A method and apparatus for processing data uploaded by a user terminal

Abstract

The application discloses a method and device for processing data uploaded by a user terminal, and aims to improve the privacy protection of sensitive information such as data content uploaded by the user terminal. The method provided by the application comprises the following steps: when a security detection request of data content to be uploaded by a user terminal is received, an algorithm power percentage coefficient is allocated to the user terminal, wherein the algorithm power percentage coefficient is the percentage of the total algorithm power of an edge server allocated to the security detection of the data content to be uploaded by the user terminal; the current security detection task allocation strategy is determined by using the security detection related calculation amount information of the user terminal and the algorithm power percentage coefficient, and the security detection task allocation strategy is used to indicate the edge server performing the security detection task on the data content to be uploaded by the user terminal.

Description

Technical Field

[0001] This application relates to the field of user privacy protection technology, and in particular to a method and apparatus for processing data uploaded by user terminals. Background Technology

[0002] With the development of multimedia and network technologies, short video services, live streaming, and video-based social networking have experienced rapid growth. This has led to insufficient cloud computing power and increased pressure on network communication load. To ensure the legality and health of online content, existing industries typically deploy sensitive information detection algorithms on software platforms to check the short video / live streaming content uploaded by each user. However, this approach is increasingly constrained by the concurrency of cloud algorithms and the pressure on network services, especially given the ever-increasing number of video services and users. Some software service providers have attempted to deploy detection algorithms within apps, allowing devices to perform edge (terminal) detection using a pruned and distilled lightweight model. However, the accuracy of lightweight models differs significantly from cloud models, and edge-based sensitive information detection is inevitably limited by edge computing power and increases device energy consumption. Summary of the Invention

[0003] This application provides a method and apparatus for processing data uploaded by user terminals, which improves the privacy protection of sensitive information such as data content uploaded by user terminals, and improves the efficiency of sensitive information detection, avoiding the problems of insufficient computing power and low detection efficiency caused by data concurrency during peak periods.

[0004] On the edge server side, this application embodiment provides a method for processing data uploaded by a user terminal, including:

[0005] When a security detection request is received from a user terminal for the data content to be uploaded, a computing power percentage coefficient is allocated to the user terminal. The computing power percentage coefficient is the percentage of the total computing power of the edge server that is allocated to the user terminal for security detection of the data content to be uploaded.

[0006] Using the computational load information related to security detection of the user terminal and the computing power percentage coefficient, a current security detection task allocation strategy is determined. The security detection task allocation strategy is used to instruct the edge server to perform security detection tasks on the data content to be uploaded by the user terminal.

[0007] This method enables edge servers to allocate a percentage of computing power to a user terminal when receiving a security detection request for data content to be uploaded. By utilizing the user terminal's computational load information related to security detection, along with the percentage of computing power, a current security detection task allocation strategy is determined. This strategy prevents the user terminal or a third party from predicting which edge server will be responsible for sensitive information detection, thereby improving privacy protection for sensitive information such as data uploaded by the user terminal. Furthermore, by using the user terminal's computational load information related to security detection, along with the percentage of computing power, the current security detection task allocation strategy can allocate the user terminal to the globally optimal edge server (i.e., an edge alliance composed of multiple edge servers) to perform the security detection task. This also improves the efficiency of sensitive information detection and avoids the problems of insufficient computing power and low detection efficiency caused by data concurrency during peak periods.

[0008] In some embodiments, the current security detection task allocation strategy is determined using the computational load information related to security detection of the user terminal and the computing power percentage coefficient, including:

[0009] Using the computational load information related to the security detection of the user terminal and the computing power percentage coefficient, calculate the user service benefits generated by the local edge server for the user terminal that sends the security detection request to the local edge server;

[0010] Using a penalty function, iterate through combinations of multiple user terminals and multiple edge servers, and determine the sum of user service benefits for each combination. Then, use the sum of user service benefits for each combination to determine the set of computing power percentage coefficients that maximize the current global revenue.

[0011] Using the set of computing power percentage coefficients, the current security detection task allocation strategy is determined. The security detection task allocation strategy includes: the correspondence between edge servers and user terminals, and the computing power percentage coefficient allocated by the edge server to the corresponding user terminal.

[0012] In some embodiments, the method further includes:

[0013] According to the security detection task allocation strategy, determine the user terminal corresponding to the local edge server;

[0014] Send a security detection response to the user terminal corresponding to the local edge server, and obtain the data content uploaded by the user terminal corresponding to the local edge server;

[0015] The data uploaded by the user terminal corresponding to the local edge server is encoded using a preset method and then uploaded to the cloud server.

[0016] In some embodiments, the data content uploaded by the user terminal corresponding to the local edge server is encoded using a preset method, including:

[0017] Using the current local time and the number of members in the edge alliance, a binary key of a preset length is generated;

[0018] The binary encoding of the data content uploaded by the user terminal corresponding to the local edge server is XORed with the binary key bitwise to form the encrypted data content;

[0019] The decryption key, the encrypted data content, the preset identifier check character, and the first information encoding block, the second information encoding block, and the third information encoding block are used to generate encoded data according to a preset format.

