Edge Tampering Detection Method, Device, Electronic Device and Storage Medium

Abstract

The present invention relates to the field of communication technologies, and provides an edge tampering detection method, apparatus, electronic device, and storage medium. The method includes: obtaining a detection policy and a feature mapping table of a cloud server; determining shared features of each original feature combination in an original data sequence, and obtaining multiple pieces of camouflaged data for each original feature combination based on the feature mapping table; inserting the multiple pieces of camouflaged data for each original feature combination into the original data sequence based on the detection policy to obtain a target data sequence; and forwarding the target data sequence to the cloud server through an edge computing system for the cloud server to perform edge tampering detection based on the target data sequence. The edge tampering detection method provided by the embodiments of the present invention converts shared features into camouflaged data and inserts them into the original data sequence, and realizes common edge tampering detection based on the original data, shared features, and camouflaged data in the original feature combination, thereby improving the reliability of edge tampering detection.

Description

Technical Field

[0001] The present invention relates to the field of communication technologies, and in particular, to a method, device, electronic device, and storage medium for edge tampering detection. Background Art

[0002] Currently, in the data preprocessing of edge-cloud collaboration, the edge executes part of the cloud's computing tasks and uploads the preprocessing results to the cloud. However, the content and format of the edge preprocessing results may be different from the original data. For example, the edge outputs input image data as text data.

[0003] Existing edge tampering detection methods focus on detecting whether the data transmission process, the outsourcing computing process of two-party interaction, etc. have been tampered with. However, the existing edge tampering detection methods do not well solve the edge tampering detection of the three-party interaction between the edge, the cloud, and the user terminal, resulting in low reliability of edge tampering detection. Summary of the Invention

[0004] The present invention provides a method, device, electronic device, and storage medium for edge tampering detection, aiming to improve the reliability of edge tampering detection.

[0005] In a first aspect, the present invention provides an edge tampering detection method, which is applied to a user terminal. The edge tampering detection method includes:

[0006] Obtaining a detection strategy and a feature mapping table through an edge computing system; the detection strategy and the feature mapping table are retrieved and fed back by the edge computing system from a cloud server;

[0007] Determining the shared features of each original feature combination in the original data sequence generated when performing an edge task, and obtaining multiple disguised data for each original feature combination based on the feature mapping table; the feature mapping table includes the mapping relationship between each shared feature and multiple disguised data;

[0008] Based on the detection strategy, inserting the multiple disguised data of each original feature combination into the original data sequence to obtain a target data sequence;

[0009] Forwarding the target data sequence to the cloud server through the edge computing system for the cloud server to perform edge tampering detection based on the target data sequence.

[0010] In one embodiment, the detection strategy includes a feature position selection strategy;

[0011] The step of inserting the multiple disguised data of each original feature combination into the original data sequence based on the detection strategy to obtain a target data sequence includes:

[0012] Based on the above-mentioned feature position selection strategy, multiple camouflage data of each original feature combination are inserted into different feature positions of the original data sequence to obtain the target data sequence.

[0013] In the original data sequence generated when performing edge tasks, to determine the shared features of each original feature combination and obtain multiple camouflage data of each original feature combination based on the feature mapping table, the steps include:

[0014] Determine the original data features of each original data in the original data sequence, and split the original data features in the original data sequence into multiple original feature combinations;

[0015] Determine the shared features of the original data features in each original feature combination;

[0016] Based on the feature mapping table, convert the shared features of each original feature combination into multiple camouflage data of each original feature combination.

[0017] The detection strategy includes a feature construction strategy;

[0018] After determining the shared features of the original data features in each original feature combination, the following steps are further included:

[0019] Based on the feature construction strategy, construct a preset mapping relationship between the original data features in each original feature combination and the shared features of each original feature combination;

[0020] Carry the preset mapping relationship of each original feature combination in the target data sequence and send it to the cloud server. After the cloud server receives the target data sequence sent by the user terminal, perform edge tampering detection on the target data sequence through the preset mapping relationship of each original feature combination determined by the feature construction strategy.

[0021] In a second aspect, the present invention provides an edge tampering detection method, which is applied to a cloud server. The edge tampering detection method includes:

[0022] Receive the target data sequence sent by the edge computing system; multiple camouflage data of each original feature combination at different feature positions are carried in the target data sequence;

[0023] Based on the detection strategy, determine multiple camouflage data of each original feature combination at different feature positions from the target data sequence;

[0024] Map multiple camouflaged data of each original feature combination based on the feature mapping table to obtain the shared features of each original feature combination; the feature mapping table includes the mapping relationship between each shared feature and multiple camouflaged data;

[0025] Perform edge tampering detection based on the shared features of each original feature combination and the original data features in each original feature combination.

