert and delete a data block without affecting other data blocks. However, the efficiency is low when searching for the location of the data block. The index array-based structure is just the opposite. This structure supports fast positioning of data blocks, but the modification of data blocks in this structure is implemented by "deletion first and then insertion", which undoubtedly leads to more calculation and communication overhead.

[0096] As shown in Figure 2 , both the above single verification method and batch verification method can adopt index-linked list storage form, number represents serial number, IN represents index number of data block, time-M is the time of last modification, time-V is the time of last verification, and PR is a pointer to the physical storage location of the data block in the single linked list, as shown in the following figure:

[0097] Insertion operation: suppose a new data block F is inserted at the i-th position of the single linked list. First, operate on the single linked list: store the data block F in any empty physical location, and disconnect the pointer between the data blocks F i-1 and F i . The pointer of F i-1 points to the physical storage location of F, and the pointer of F points to the physical storage location of F ithe physical storage location of F. Then the index table is operated: check whether the PR of the last row of the index table satisfies PR = Null. If PR = Null, PR points to the physical storage location of F. Otherwise, add a row at the end of the index table, make the newly added PR point to the physical storage location of F, and perform +1 operation on IN greater than i.

[0098] Deletion operation: suppose that the data block F at the i-th position in the single chain table is to be deleted i . First, the single chain table is operated: disconnect the pointers between F i-1 and F i , and between F i and F i+1 , so that the pointer of F i-1 points to the physical storage location of F i+1 . Then PR is changed to Null, IN becomes -1, and -1 operation is performed on IN greater than i.

[0099] Modification operation: the data block F i is modified to F, where F i is the data block at the i-th position in the single chain table, it is only necessary to find the physical storage location of F i , and replace its storage value with F, and modify the time corresponding to time-M.

Claims

1. A method for data integrity verification in an edge computing environment, characterized in that, The system to which the method is applied comprises a plurality of user terminals, a third-party auditor and an edge node, which are communicatively connected, and each of which has a public key and a private key; The method comprises an initialization process, a label generation process, an uploading process, a request process, and a response process; In the initialization process, given the initial parameters ec, G, M, H, Φ, f; ec is an elliptic curve, G is a generator on the elliptic curve, M is another point on the elliptic curve; H is a hash for collision prevention, H: (0, 1) * ∈ Zq, is a pseudo-random sequence, Φ is a pseudo-random function, f is a pseudo-random function used by the user end to blind the data; Zq refers to an integer less than q; In the label generation process, the user terminal divides the data file to be verified into n blocks, each block corresponding to an index position, and the user terminal blurs the data blocks and generates corresponding labels for the blurred data blocks; In the uploading process, the user terminal uploads the data file to the edge node and uploads the labels to the third-party auditor; In the request process, the third-party auditor generates a challenge index I, random numbers corresponding to each index, random numbers r1 and r2, encrypts Req with the public key of the edge node, and sends a request to the edge node, wherein Req={I, V, r1}; Req is a request, and V is a set of random numbers; In the response flow, after receiving the request, the edge node first decrypts with the private key, generates the signature R, P according to the challenge index and the random number, and calculates RES, and then uses the public key pk of the third-party auditor TPA The encrypted RES is sent to the third-party auditor; RES is the result returned by the edge node; In the verification process, the third-party auditor calculates the following formula: The third-party auditor compares the result returned by the edge node with T and returns the comparison result to the user terminal; T is a label generated by the blurred data; T i denotes the label of the i-th data block, v i is a random number. 2.The method for data integrity verification in edge computing environment according to claim 1, wherein, The label generation process specifically comprises: Assume that the file uploaded by the user end is F, the user end divides it into n blocks F=(m1, m2, …m n n), where i is the index position corresponding to the data block, and i takes values from 1 to n; the user first selects a random number k∈Z q as the key of the pseudo-random function f, generates a corresponding blinding factor b i = f k (j, name) for each block, and the blinded data block is F i = m i +b i , F i represents the ith block f k of the blinded file; j is the number of edge nodes; a label T=(T1, T2, …T n ) is generated for the blinded data. T i = H(F id || F i )+ sk user wherein F id is a unique identity of a file F, T i represents a label for the i-th data block, name is a file name, and sk user is a user-end private key. 3.The method for data integrity verification in edge computing environment of claim 2, wherein, The request flow is specifically as follows: the third-party auditor first generates a number c of data blocks to be verified at random for the data file to be verified, selects the index sequence blocks meeting the requirements according to the stored time and the time verified in the last round to generate a challenge index I = {s1, s2,..., s c}, c ∈ {1, 2,..., n}; generates a random number according to a calculation formula V i = φ(k, i), one index corresponding to one random number, wherein Zq* refers to a positive integer less than q, each data block corresponds to a group of random numbers, and a number r1 is randomly generated for the edge node, a random number r2 is saved by itself, and the public key pk edge of the edge node is used to encrypt Req = {I, V, r1} and send a request to the corresponding edge node, wherein V is a set of V i . 4.The method for data integrity verification in edge computing environment according to claim 3, wherein, The response flow is specifically as follows: after receiving the request, the edge node decrypts with the private key, generates a signature R and P according to the challenge index and the random number, and then uses the public key pk of the third-party auditor TPA The calculation result RES is sent in an encrypted manner, and the calculation formula is as follows: 5.The method for data integrity verification in edge computing environment according to claim 4, wherein, In the verification process, the third-party auditor verifies whether the following equation is equal: T-RES≡(r2-r1)M The verification result is returned to the user terminal, and True is returned if the equation is established, and False is returned if the equation is not established.