[0020] In some embodiments, the first information encoding block is generated using the following information:

[0021] The edge servers in the security detection task allocation strategy use preset natural language processing, voice, and image sensitive information detection models to detect the text, audio, and video in the data content provided by the corresponding user terminals in the security detection task allocation strategy, and obtain the detection results.

[0022] The second information encoding block is generated using the following information:

[0023] The number of security detection requests processed by each user terminal in the security detection task allocation strategy;

[0024] In the security detection task allocation strategy, each user terminal corresponds to 100*α, where α represents the computing power percentage coefficient.

[0025] The user terminal's unique identifier in the security detection task allocation strategy;

[0026] The third information encoding block is generated using the following information:

[0027] The edge server's region encoding in the security detection task allocation strategy;

[0028] The average available computing power of edge servers in the security detection task allocation strategy;

[0029] The edge servers in the security detection task allocation strategy are numbered in the network node structure of the edge alliance.

[0030] Accordingly, in the cloud, this application provides a method for processing data uploaded by a user terminal, including:

[0031] The system receives an encoded block sent by an edge server, the encoded block being obtained by the edge server encoding data content from a user terminal; wherein the data content is data content that has passed security testing by the edge server; and the security testing is performed according to a determined security testing task allocation strategy, the security testing task allocation strategy being determined using the method described above on the edge server side.

[0032] The encoded block is then decoded.

[0033] In some embodiments, the decoding process includes:

[0034] Determine whether the header encoding of the encoded block is a preset encoding value used to indicate that the data content has passed the security check by the edge server;

[0035] If so, then search the encoding block for the existence of a first preset identifier check character and a second preset identifier check character according to the preset position; if so, then obtain the encoded value between the first preset identifier check character and the second preset identifier check character.

[0036] When it is confirmed that the last bit of the encoded value between the first preset identifier checksum and the second preset identifier checksum is the number of the edge server in the preset edge alliance, a preset number of encoded values ​​are read from the header encoding as a key to decrypt the encrypted data content from the user terminal, thereby obtaining the decrypted data content. The encrypted data content from the user terminal is located between the key and the first preset identifier checksum in the encoding block.

[0037] Accordingly, on the terminal side, the present application provides a method for a user terminal to upload data, including:

[0038] When data content needs to be uploaded to the network side, a security detection request for the data content to be uploaded is sent to the edge server. This is used to instruct the edge server to determine the security detection task allocation strategy according to the security detection request and the method described above on the edge server side. This includes information about the edge server that performs the security detection task for the data content on the local user terminal, and the percentage of computing power allocated by the edge server that performs the security detection task for the local user terminal to the local user terminal.

[0039] The system receives a response from the edge server that performs the security detection task for the local user terminal, and sends the data content to the edge server that performs the security detection task for the local user terminal.

[0040] In some embodiments, before sending the security detection request, the method further includes:

[0041] The computational load information related to security detection of the local user terminal is sent to the edge server. The computational load information includes: the computing power required to perform the security detection task, the size of the data content, and the computational latency required for the local user terminal to perform the security detection task.

[0042] Another embodiment of this application provides a computing device including a memory and a processor, wherein the memory is used to store program instructions, and the processor is used to call the program instructions stored in the memory and execute any of the methods described above according to the obtained program.

[0043] Furthermore, according to embodiments, for example, a computer program product for a computer is provided, which includes software code portions that, when the product is run on the computer, perform the steps of the methods defined above. The computer program product may include a computer-readable medium on which the software code portions are stored. Furthermore, the computer program product may be directly loaded into the computer's internal memory and / or sent via a network through at least one of an upload process, a download process, and a push process.

[0044] Another embodiment of this application provides a computer-readable storage medium storing computer-executable instructions for causing the computer to perform any of the methods described above. Attached Figure Description

[0045] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0046] Figure 1 This is a schematic diagram illustrating the content architecture of the technical solution provided in the embodiments of this application;

[0047] Figure 2 This is a schematic diagram of the end-to-cloud architecture provided in an embodiment of this application;

[0048] Figure 3 This is a schematic diagram illustrating the services that the cloud server provided in the embodiments of this application can offer;

[0049] Figure 4 The edge server of the end-to-cloud architecture middle layer provided in the embodiments of this application can be located on the base station side, and the same base station can contain multiple edge servers.

[0050] Figure 5 This is a schematic diagram of the end-to-cloud architecture provided in an embodiment of this application;

[0051] Figure 6 A schematic diagram illustrating the uploading of data by an unauthorized user terminal in an end-to-cloud architecture, as provided in an embodiment of this application.

[0052] Figure 7 A schematic diagram illustrating the process of video content processing by an edge server as provided in this embodiment of the application;

[0053] Figure 8 This application provides a schematic diagram illustrating the specific process for determining user service benefits using an edge server in an embodiment of the present application.