[0026] In one embodiment, the performing edge tampering detection based on the shared features of each original feature combination and the original data features in each original feature combination includes:

[0027] Determine the preset mapping relationship of each original feature combination based on the feature construction strategy;

[0028] Determine whether the original data features in each original feature combination and the shared features of each original feature combination satisfy the preset mapping relationship of each original feature combination;

[0029] If the original data features in each original feature combination and the shared features of each original feature combination all satisfy the preset mapping relationship of each original feature combination, determine that the data at the edge is not tampered;

[0030] If there is at least one target original feature combination, determine that the data at the edge is tampered, where the original data features in the target original feature combination and the shared features of the target original feature combination do not satisfy the preset mapping relationship of the target original feature combination

[0031] In a third aspect, the present invention provides an edge tampering detection device, which is applied to a user terminal. The edge tampering detection device includes:

[0032] An acquisition module, configured to acquire a detection strategy and a feature mapping table through an edge computing system; the detection strategy and the feature mapping table are retrieved and fed back by the edge computing system from a cloud server;

[0033] A conversion module, configured to determine the shared features of each original feature combination in the original data sequence generated when performing an edge task, and obtain multiple camouflaged data of each original feature combination based on the feature mapping table; the feature mapping table includes the mapping relationship between each shared feature and multiple camouflaged data;

[0034] An import module, based on the detection strategy, inserts multiple camouflaged data of each original feature combination into the original data sequence to obtain a target data sequence;

[0035] A sending module, configured to forward the target data sequence to the cloud server through the edge computing system, so that the cloud server performs edge tampering detection based on the target data sequence.

[0036] In a fourth aspect, the present invention provides an edge tampering detection device, which is applied to a user terminal. The edge tampering detection device includes:

[0037] A receiving module, configured to receive a target data sequence sent by an edge computing system; the target data sequence carries a plurality of camouflage data of each original feature combination at different feature positions;

[0038] An extraction module, configured to determine, from the target data sequence based on a detection strategy, a plurality of camouflage data of each original feature combination at different feature positions;

[0039] A mapping module, configured to map the plurality of camouflage data of each original feature combination based on a feature mapping table to obtain a shared feature of each original feature combination; the feature mapping table includes a mapping relationship between each shared feature and a plurality of camouflage data;

[0040] An edge detection module, configured to perform edge tampering detection based on the shared feature of each original feature combination and the original data feature in each original feature combination.

[0041] In a fifth aspect, the present invention further provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the program, the edge tampering detection methods described in the first aspect and the second aspect are implemented.

[0042] In a sixth aspect, the present invention further provides a non-transitory computer-readable storage medium, which includes a computer program. When the computer program is executed by the processor, the edge tampering detection methods described in the first aspect and the second aspect are implemented.

[0043] In a seventh aspect, the present invention further provides a computer program product, which includes a computer program. When the computer program is executed by the processor, the edge tampering detection methods described in the first aspect and the second aspect are implemented.

[0044] The edge tampering detection method, device, electronic device and storage medium provided by the present invention obtain a detection strategy and a feature mapping table through an edge computing system; the detection strategy and the feature mapping table are retrieved and fed back by the edge computing system from a cloud server; determine the shared features of each original feature combination in the original data sequence generated when performing an edge task, and obtain multiple disguised data for each original feature combination based on the feature mapping table; the feature mapping table includes the mapping relationship between each shared feature and multiple disguised data; based on the detection strategy, insert the multiple disguised data of each original feature combination into the original data sequence to obtain a target data sequence; forward the target data sequence to the cloud server through the edge computing system for the cloud server to perform edge tampering detection based on the target data sequence.

[0045] During the process of edge tampering detection, the shared features of each original feature combination are converted into disguised data and inserted into the original data sequence sensed by the terminal, and sent to the edge together with the original data in the original data sequence. If any one of the original data or the disguised data in the edge is tampered with, the association between the shared features and the relevant original data features will be destroyed, which also means that the edge is unreliable. Therefore, common edge tampering detection is achieved through the original data, its shared features, and the disguised data in the original feature combination, improving the reliability of edge tampering detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the technical solutions of the present invention, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0047] Figure 1 is one of the flow diagrams of the edge tampering detection method provided by the present invention;

[0048] Figure 2 is the schematic diagram of the construction of the disguised data provided by the present invention;

[0049] Figure 3 is the schematic diagram of data position selection and data position distribution provided by the present invention;

[0050] Figure 4 is the overall framework schematic diagram of the edge tampering detection provided by the present invention;

[0051] Figure 5 is the second flow diagram of the edge tampering detection method provided by the present invention;

[0052] Figure 6 is one of the structural diagrams of the edge tampering detection device provided by the present invention;

[0053] Figure 7 It is the second structural schematic diagram of the edge tampering detection device provided by the present invention;

[0054] Figure 8 It is the structural schematic diagram of the electronic device provided by the present invention. Specific embodiments

[0055] To make the objectives, technical solutions and advantages of the present invention clearer, the technical solutions in the present invention will be clearly and completely described below with reference to the accompanying drawings in the present invention. Apparently, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present invention without making creative efforts shall fall within the protection scope of the present invention.

[0056] The embodiments of the present invention provide embodiments of the edge tampering detection method. It should be noted that although the logical order is shown in the flowchart, under certain data, the steps shown or described may be completed in a different order than here.

[0057] Refer to Figure 1 , Figure 1 It is one of the flowchart diagrams of the edge tampering detection method provided by the present invention. The edge tampering detection method provided by the embodiments of the present invention includes:

[0058] Step 101, obtaining a detection strategy and a feature mapping table through an edge computing system;

[0059] Step 102, determining the shared features of each original feature combination in the original data sequence generated when performing an edge task, and obtaining multiple camouflage data for each original feature combination based on the feature mapping table;

[0060] Step 103, inserting the multiple camouflage data of each original feature combination into the original data sequence based on the detection strategy to obtain a target data sequence;

[0061] Step 104, forwarding the target data sequence to the cloud server through the edge computing system for the cloud server to perform edge tampering detection based on the target data sequence.