6. The data integrity verification method in the edge computing environment according to any one of claims 1-5, characterized in that, The method uses an index-chain table storage form, each index corresponds to a part of the chain table, and each part contains a sequence number, an index number of a data block, a last modification time, a last verification time, and a pointer to a physical storage location of the data block in the single chain table.

7. A method for batch verification of data integrity in edge computing environment, the system to which the method is applied comprises a plurality of user terminals, a third-party auditor, a plurality of edge nodes, the three parties are communicatively connected, each device has its own public key and private key; characterized in that, The method comprises an initialization process, a label generation process, an uploading process, a batch request process, a batch response process, and a batch verification process; In the initialization process, given the initial parameters ec, G, M, H, Φ, f; ec is an elliptic curve, G is a generator on the elliptic curve, M is another point on the elliptic curve; H is a hash for collision prevention, H: (0, 1) * ∈ Zq, is a pseudo-random sequence, Φ is a pseudo-random function, f is a pseudo-random function for user-side data blinding; Zq refers to an integer less than q; In the label generation process, the user terminal divides the data file to be verified into n blocks, blurs the data blocks, and generates text labels for the blurred data blocks; In the uploading process, the user terminal uploads the data file to the edge node and uploads the data labels to the third-party auditor; In the batch request process, the third-party auditor generates request information for each edge node, randomly generates a number of data blocks to be verified for each data file to be verified, generates a challenge index I, random numbers corresponding to each index, random numbers r1 and r2, encrypts Req with the public key, and sends a request to the corresponding edge node; Req is a request, and V is a set of random numbers; In the batch response process, after receiving the request, the edge node decrypts it with the private key, generates a signature R and P according to the challenge index and the random number, calculates RES, and then encrypts and sends it to the third-party auditor with the public key of the third-party auditor; RES is the result returned by the edge node; In the batch verification process, the third-party auditor calculates the following formula: The third-party auditor compares the sum of the results returned by each edge node with T, and returns the comparison result to the user terminal; T is a label generated by the blurred data; T i denotes the label of the i-th data block, v i is a random number. 8.The method of claim 7, wherein, The batch response process is specifically: the edge node receiving the request first decrypts with a private key, each edge node generates a signature R, P according to a challenge index and a random number, then hides the generated information, calculates RES, and uses the public key pk of the third-party auditor TPA Encrypts and sends to the third-party auditor: the calculation formula is as follows: RES=r1*M+|R+P|*G; Pk user is the user's public key; F id is the unique identity of the file F; F i is the i-th block of the blinded file. 9.The method of claim 8, wherein, In the batched version of the protocol, the third party auditor compares the sum of the results returned by each edge node to T in the following way: verify if the following equation is true: T-res≡(r2-j*r1)M res is the sum of the RES produced by each edge node. If the equation is true, return True to the user, if not, return False.

10. The method for batch verification of data integrity in edge computing environment according to any one of claims 7-9, characterized in that, The method uses an index-linked list storage form, each index corresponds to a part of the linked list, and each part contains the sequence number, the index number of the data block, the time of the last modification, the time of the last verification, and the pointer to the physical storage location of the data block in the single linked list.

Images