[0054] Figure 9 This is a schematic diagram of the encoded data format provided in an embodiment of this application;

[0055] Figure 10 This is a schematic diagram of the coding block of the algorithm allocation scheme in an embodiment of this application;

[0056] Figure 11 This is a schematic diagram of the node information encoding block in an embodiment of this application;

[0057] Figure 12 A schematic diagram of the edge server numbering provided in the embodiments of this application;

[0058] Figure 13 A schematic diagram illustrating the network node structure numbering of the edge server in the edge alliance, provided in an embodiment of this application;

[0059] Figure 14 This is a schematic diagram of the data content upload processing flow provided in the embodiments of this application;

[0060] Figure 15 A flowchart illustrating a method for processing data uploaded by a user terminal on the edge server side, as provided in an embodiment of this application;

[0061] Figure 16 A flowchart illustrating a method for processing data uploaded by a user terminal on a cloud server side, as provided in an embodiment of this application;

[0062] Figure 17 A flowchart illustrating a method for processing data uploaded by a user terminal on the user terminal side, as provided in an embodiment of this application;

[0063] Figure 18 A schematic diagram of the structure of a computing device (e.g., a network-side device or a user terminal) provided for an embodiment of this application;

[0064] Figure 19 A schematic diagram of the structure of a network-side device (e.g., an edge server, a base station, or a cloud server) provided in an embodiment of this application;

[0065] Figure 20 This is a schematic diagram of the structure of a user terminal provided in an embodiment of this application. Detailed Implementation

[0066] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0067] This application provides a method and apparatus for processing data uploaded by user terminals, which improves the privacy protection of sensitive information such as data content uploaded by user terminals, and improves the efficiency of sensitive information detection, avoiding the problems of insufficient computing power and low detection efficiency caused by data concurrency during peak periods.

[0068] The method and apparatus are based on the same concept of the application. Since the methods and apparatus solve problems in similar ways, the implementation of the apparatus and methods can refer to each other, and the repeated parts will not be described again.

[0069] The terms "first," "second," etc. (if present) in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0070] The following examples and embodiments are to be understood as illustrative only. While this specification may refer to "a," "an," or "some" examples or embodiments in several places, this does not mean that every such reference relates to the same example or embodiment, nor does it mean that the feature applies only to a single example or embodiment. Individual features of different embodiments may also be combined to provide other embodiments. Furthermore, terms such as "comprising" and "including" should be understood not to limit the described embodiments to consisting only of those features mentioned; such examples and embodiments may also include features, structures, units, modules, etc., not specifically mentioned.

[0071] The technical solutions provided in this application can be applied to various systems, especially 5G systems. For example, applicable systems include Global System for Mobile communication (GSM), Code Division Multiple Access (CDMA), Wideband Code Division Multiple Access (WCDMA) General Packet Radio Service (GPRS), Long Term Evolution (LTE), LTE Frequency Division Duplex (FDD), LTE Time Division Duplex (TDD), Universal Mobile Telecommunication System (UMTS), Worldwide Interoperability for Microwave Access (WiMAX), 5G systems, and 5G NR systems. All of these systems include terminal equipment and network equipment.

[0072] The terminal devices involved in the embodiments of this application can be devices that provide voice and / or data connectivity to users, handheld devices with wireless connectivity, or other processing devices connected to a wireless modem. The names of the terminal devices may differ in different systems; for example, in a 5G system, a terminal device can be called User Equipment (UE). Wireless terminal devices can communicate with one or more core networks via the RAN. Wireless terminal devices can be mobile terminal devices, such as mobile phones (or "cellular" phones) and computers with mobile terminal devices, for example, portable, pocket-sized, handheld, computer-embedded, or vehicle-mounted mobile devices that exchange voice and / or data with the radio access network. Examples include Personal Communication Service (PCS) phones, cordless phones, Session Initiated Protocol (SIP) phones, Wireless Local Loop (WLL) stations, Personal Digital Assistants (PDAs), and other devices. Wireless terminal equipment can also be referred to as a system, subscriber unit, subscriber station, mobile station, mobile station, remote station, access point, remote terminal, access terminal, user terminal, user agent, or user device, but is not limited to these terms in the embodiments of this application.

[0073] The network devices involved in this application include edge servers, base stations, and cloud servers, wherein the edge servers can be located in the base stations. The base stations can include multiple cells. Depending on the specific application, a base station can also be called an access point, or it can refer to a device in the access network that communicates with wireless terminal devices through one or more sectors on the air interface, or other names. The network devices can be used to convert received air frames to and from Internet Protocol (IP) packets, acting as routers between wireless terminal devices and the rest of the access network, where the rest of the access network may include an Internet Protocol (IP) communication network. The network devices can also coordinate the attribute management of the air interface. For example, the network equipment involved in the embodiments of this application can be a base transceiver station (BTS) in Global System for Mobile Communications (GSM) or Code Division Multiple Access (CDMA), or a node B in Bandwidth Code Division Multiple Access (WCDMA), or an evolved Node B (eNB or e-NodeB) in a Long Term Evolution (LTE) system, a 5G base station in a 5G network architecture (next generation system), or a Home evolved Node B (HeNB), relay node, femto, pico, etc., and is not limited in the embodiments of this application.