[0062] In this embodiment, the user terminal is used as the execution subject to describe the edge tampering detection method. In actual operation, the execution subject can be a mobile user terminal and a non-mobile user terminal.

[0063] It should be noted that when the user terminal registers to the edge computing system, it needs to negotiate a set of detection policies with the cloud server of the edge computing system. Among them, the detection policies include a feature construction policy and a feature location selection policy. The feature construction policy is a policy for constructing the mapping relationship between the original data features and the shared features, and the feature location selection policy is a policy for inserting multiple disguised data into the original data sequence.

[0064] Therefore, when the user terminal executes some edge tasks, it needs to send a registration instruction to the edge computing system. After receiving the registration instruction sent by the user terminal, the edge computing system establishes a communication connection between the user terminal and the cloud server, so that the user terminal and the cloud server negotiate the detection policies, that is, the cloud server and the user terminal negotiate the feature construction policy and the feature location selection policy. Further, the cloud server returns the negotiated detection policies to the user terminal through the edge computing system. Further, the cloud server also returns the feature mapping table to the user terminal, where the feature mapping table includes the mapping relationship between each shared feature and multiple disguised data. At the same time, the cloud server needs to determine the feature range of each data when the user terminal executes some edge tasks. Therefore, the user terminal can obtain the detection policies and the feature mapping table returned by the cloud server through the registration instruction.

[0065] Further, after the user terminal obtains the detection policies and the feature mapping table, it determines the original data sequence generated when executing some edge tasks. There are n original data in the original data sequence, and each original data corresponds to its original data feature.

[0066] Further, the user terminal determines the common feature of the n original data features, and constructs the mapping relationship between the n original data features and their common feature according to the detection policies. In one embodiment, if there is a common feature among the n original data features, the constructed mapping relationship is an n - 1 mapping relationship. Further, the user terminal converts the common feature of the n original data features into disguised data according to the feature mapping table.

[0067] In one embodiment, referring to Figure 2 it can be seen that when performing some edge tasks, the n original data □ are converted into n original data features ○, and the n original data features are determined to be converted into shared features ■. The shared features are converted into disguised data ● (false data) according to the feature mapping table, and the false data are inserted into the tail of the n original data to obtain a new data sequence.

[0068] In one embodiment, the user terminal splits n original data features into multiple original feature combinations. To avoid unreliable edges, the terminal data is honestly preprocessed once, and then the correct but outdated preprocessing results are sent to the cloud server multiple times. In the embodiments of the present invention, the length of the divided original data segment is related to time, that is, it is not a fixed length. That is to say, in the embodiments of the present invention, the number of features of the original data features in each original feature combination may be equal or unequal.

[0069] It should be noted that there are multiple different original feature combinations for n original data features. If each original feature combination is mapped to a unique fake data, the volume of the feature mapping table will be extremely large. Therefore, in the embodiments of the present invention, an original feature combination of n original data features is mapped from a unique fake data to multiple fake data and inserted into different feature positions. At this time, a fake data can represent different feature information at different feature positions, and the fake data at multiple feature positions cooperate to express shared features. That is, multiple feature insertion positions reduce the volume of the feature mapping table.

[0070] Therefore, in the embodiments of the present invention, the user terminal determines the shared features of each original feature combination in the original data sequence, and converts the shared features of each original feature combination into multiple fake data of each original feature combination according to the feature mapping table.

[0071] Further, the user terminal inserts multiple fake data of each original feature combination into different feature positions of the original data sequence according to the detection strategy to obtain the target data sequence. It should be noted that the fake data at different feature positions in the target data sequence is set according to the detection strategy, which can help the cloud server perform edge tampering detection. In the embodiments of the present invention, the normal target data sequence is divided into multiple data segments, a shared feature is constructed for each data segment and appended to the end of the data segment. Referring to Figure 3 the upper part of the content, the fake data of each segment of the original data is inserted at the end of the original data of that segment.

[0072] Further, the user terminal sends the target data sequence to the edge computing system, and the edge computing system forwards the target data sequence to the cloud server. The cloud server receives the target data sequence sent by the edge computing system and determines multiple fake data of each original feature combination at different feature positions from the target data sequence according to the feature position selection strategy.

[0073] Further, the cloud server maps multiple fake data of each original feature combination according to the feature mapping table to obtain the shared features of each original feature combination.

[0074] Further, the cloud server performs edge tampering detection based on the shared features of each original feature combination and the original data features in each original feature combination, that is, determines whether the edge has modified the data in the target data sequence sent by the user terminal to the cloud server.

[0075] In one embodiment, referring to Figure 4 , when the user terminal executes some edge tasks, it needs to send a registration instruction to the edge computing system. After receiving the registration instruction sent by the user terminal, the edge computing system establishes a communication connection between the user terminal and the cloud server, so that the user terminal and the cloud server negotiate a detection strategy, that is, the cloud server and the user terminal negotiate a feature construction strategy and a feature position selection strategy. Further, the cloud server returns the negotiated detection strategy to the user terminal through the edge computing system. Further, the cloud server also returns the feature mapping table to the user terminal, where the feature mapping table includes the mapping relationship between each shared feature and multiple disguised data. At the same time, the cloud server needs to determine the feature range of each data when the user terminal executes some edge tasks. Therefore, the user terminal can obtain the detection strategy and the feature mapping table returned by the cloud server through the registration instruction.

[0076] The user terminal determines the common features of n original data features, converts the common features into disguised data and inserts them into the original data sequence to obtain the target data sequence, and forwards the target sequence data to the edge computing system. Further, the edge computing system forwards the target sequence data to the cloud server. The cloud server receives the target data sequence sent by the edge computing system, and determines the disguised data at different feature positions from the target data sequence according to the feature position selection strategy. Further, the cloud server maps the disguised data according to the feature mapping table to obtain the shared features. Further, the cloud server performs edge tampering detection based on the shared features and the original data features in each original feature combination.

[0077] The edge tampering detection method provided by the embodiment of the present invention obtains a detection strategy and a feature mapping table through the edge computing system; the detection strategy and the feature mapping table are retrieved and fed back by the edge computing system from the cloud server; determine the shared features of each original feature combination in the original data sequence generated when performing edge tasks, and obtain multiple disguised data of each original feature combination based on the feature mapping table; the feature mapping table includes the mapping relationship between each shared feature and multiple disguised data; based on the detection strategy, insert multiple disguised data of each original feature combination into the original data sequence to obtain the target data sequence; through the edge computing system, forward the target data sequence to the cloud server for the cloud server to perform edge tampering detection based on the target data sequence.

[0078] During the edge tampering detection process, the shared features of each original feature combination are converted into disguised data and inserted into the original data sequence sensed by the terminal. Together with the original data in the original data sequence, they are sent to the edge. If any of the original data or the disguised data at the edge is tampered with, the association between the shared features and the relevant original data features will be destroyed, which also means that the edge is unreliable. Therefore, common edge tampering detection is achieved through the original data, its shared features, and the disguised data in the original feature combination, improving the reliability of edge tampering detection.

[0079] Further, based on the shared features of each original feature combination in the original data sequence determined in step 102 during the execution of the edge task, multiple disguised data of each original feature combination are obtained based on the feature mapping table, including:

[0080] Determine the original data features of each original data in the original data sequence, and split the original data features in the original data sequence into multiple original feature combinations;

[0081] Determine the shared features of the original data features in each original feature combination;

[0082] Based on the feature mapping table, convert the shared features of each original feature combination into multiple disguised data of each original feature combination.

[0083] Specifically, the user terminal determines n original data in the original data sequence for executing part of the edge task, and each original data corresponds to its original data features.

[0084] Further, the user terminal splits the n original data features into multiple original feature combinations. To avoid unreliable edge honest preprocessing of the terminal data once, the correct but outdated preprocessing results are sent to the cloud server multiple times. In the embodiment of the present invention, the length of the divided original data segment is related to time, that is, it is not a fixed length. That is, the number of original data features in each original feature combination in the embodiment of the present invention can be equal or not equal.

[0085] It should be noted that there are multiple different original feature combinations for the n original data features. If each original feature combination is mapped to a unique disguised data (false data), the volume of the feature mapping table will be very large. Therefore, in the embodiment of the present invention, an original feature combination of the n original data features is mapped from a unique false data to multiple false data and inserted into different feature positions. At this time, a false data can represent different feature information at different feature positions, and the false data at multiple feature positions cooperate to express the shared features. That is, multiple feature insertion positions reduce the volume of the feature mapping table.

[0086] Therefore, in the embodiments of the present invention, the user terminal determines the shared features of each original feature combination in the original data sequence, and converts the shared features of each original feature combination into multiple disguised data according to the feature mapping table. In one embodiment, referring to Figure 3 , from Figure 3 , it can be seen from the lower part that the common features of the first data segment (the first original feature combination) are ■...▲ and ●, and the common features ■...▲ and ● are converted into false data 1... false data i... false data m.

[0087] In the process of edge tampering detection in the embodiments of the present invention, the shared features of each original feature combination are converted into disguised data and inserted into the original data sequence sensed by the terminal. If the edge tampers with the disguised data, the association between the shared features and the relevant original data features will be destroyed, which also means that the edge is unreliable. Therefore, common edge tampering detection is achieved through the disguised data, improving the reliability of edge tampering detection. At the same time, the shared features are inserted into the target data sequence as false data without interfering with the normal service of the system.

[0088] Further, after determining the shared features of the original data features in each original feature combination, it further includes:

[0089] Based on the feature construction strategy, construct a preset mapping relationship between the original data features in each original feature combination and the shared features of each original feature combination;

[0090] Carry the preset mapping relationship of each original feature combination in the target data sequence and send it to the cloud server, so that after the cloud server receives the target data sequence sent by the user terminal, it can perform edge tampering detection on the target data sequence through the preset mapping relationship of each original feature combination determined by the feature construction strategy.