[0074] In the technical solution provided in this application, both the detection models under the edge-cloud and edge-side architectures need to detect the video content uploaded by users. During this process, information such as the video ID, video content, and user ID containing sensitive content may be stored and recorded, posing a privacy risk. This application addresses this risk by utilizing a cross-processing strategy within an edge consortium.

[0075] In the technical solution provided in this application embodiment, services within the edge alliance can share algorithms and computing power for sensitive information detection, effectively solving the concurrency problem during peak hours (such as the peak period of live streaming and short video services from 8 pm to 11 pm) under the end-to-cloud model (i.e., a network architecture that includes terminals and the cloud).

[0076] In the technical solution provided in this application embodiment, sensitive information detection is achieved based on hash encoding with two dimensions (encryption encoding and network encoding), which improves the accuracy of detection.

[0077] The various embodiments of this application will now be described in detail with reference to the accompanying drawings. It should be noted that the order in which the embodiments are presented represents only a chronological order and does not indicate the superiority or inferiority of the technical solutions provided by the embodiments.

[0078] This application uses the security detection of short video content as an example for illustration, but it is not limited to this; the content to be detected can also be information data in other formats. The security detection described herein can also be referred to as user privacy detection or sensitive information detection, etc.

[0079] See Figure 1 This application provides a privacy-preserving edge alliance short video content security detection scheme, which mainly includes the following three parts: video service communication mechanism under the edge alliance architecture, video content security detection task scheduling, and video content encoding and decoding. Examples are given below for each.

[0080] Part 1: Video Service Communication Mechanism under the Edge Consortium Architecture

[0081] In traditional edge-cloud architectures, there are no intermediate layer edge servers. Therefore, there is no intermediate computing power structure to share the computing tasks, and all computing pressure is on the cloud or the local terminal. Furthermore, the detection of sensitive data in video content uploaded by user terminals is carried out in the cloud, which may result in the cloud recording the identity of the producer of illegal video content, which is not conducive to the protection of user privacy.

[0082] In this embodiment of the application, the edge-cloud architecture is described below. Figure 2 This includes at least one cloud server, an edge server (also known as an edge computing server), and a user terminal (or user equipment). See also... Figure 3 Cloud servers can provide services such as short video streaming; see [link to relevant documentation]. Figure 4 In the edge server of the end-to-cloud architecture, the middle layer can be located on the base station side, and the same base station can contain multiple edge servers.

[0083] For example, see Figure 5 When user terminal i provides video services, the communication link evolves from the traditional end-to-cloud architecture to... Figure 5 The architecture shown in this application embodiment adds a large number of edge servers to the middle layer of the existing architecture. These edge servers are deployed in computer rooms or next to or inside communication base stations in communities, neighborhoods, streets, etc., and these edge servers are a friendly and communicative alliance. The uplink of data uploaded by the user terminal is transmitted to the regional base station via the edge servers deployed in the community, neighborhood, or street (e.g., edge server 3) or directly. However, in this application embodiment, edge servers are deployed in the base station, such as edge server 1, edge server 2, and edge server n.

[0084] Under the existing edge-cloud architecture and edge-side architecture, it is necessary to perform security checks on the video content uploaded by user terminals. In this process, information such as video ID, video content, and user terminal ID containing sensitive content may be stored and recorded, posing a risk of user privacy leakage.

[0085] Therefore, this application embodiment utilizes a cross-processing strategy within an edge alliance to address this risk. Specifically: video data from the user terminal is uploaded to an edge server, where sensitive information is detected. If the detection passes, the relevant video content can be uploaded from the base station where the edge server is located to a cloud server, for example, see [link to relevant documentation]. Figure 5 The path for uploading video content to the base station is: User Terminal 1 - Edge Server 1 - Base Station 1; or, the video content is uploaded from the base station covering the edge server to the cloud server, for example, see [link to relevant documentation]. Figure 5 The path for uploading video content to the base station is: user terminal n - edge server 3 - base station n.

[0086] All edge servers are deployed by one or more video service providers. These edge servers are friendly and reliable, forming an edge alliance to provide services to all user terminals. Therefore, the cross-processing strategy means that neither the user terminal nor any third party can predict which edge server will be responsible for processing sensitive information in the video data. Consequently, data uploaded by the user terminal is not stored on a designated edge server, thus mitigating the risk of third parties illegally accessing server data and stealing the user terminal's private data.

[0087] Different services within an edge alliance can share algorithms and computing power for sensitive information detection. Specifically, the cloud servers belong to the software vendors providing video services. In scenarios with multiple vendors, vendors can share edge servers 1 through n to deploy their own algorithms, or they can deploy independent edge servers n+1 (in multi-vendor scenarios). However, all vendors need to share multiple edge servers to process the computing requests from alliance members. Simultaneously, each individual edge server needs to distribute its total computing power among various services. In other words, edge servers joining an edge alliance must comply with the alliance's allocation and compute service requests from it. Service providers joining an edge alliance do not necessarily use their own edge servers when processing their own services; instead, they use edge servers deployed by other service providers within the alliance. Regardless of their ownership, these edge servers must deploy the services of that vendor. Therefore, this application embodiment utilizes an edge alliance composed of edge servers to achieve concurrent computing power sharing or offloading, effectively solving the service concurrency problem in edge-cloud architectures during peak hours (e.g., peak periods for live streaming and short video services from 8-11 PM).