[0091] Specifically, the user terminal constructs a preset mapping relationship between the original data features in each original feature combination and the shared features of each original feature combination according to the feature construction strategy. Among them, the preset mapping relationship is determined according to the feature construction strategy. For example, the preset relationship of a certain data segment is that the first data segment corresponds to the shared features ABC, and the second data segment corresponds to the shared features ACDE. At the same time, the user terminal carries the preset mapping relationship of each original feature combination in the target data sequence and sends it to the cloud server.

[0092] The preset mapping relationship of each original feature combination determined by the cloud server according to the feature construction policy. Further, the cloud server determines whether the original data features in each original feature combination and the shared features of each original feature combination satisfy the preset mapping relationship of each original feature combination, where the preset mapping relationship is determined based on the feature construction policy, which is not limited in this embodiment. If it is determined that the original data features in each original feature combination and the shared features of each original feature combination all satisfy the preset mapping relationship of each original feature combination, the cloud server determines that the edge data is not tampered with. If it is determined that there is at least one target original feature combination, and the original data features in the target original feature combination and the shared features of the target original feature combination do not satisfy the preset mapping relationship of the target original feature combination, the cloud server determines that the edge data is tampered with.

[0093] In the process of edge tampering detection in the embodiments of the present invention, the shared features of each original feature combination are converted into disguised data and inserted into the original data sequence sensed by the terminal. If any one of the edge-tampered original data or the disguised data, the association between the shared features and the relevant original data features will be destroyed, which also means that the edge is unreliable. Therefore, the common edge tampering detection is realized through the original data and its shared features in the original feature combination, improving the reliability of edge tampering detection.

[0094] Step 103 records that based on the detection policy, inserting multiple disguised data of each original feature combination into the original data sequence to obtain a target data sequence, including:

[0095] Inserting multiple disguised data of each original feature combination into different feature positions of the original data sequence based on the feature position selection policy to obtain a target data sequence.

[0096] Specifically, the user terminal inserts multiple disguised data of each original feature combination into different feature positions of the original data sequence based on the feature position selection policy to obtain a target data sequence. Refer to Figure 3 The following content, converting the common features of the first data segment ((the first original feature combination)) ■...▲ and ● into false data 1... false data i... false data m, and inserting the false data 1... false data i... false data m into different feature positions of the original data sequence to obtain a target data sequence.

[0097] In the process of edge tampering detection in the embodiments of the present invention, the shared features of each original feature combination are converted into disguised data and inserted into the original data sequence perceived by the terminal, and are sent to the edge together with the original data in the original data sequence. If any of the original data or the disguised data in the edge is tampered with, the association between the shared features and the relevant original data features will be destroyed, which also means that the edge is unreliable. Therefore, common edge tampering detection is achieved through the original data, its shared features, and the disguised data in the original feature combination, improving the reliability of edge tampering detection.

[0098] Referring to Figure 5 , Figure 5 is the second flowchart of the edge tampering detection method provided by the present invention. The edge tampering detection method provided by the embodiments of the present invention includes:

[0099] Step 105, receiving a target data sequence sent by an edge computing system;

[0100] Step 106, determining multiple disguised data of each original feature combination at different feature positions from the target data sequence based on a detection strategy;

[0101] Step 107, mapping the multiple disguised data of each original feature combination based on a feature mapping table to obtain the shared features of each original feature combination;

[0102] Step 108, performing edge tampering detection based on the shared features of each original feature combination and the original data features in each original feature combination.

[0103] After receiving the target data sequence sent by the edge computing system, the cloud server determines multiple disguised data of each original feature combination at different feature positions from the target data sequence according to the feature position selection strategy in the detection strategy and the fact that the target data sequence carries multiple disguised data of each original feature combination at different feature positions.

[0104] Furthermore, the cloud server maps the multiple disguised data of each original feature combination according to the feature mapping table to obtain the shared features of each original feature combination.

[0105] Furthermore, the cloud server performs edge tampering detection according to the shared features of each original feature combination and the original data features in each original feature combination, that is, determines whether the edge has modified the data in the target data sequence sent by the user terminal to the cloud server.

[0106] The embodiments of the present invention achieve the verification of the target data sequence forwarded by the edge without the original data of the user terminal, and determine whether the edge correctly uploads the target data sequence to the cloud server, improving the reliability of edge tampering detection.

[0107] Further, the edge tampering detection based on the shared features of each original feature combination and the original data features in each original feature combination recorded in step 108 includes:

[0108] Determine the preset mapping relationship of each original feature combination based on the feature construction strategy;

[0109] Determine whether the original data features in each original feature combination and the shared features of each original feature combination satisfy the preset mapping relationship of each original feature combination;

[0110] If the original data features in each original feature combination and the shared features of each original feature combination satisfy the preset mapping relationship of each original feature combination, determine that the edge data is not tampered;

[0111] If there is at least one target original feature combination, determine that the edge data is tampered, where the original data features in the target original feature combination and the shared features of the target original feature combination do not satisfy the preset mapping relationship of the target original feature combination.

[0112] Specifically, the cloud server determines whether the original data features in each original feature combination and the shared features of each original feature combination satisfy the preset mapping relationship of each original feature combination according to the feature construction strategy, where the preset mapping relationship is determined based on the feature construction strategy and is not limited in this embodiment.

[0113] If it is determined that the original data features in each original feature combination and the shared features of each original feature combination satisfy the preset mapping relationship of each original feature combination, the cloud server determines that the edge data is not tampered.