[0088] If the uplink video content passes the sensitive information detection of an edge server, the data is uploaded to the cloud backend service via the corresponding base station for subsequent operations such as video forwarding and storage. If the video content fails the sensitive information detection, the uplink data communication process stops here, and further transmission is impossible.

[0089] Furthermore, regardless of whether there are multiple suppliers, since the definitions of sensitive information in video service scenarios are similar, different video services can all perform sensitive information detection within the same edge alliance architecture system. To prevent some video services or malicious user terminals from bypassing the edge alliance and directly uploading undetected data through backend interface addresses known to other services or illegally exposed backend interface addresses, this application embodiment mandates that user terminal data be uploaded to the edge server for security detection and encoded using a preset encoding method before being uploaded to the cloud server.

[0090] like Figure 6 As shown, Figure 6 User terminal E is an unauthorized user terminal. After illegally obtaining the cloud server interface, it bypasses the edge server and directly sends data to the cloud server. Since sensitive information detection in this embodiment is decentralized to the edge server (rather than the cloud), this bypasses sensitive information detection, allowing for the illegal uploading of illegal videos. Therefore, in this embodiment, the data uploaded by the user terminal is encoded using the encoding method provided in this embodiment on the edge server. Since the unauthorized user terminal uploads the data directly to the cloud server without encoding by the edge server using the provided encoding method, the cloud server cannot decode it and will not receive the video. In other words, this embodiment mandates that user terminal data be uploaded to the edge server for security testing. Only after passing the test and being encoded using the provided encoding method can the data be uploaded to the cloud server. The cloud server then uses a decoding method corresponding to the encoding method provided in this embodiment for decoding. In other words, after the user terminal's original data passes the security test of the edge server in the certified edge alliance, the edge server encodes the original data using the efficient encoding method provided in this embodiment and uploads it. The cloud server then successfully decodes the data to obtain the original video. Therefore, video data transmitted directly to the cloud by a user terminal without using an edge server cannot be decoded and thus cannot be read correctly. In other words, without an edge server, the user terminal data is not encoded, so the cloud cannot decode it and will not receive the video. The user terminal described in this embodiment can also be referred to as a client.

[0091] Part Two: Video Content Security Detection Task Scheduling

[0092] The video content security detection task scheduling provided in this application embodiment is used to solve the problem of selecting which edge server in the edge alliance can obtain the globally optimal service efficiency when a group of user terminals simultaneously request video services.

[0093] In this embodiment, "end-side" refers to devices such as mobile phones, computers, and tablets that have the function of uploading short videos and live streaming, or it can refer to a software module installed on the user terminal, denoted as i.

[0094] In this embodiment, the edge side (i.e., the edge server side) represents the edge server deployed in areas such as office buildings, communities, and base stations, denoted as j, and the computing power of the edge server is denoted as C. j .

[0095] Let R be the computing power required for the sensitive information detection algorithm to pass the video uploaded by the user terminal. Then:

[0096] R = ∑f*num;

[0097] Where f represents the computing power (Flops) required by the sensitive information detection algorithm to process 1 unit (e.g., 1 frame) of video content, and num represents the number of video units detected (e.g., 15 frames or 20 seconds).

[0098] The computational latency required for edge server j in the Edge Consortium to perform sensitive information detection on the video data sent by user terminal i is:

[0099] t 计算 =R i / (α j *C j );

[0100] Where, α j The percentage coefficient of computing power allocated by edge server j to video content security detection tasks from user terminal i; for example, if edge server j has a computing power of 100, and if edge server j receives video data from three user terminals, then it needs to complete three video content security detection tasks. Therefore, it can choose to allocate 30%, 30%, and 40% of the computing power respectively. That is, for these three user terminals, the corresponding α is 30%, 30%, and 40%, respectively. In other words, the computing power allocated by edge server j to these three video content security detection tasks is 30%, 30%, and 40, respectively.

[0101] R i This represents the computing power required for user terminal i to detect the video content provided by the terminal using a sensitive information detection algorithm.

[0102] C j This represents the computing power possessed by edge server j.

[0103] Let B be the communication rate at which video is transmitted from user terminal i to edge server j. j The size of the video uploaded by user terminal i is S. i bits, then the completion latency of the video content security detection task is:

[0104] t j =t 计算 +t 传输 =R i / (α j *C j )+S i / B j

[0105] Among them, B j =B i log(1+S / N i ), B i This represents the link bandwidth for communication between user terminal i and edge server j. S / N i This represents the signal-to-noise ratio (SNR) during communication between user terminal i and edge server j. In the embodiments of this application, the SNR is primarily related to the distance d from user terminal i to the base station. i Relatedly, N represents the noise signal received by the base station under broadcast signals, including interference from other user terminals and Gaussian white noise.

[0106] Under the edge consortium architecture, the computation time for video content security inspection should be more efficient than performing video content security inspection locally on the edge under the traditional architecture. Let C be the computing power of the edge device for video content security inspection. i The completion time of the video content security detection task on the user terminal is then:

[0107] t i =R i / C i ;

[0108] Let the user latency benefit of edge server j be β. j (Fixed value), then the user service benefit produced by edge server j in this embodiment of the application is: -β j *(t j -t i ).

[0109] Wherein, the user latency benefit of the edge server j is β jThis refers to the lost benefits due to latency during user service usage. For example, in live streaming services, user satisfaction is a simple example of user benefit. Higher latency results in lower satisfaction and higher latency benefits. Therefore, after accelerating video content security detection using the embodiments of this application, the user service benefit produced by edge server j can be recorded as: negative loss benefit (latency benefit) * time difference saved after acceleration. Wherein, the negative loss benefit is -β... j The time difference saved after acceleration, i.e., t j -t i .

[0110] Therefore, considering the constrained optimization problem, we can obtain the set of α that maximizes the global benefit in the current time slot. This set includes the set of computing power percentage coefficients α of multiple video content security detection tasks provided by multiple edge servers in the edge alliance to multiple user terminals.

[0111] Regarding the maximization of global profit, since it is a negative number (-β) j Therefore, the goal is to minimize the time delay benefit (i.e., minimize the loss). Thus, we have:

[0112] min∑β j *(t j -t i )

[0113] st∑α j =1 (1)

[0114] t j <t i (2)

[0115] Where, ∑β j *(t j -t i ), iterate through j and i respectively and sum; st represents the constraint condition, constraint (1) represents the percentage coefficient of computing power allocated by any edge server to the video content security detection task provided by multiple user terminals, and the sum is 1. For example, if an edge server receives video data provided by 5 user terminals, then it can choose to distribute the computing power equally, with each user terminal receiving 20% ​​of the computing power. That is, for each of these 5 user terminals, the corresponding α is 20%; constraint (2) represents the time t required for edge server j to perform security detection on the video content provided by user terminal i. j It must be less than the time t required for the user terminal i to perform local security checks on the video content. i .

[0116] Therefore, in summary, see Figure 7On any edge server side, the video content processing method provided in this application embodiment includes, for example:

[0117] Step S701: Edge server j receives the computational load information related to the security detection of the video content to be uploaded by user terminal i, broadcast by user terminal i.

[0118] The computational load information includes:

[0119] Security inspection task size R i (That is, the computing power required for the video content to be uploaded by user terminal i to pass the sensitive information detection algorithm, which is also the computing power required to execute the security detection task corresponding to user terminal i).

[0120] The size S of the video content to be uploaded by user terminal i i ;

[0121] The computational latency t required for the user terminal to locally complete the security detection task for the video content i .

[0122] Step S702: After receiving a video upload request (also equivalent to a sensitive information detection request, i.e., a security detection request) from at least one user terminal (e.g., k user terminals, where k is an integer greater than or equal to 1) in the current time slot, edge server j sets the initial iteration computing power percentage coefficient α for each user terminal according to the weighting rules. j For example, if k = 3, then for these three user terminals, the corresponding α... j They are 30%, 30%, and 40% respectively.

[0123] The weighting rule can be determined according to actual needs. For example, for edge server j, the closer a user terminal is to edge server j, or the greater the computational load of video content security detection, the higher the percentage coefficient α of the computational power allocated to that user terminal for video content security detection. j The larger it is, the smaller it is.

[0124] Step S703: Edge server j utilizes the computing power percentage coefficient α j And R as described in step S701 i S i t i Calculate the sum of user service benefits generated by edge server j for each user terminal (i.e., the k user terminals that send video upload requests to edge server j), i.e., ∑β j *(t j -t i ).

[0125] See Figure 8This step S703 specifically includes, for example:

[0126] S801, the percentage coefficient α of computing power achievable by this edge server j for video content security detection. j The computing power C possessed by edge server j j And R, the size of the video content security detection task broadcast by user terminal i in step S701. i Calculate the computation latency t required for edge server j to perform video content security detection. 计算 ;

[0127] S802, The video size S to be uploaded by user terminal i using the broadcast in step S701. i The communication rate B from user terminal i to edge server j j Calculate the latency t of video transmission from user terminal i to edge server j. 传输 ;

[0128] S803, using t 计算 and t 传输 The completion latency t of the video content security inspection task on edge server j is calculated. j ;

[0129] S804, Utilizing the user latency benefits of edge server j (β) j , and t j And the computational delay t required for the user terminal i to locally complete the video content security detection task, broadcast in step S701. i Determine the user service benefits generated by edge server j for user terminal i: β j |t j -t i |;

[0130] S805, Obtain the percentage coefficient α of the currently allocated computing power. j In the case of edge server j, the sum of user service benefits generated by user terminals making video upload requests for all current time slots is ∑β. j *(t j -t i ).

[0131] Step S704: Use a penalty function to solve the constraint optimization problem. Under the distance constraint rule, iterate the combination of multiple user terminals and multiple edge servers, and determine the user service benefit corresponding to each combination. Use the user service benefit corresponding to each combination to obtain the α set that maximizes the global revenue in the current time slot.