[0114] If it is determined that there is at least one target original feature combination, and the original data features in the target original feature combination and the shared features of the target original feature combination do not satisfy the preset mapping relationship of the target original feature combination, the cloud server determines that the edge data is tampered.

[0115] The embodiment of the present invention determines whether the edge correctly uploads the target data sequence to the cloud server through the original data features, shared features, and preset mapping relationship of each original feature combination, improving the reliability of edge tampering detection.

[0116] Further, the edge tampering detection device provided by the present invention corresponds to and refers to the edge tampering detection method provided by the present invention.

[0117] As Figure 6 shown, Figure 6It is one of the structural schematic diagrams of the edge tampering detection device provided by the present invention. The edge tampering detection device is applied to a user terminal, and the edge tampering detection device includes:

[0118] An acquisition module 601, configured to obtain a detection policy and a feature mapping table through an edge computing system; the detection policy and the feature mapping table are retrieved and fed back by the edge computing system from a cloud server;

[0119] A conversion module 602, configured to determine the shared features of each original feature combination in the original data sequence generated when performing an edge task, and obtain multiple camouflage data for each original feature combination based on the feature mapping table; the feature mapping table includes the mapping relationship between each shared feature and multiple camouflage data;

[0120] An import module 603, configured to insert the multiple camouflage data of each original feature combination into the original data sequence based on the detection policy to obtain a target data sequence;

[0121] A sending module 604, configured to forward the target data sequence to the cloud server through the edge computing system for the cloud server to perform edge tampering detection based on the target data sequence.

[0122] Further, the conversion module 602 is further configured to:

[0123] Determine the original data features of each original data in the original data sequence, and split the original data features in the original data sequence into multiple original feature combinations;

[0124] Determine the shared features of the original data features in each original feature combination;

[0125] Convert the shared features of each original feature combination into multiple camouflage data of each original feature combination based on the feature mapping table.

[0126] Further, the edge tampering detection device further includes a construction module, configured to:

[0127] Construct a preset mapping relationship between the original data features in each original feature combination and the shared features of each original feature combination based on the feature construction policy;

[0128] Carry the preset mapping relationship of each original feature combination in the target data sequence and send it to the cloud server, so that after the cloud server receives the target data sequence sent by the user terminal, it can perform edge tampering detection on the target data sequence through the preset mapping relationship of each original feature combination determined by the feature construction policy.

[0129] Further, the import module 603 is further configured to:

[0130] Based on the feature position selection strategy, insert multiple disguised data of each original feature combination into different feature positions of the original data sequence to obtain the target data sequence.

[0131] The specific embodiments of the edge tampering detection device provided by the present invention are basically the same as those of the above-mentioned edge tampering detection method embodiments, and will not be elaborated here.

[0132] As Figure 7 shown, Figure 7 FIG. 2 is a second schematic structural diagram of the edge tampering detection device provided by the present invention. The edge tampering detection device is applied to a user terminal, and the edge tampering detection device includes:

[0133] A receiving module 701, configured to receive a target data sequence sent by an edge computing system; the target data sequence carries multiple disguised data of each original feature combination at different feature positions;

[0134] An extraction module 702, configured to determine multiple disguised data of each original feature combination at different feature positions from the target data sequence based on a detection strategy;

[0135] A mapping module 703, configured to map multiple disguised data of each original feature combination based on a feature mapping table to obtain a shared feature of each original feature combination; the feature mapping table includes a mapping relationship between each shared feature and multiple disguised data;

[0136] An edge detection module 704, configured to perform edge tampering detection based on the shared feature of each original feature combination and the original data feature in each original feature combination.

[0137] Further, the edge detection module 704 is further configured to:

[0138] Determine a preset mapping relationship of each original feature combination based on a feature construction strategy;

[0139] Determine whether the original data feature in each original feature combination and the shared feature of each original feature combination satisfy the preset mapping relationship of each original feature combination;

[0140] If the original data feature in each original feature combination and the shared feature of each original feature combination both satisfy the preset mapping relationship of each original feature combination, it is determined that the edge is not tampered data;

[0141] If there is at least one target original feature combination, it is determined that the data has been edge tampered with, where the original data features in the target original feature combination and the shared features of the target original feature combination do not satisfy the preset mapping relationship of the target original feature combination.

[0142] The specific embodiments of the edge tampering detection device provided by the present invention are basically the same as those of the above-mentioned edge tampering detection methods, and will not be elaborated here.

[0143] Figure 8 An example of the physical structure diagram of an electronic device is shown as Figure 8 As shown, the electronic device may include: a processor 810, a communication interface 820, a memory 830, and a communication bus 840. Among them, the processor 810, the communication interface 820, and the memory 830 complete mutual communication through the communication bus 840. The processor 810 can call the logical instructions in the memory 830 to execute the edge tampering detection method, which is applied to the user terminal and includes:

[0144] Obtain a detection strategy and a feature mapping table through an edge computing system; the detection strategy and the feature mapping table are retrieved and fed back by the edge computing system from a cloud server;

[0145] Determine the shared features of each original feature combination in the original data sequence generated when performing an edge task, and obtain multiple camouflaged data for each original feature combination based on the feature mapping table; the feature mapping table includes the mapping relationship between each shared feature and multiple camouflaged data;

[0146] Based on the detection strategy, insert the multiple camouflaged data of each original feature combination into the original data sequence to obtain a target data sequence;

[0147] Forward the target data sequence to the cloud server through the edge computing system for the cloud server to perform edge tampering detection based on the target data sequence.