[0132] Among these, the distance restriction rules, for example, prohibit user terminal i and edge server j from being combined if they exceed a preset distance range (e.g., 2km).

[0133] Step S705: Determine the allocation strategy for video content security detection tasks based on the optimal solution set (i.e., set α).<i,j> And the percentage coefficient α of computing power allocated by edge server j to user terminal i for video content security detection tasks in this allocation strategy. j .

[0134] Wherein, the computational allocation strategy<i,j> This indicates that the edge server j is determined to perform sensitive information detection on the data of user terminal i, thus establishing the correspondence between user terminal i and edge server j.

[0135] Steps S704 and S705 involve determining a globally optimal solution. This solution determines which edge server j should perform sensitive information detection on the user terminal i's data, and how much computing power edge server j needs to expend to achieve global optimization.

[0136] For example, edge server j1 will iterate over multiple combinations of edge servers and user terminals i. For example, in the first iteration, i = 2, 3, 5, 7, α = [20%, 20%, 20%, 40%], and the benefit M0 at this time can be obtained.

[0137] The edge server uses a penalty function to iterate continuously to find the maximum user service benefit M = M1, and obtains the α combination at this time [30%, 10%, 20%, 40%].

[0138] Perform a second iteration, where i = 1, 3, 5, 6, and repeat the above steps to obtain the maximum user service benefit M2, and obtain the α combination at this time [30%, 10%, 20%, 40%].

[0139] If M2 > M1, then denote M = M2;

[0140] The above iteration is performed on each edge server j1 to j to obtain the global (entire edge consortium) maximum M. Here, the α corresponding to the maximum M of edge server j1 can be denoted as α1, and the α corresponding to the maximum M of edge server j can be denoted as α2. j .

[0141] Each edge server will notify other edge servers of its own calculation results (the results of the above iterative processing), so that each edge server can know the global maximum M, and then each edge server can determine the optimal allocation strategy for video content security detection tasks.

[0142] Step S706: According to the allocation strategy<i,j> Edge server j sends a video data upload response to the corresponding user terminal i, and user terminal i uploads video data to the corresponding edge server j.

[0143] Step S707: After receiving the video data and completing the sensitive information detection, the edge server j encodes the video that has passed the sensitive information detection and uploads it to the cloud platform server.

[0144] Part Three: Video Content Encoding and Decoding

[0145] In this embodiment, sensitive information detection is achieved based on two dimensions of hash encoding, improving the accuracy of detection. The first dimension is encryption encoding, which provides verification and security through the designed encoding method. The second dimension is network encoding, which ensures the traceability and queryability of the results by incorporating a node allocation scheme. Specifically:

[0146] In some embodiments, the video sent by the user terminal is allocated according to a determined strategy.<i,j> The data is uploaded to the selected edge server, and after the selected edge server performs sensitive information detection, it is uploaded to the cloud platform using the following encoding method.

[0147] The encoded content includes: video content, detection results, detection time slots, detection nodes, calculation allocation scheme results, anti-counterfeiting verification, decoding keys, and other information, ensuring the security and reliability of uploaded videos and the traceability of sensitive data detection records. If the cloud discovers falsely detected video content (i.e., video content with incorrect sensitive information detection) during offline sampling, it can verify the edge alliance detection status and the corresponding node's detection status in the time slot where the video was uploaded based on the encoding. This provides traceability support for further improving the calculation accuracy of edge servers or penalizing malicious missed detections by alliance members.

[0148] In this embodiment, the video service requires high computational efficiency, and is encoded using encryption. The encryption key is periodically updated in the cloud and within the alliance according to alliance rules. Examples of encryption methods include:

[0149] (1) Using time (i.e. the detection time slot mentioned above) and the number of members of the edge alliance as input, generate a binary key of a specified number of bits (e.g., 256 bits).

[0150] (2) The video data binary encoding is XORed with the key bitwise to form encrypted video stream data.

[0151] (3) Combine the decryption key, encrypted video stream data, preset identifier check character, and three sets of encoding blocks according to the preset format (i.e., the order of their positions in the encoded data) to obtain the encoded data.

[0152] The encoded data format is as follows: Figure 9 As shown, both the identifier checksum F and N are used for anti-counterfeiting verification. The identifier checksum F also serves to separate the encrypted video stream and the algorithm allocation scheme block. This is because the length of the encrypted video stream varies with the length of the video uploaded by the user. In order to quickly extract the position of the module, the module content can be quickly obtained by directly searching the neighboring position of the checksum. The identifier checksum N also serves to separate the algorithm allocation scheme block and the node information encoding block.

[0153] The three sets of coded blocks are explained below.

[0154] Block 1 (Detection Result Encoding Block):

[0155] The detection result consists of three parts: text (title and introduction), audio, and video. They are respectively tested using natural language processing, speech, and image sensitive information detection models. If all three pass the detection, the encoding value of the detection result encoding block is 111; otherwise, they are set to 0 in sequence.

[0156] Block 2 (Algorithm Allocation Scheme Encoding Block):

[0157] The algorithm's allocation scheme block is represented by a matrix. Using the three parameters j, i, and α in the allocation strategy determined above, a three-dimensional matrix A is formed. j,i，α This is used to indicate the percentage of computing power allocated to edge server j when processing video content on user terminal i.