[0148] The processor 810 can call the logical instructions in the memory 830 to execute the edge tampering detection method, which is applied to the cloud server and includes:

[0149] Receive the target data sequence sent by the edge computing system; the target data sequence carries multiple camouflaged data of each original feature combination at different feature positions;

[0150] Based on the detection strategy, determine multiple camouflaged data of each original feature combination at different feature positions from the target data sequence;

[0151] Map multiple camouflaged data of each original feature combination based on the feature mapping table to obtain the shared features of each original feature combination; the feature mapping table includes the mapping relationship between each shared feature and multiple camouflaged data.

[0152] Perform edge tampering detection based on the shared features of each original feature combination and the original data features in each original feature combination.

[0153] In addition, when the logical instructions in the above-mentioned memory 830 are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs that can store program codes.

[0154] On the other hand, the present invention also provides a computer program product. The computer program product includes a computer program stored on a non-transitory computer-readable storage medium. The computer program includes program instructions. When the program instructions are executed by a computer, the computer can execute the edge tampering detection method provided by the above-mentioned various methods. This method is applied to a user terminal and includes:

[0155] Obtain a detection strategy and a feature mapping table through an edge computing system; the detection strategy and the feature mapping table are retrieved and fed back by the edge computing system from a cloud server.

[0156] Determine the shared features of each original feature combination in the original data sequence generated when performing an edge task, and obtain multiple camouflaged data of each original feature combination based on the feature mapping table; the feature mapping table includes the mapping relationship between each shared feature and multiple camouflaged data.

[0157] Based on the detection strategy, insert multiple camouflaged data of each original feature combination into the original data sequence to obtain a target data sequence.

[0158] Forward the target data sequence to the cloud server through the edge computing system for the cloud server to perform edge tampering detection based on the target data sequence.

[0159] On the other hand, the present invention also provides a computer program product. The computer program product includes a computer program stored on a non-transitory computer-readable storage medium. The computer program includes program instructions. When the program instructions are executed by a computer, the computer can execute the edge tampering detection method provided by each of the above methods. This method is applied to a cloud server and includes:

[0160] Receive the target data sequence sent by the edge computing system; multiple camouflage data of each original feature combination at different feature positions are carried in the target data sequence;

[0161] Based on the detection strategy, determine multiple camouflage data of each original feature combination at different feature positions from the target data sequence;

[0162] Map the multiple camouflage data of each original feature combination based on the feature mapping table to obtain the shared feature of each original feature combination; the feature mapping table includes the mapping relationship between each shared feature and multiple camouflage data;

[0163] Perform edge tampering detection based on the shared feature of each original feature combination and the original data feature in each original feature combination.

[0164] On another aspect, the present invention also provides a non-transitory computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it is configured to execute the edge tampering detection method provided by each of the above. This method is applied to a user terminal and includes:

[0165] Obtain the detection strategy and the feature mapping table through the edge computing system; the detection strategy and the feature mapping table are retrieved and fed back by the edge computing system from the cloud server;

[0166] Determine the shared feature of each original feature combination in the original data sequence generated when performing the edge task, and obtain multiple camouflage data of each original feature combination based on the feature mapping table; the feature mapping table includes the mapping relationship between each shared feature and multiple camouflage data;

[0167] Based on the detection strategy, insert multiple camouflage data of each original feature combination into the original data sequence to obtain a target data sequence;

[0168] Forward the target data sequence to the cloud server through the edge computing system for the cloud server to perform edge tampering detection based on the target data sequence.

[0169] In another aspect, the present invention also provides a non-transitory computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it is used to execute the edge tampering detection method provided above. The method is applied to a cloud server and includes:

[0170] Receiving a target data sequence sent by an edge computing system; multiple camouflage data at different feature positions of each original feature combination are carried in the target data sequence;

[0171] Based on a detection strategy, determining multiple camouflage data at different feature positions of each original feature combination from the target data sequence;

[0172] Mapping multiple camouflage data of each original feature combination based on a feature mapping table to obtain a shared feature of each original feature combination; the feature mapping table includes a mapping relationship between each shared feature and multiple camouflage data;

[0173] Performing edge tampering detection based on the shared feature of each original feature combination and the original data feature in each original feature combination.

[0174] The processor-readable storage medium can be any available medium or data storage device accessible by the processor, including but not limited to magnetic memories (such as floppy disks, hard disks, magnetic tapes, magneto-optical disks (MO), etc.), optical memories (such as CDs, DVDs, BDs, HVDs, etc.), and semiconductor memories (such as ROMs, EPROMs, EEPROMs, non-volatile memories (NANDFLASH), solid-state drives (SSD)).

[0175] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art can understand and implement it without creative labor.

[0176] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on such an understanding, the essence of the above technical solution or the part that contributes to the prior art can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0177] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments or perform equivalent replacements for some of the technical features. These modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. An edge tampering detection method, which is applied to a user terminal, characterized in that, The edge tampering detection method includes: Obtaining a detection policy and a feature mapping table through an edge computing system; the detection policy and the feature mapping table are retrieved and fed back by the edge computing system from a cloud server; Determining the shared features of each original feature combination in the original data sequence generated during the execution of an edge task, and obtaining multiple disguised data for each original feature combination based on the feature mapping table; the feature mapping table includes the mapping relationship between each shared feature and multiple disguised data; Based on the detection policy, inserting the multiple disguised data of each original feature combination into the original data sequence to obtain a target data sequence; Forwarding the target data sequence to the cloud server through the edge computing system for the cloud server to perform edge tampering detection based on the target data sequence.

2. The edge tampering detection method according to claim 1, characterized in that The detection policy includes a feature position selection policy; The step of, based on the detection policy, inserting the multiple disguised data of each original feature combination into the original data sequence to obtain a target data sequence includes: Based on the feature position selection policy, inserting the multiple disguised data of each original feature combination into different feature positions in the original data sequence to obtain the target data sequence.

3. The edge tampering detection method according to claim 1, characterized in that The step of determining the shared features of each original feature combination in the original data sequence generated during the execution of an edge task, and obtaining multiple disguised data for each original feature combination based on the feature mapping table includes: Determining the original data features of each original data in the original data sequence, and splitting the original data features in the original data sequence into multiple original feature combinations; Determining the shared features of the original data features in each original feature combination; Based on the feature mapping table, converting the shared features of each original feature combination into multiple disguised data of each original feature combination.

4. The edge tampering detection method according to any one of claims 1 to 3, characterized in that, The detection policy includes a feature construction policy; After determining the shared features of the original data features in each original feature combination, it further includes: Based on the feature construction policy, constructing a preset mapping relationship between the original data features in each original feature combination and the shared features of each original feature combination; Carrying the preset mapping relationship of each original feature combination in the target data sequence and sending it to the cloud server, so that after the cloud server receives the target data sequence sent by the user terminal, it performs edge tampering detection on the target data sequence through the preset mapping relationship of each original feature combination determined by the feature construction policy.

5. A method for edge tampering detection, which is applied to a cloud server, and is characterized in that, The edge tampering detection method includes: Receiving a target data sequence sent by an edge computing system; the target data sequence carries multiple disguised data of each original feature combination at different feature positions; Based on the detection policy, determining the multiple disguised data of each original feature combination at different feature positions from the target data sequence; Map multiple camouflaged data of each original feature combination based on the feature mapping table to obtain the shared features of each original feature combination; the feature mapping table includes the mapping relationship between each shared feature and multiple camouflaged data. Perform edge tampering detection based on the shared features of each original feature combination and the original data features in each original feature combination.

6. The edge tampering detection method according to claim 5, characterized in that The performing edge tampering detection based on the shared features of each original feature combination and the original data features in each original feature combination includes: Determine the preset mapping relationship of each original feature combination based on the feature construction strategy. Determine whether the original data features in each original feature combination and the shared features of each original feature combination satisfy the preset mapping relationship of each original feature combination. If the original data features in each original feature combination and the shared features of each original feature combination all satisfy the preset mapping relationship of each original feature combination, determine that the edge data is not tampered. If there is at least one target original feature combination, determine that the edge data is tampered, where the original data features in the target original feature combination and the shared features of the target original feature combination do not satisfy the preset mapping relationship of the target original feature combination.

7. An edge tampering detection device, which is applied to a user terminal, and is characterized in that The edge tampering detection device includes: An acquisition module, configured to acquire a detection strategy and a feature mapping table through an edge computing system; the detection strategy and the feature mapping table are retrieved and fed back by the edge computing system from a cloud server. A conversion module, configured to determine the shared features of each original feature combination in the original data sequence generated when performing an edge task, and obtain multiple camouflaged data of each original feature combination based on the feature mapping table; the feature mapping table includes the mapping relationship between each shared feature and multiple camouflaged data. An import module, based on the detection strategy, inserts multiple camouflaged data of each original feature combination into the original data sequence to obtain a target data sequence. A sending module, configured to forward the target data sequence to the cloud server through the edge computing system for the cloud server to perform edge tampering detection based on the target data sequence.

8. An edge tampering detection device, which is applied to a user terminal, and is characterized in that, The edge tampering detection device includes: A receiving module, configured to receive the target data sequence sent by the edge computing system; the target data sequence carries multiple camouflaged data of each original feature combination at different feature positions. An extraction module, configured to determine multiple camouflaged data of each original feature combination at different feature positions from the target data sequence based on the detection strategy. A mapping module, configured to map multiple camouflaged data of each original feature combination based on the feature mapping table to obtain the shared features of each original feature combination; the feature mapping table includes the mapping relationship between each shared feature and multiple camouflaged data. An edge detection module, configured to perform edge tampering detection based on the shared features of each original feature combination and the original data features in each original feature combination.

9. An electronic device, the electronic device includes a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, When the processor executes the computer program, the edge tampering detection method according to any one of claims 1 to 6 is implemented.

10. A non-transitory computer-readable storage medium, the non-transitory computer-readable storage medium comprising a computer program, characterized in that, When the computer program is executed by a processor, the edge tampering detection method according to any one of claims 1 to 6 is implemented.

Images