[0158] The three-dimensional matrix is, for example, [[j1,[i1,0.02],[i11,0.03]…],[j2,[i5,0.02],[i12,0.03]], etc., where, for example, [j1,[i1,0.02]] indicates that edge server j1 spends 0.02 percent of computing power to process data from user terminal i1.

[0159] To represent this matrix in the encoding, it is represented using matrix encoding. For A... j,i，α Dimensionality reduction and transformation are performed to obtain A. j and A i,a Two vectors, A j As a number of bits, the position is filled with the number of requests from user terminal i processed by edge server j, which falls into the encoding bit, such as Figure 10 As shown, the first two digits of the algorithm allocation block, i.e., the values ​​corresponding to edge servers j1 and j2, are both 2; since A i,a The row is unique (i.e., the video from the same user terminal will not be assigned to two computing nodes in one time slot), so A i As the number of digits, fill in 100*α in the position, such as... Figure 10As shown, the last four digits of the values ​​corresponding to edge servers j1 and j2 are 100*α for user terminals i1, i2, i3, and i4, respectively, which are 2, 5, 3, and 7; A i The codes are sequentially placed at the tail of the two-dimensional matrix, such as... Figure 10 As shown, the four digits following 100*α for each of the user terminals i1, i2, i3, and i4 are the respective numbers of user terminals i1, i11, i15, and i12, namely 1, 11, 15, and 12.

[0160] Specifically, the original matrix is ​​as follows: [[j1,[i1,0.02],[i4,0.07]],[j2,[i2,0.05],[i3,0.03]];

[0161] Edge server j1 handles two user terminals i1 and i4, while edge server j2 handles two user terminals i2 and i3.

[0162] Block 3 (Node Information Encoding Block):

[0163] The node information, such as the number of the edge server providing the encoding service in the edge consortium.

[0164] The edge server number in the edge alliance of this application embodiment, for example... Figure 11 As shown, the information includes three parts: the region code where the edge server is located, the average available computing power of the edge server, and the network node structure number of the edge server in the edge alliance. Therefore, the node information encoding block is obtained by encoding these three parts.

[0165] The network node structure number of the edge server in the edge consortium is determined, for example, in the following way:

[0166] For example Figure 12 As shown, during the deployment phase, any edge server first starts by traversing the edge servers in the edge alliance and assigning them a unique number to ensure that each edge server has a different number.

[0167] Starting from the same edge server, perform the same traversal operation again, and when outputting, place the edge server number as the first digit in the next edge server input. For example, see [link to example]. Figure 13 Edge server 7 is input from edge server 4, denoted as 4'7; if the edge server has multiple inputs, such as the original edge server 1, it is denoted as 6'1,2'1; this naming method can quickly identify the connected edge server by searching the number of bits before the ' identifier, which is convenient for designing a fast communication mechanism.

[0168] Before the service (i.e., video data) is sent to the cloud via the base station, it is encoded. If the header encoding (i.e., the detection result encoding) is not 111, a message is returned to the user terminal i that it has failed the sensitive information detection and the relevant data is discarded.

[0169] If the header encoding is 111, the video data will be sent to the cloud after encoding.

[0170] Following the above encoding method, the cloud verifies the header encoding as 111 again, retrieves two identifier checksums (to prevent impersonation of edge server encoding rules), and confirms that the last digit of the edge server number corresponds to the edge server number within the edge alliance to which the service belongs. Then, it obtains the fourth bit (i.e., the position after the header encoding 111) of the key (e.g., bits 4 to 260), decodes the video stream, and performs subsequent services. The algorithm allocation scheme block serves as evidence, proving which edge server performed the sensitive information detection.

[0171] In summary, based on the above content, the data content upload processing flow provided in this application embodiment is as follows: Figure 14 As shown, the specifics will not be elaborated further.

[0172] The methods provided in the embodiments of this application are summarized below from different entity perspectives.

[0173] See Figure 15 On the edge server side, this application provides a method for processing data uploaded by a user terminal, including:

[0174] S101. When a security detection request is received from a user terminal for data content (e.g., video content) to be uploaded, a computing power percentage coefficient is allocated to the user terminal, wherein the computing power percentage coefficient is the percentage of the total computing power of the edge server that is allocated by the edge server to perform security detection on the data content to be uploaded by the user terminal.

[0175] The percentage coefficient for allocating computing power to the user terminal, for example, is the percentage coefficient α of computing power in the initial iteration mentioned above. j ;

[0176] S102, Utilize the computational information related to security detection of the user terminal (e.g., the aforementioned R) i S i , t i The current security detection task allocation strategy is determined by the computing power percentage coefficient and the computing power percentage coefficient. The security detection task allocation strategy is used to instruct the edge server to perform security detection tasks on the data content to be uploaded by the user terminal.

[0177] The computational load information can be pre-set on each edge server side, or it can be obtained in real time from the terminal reports by the edge server side.

[0178] This application embodiment implements a security detection task allocation strategy on the edge server side. When a user terminal receives a security detection request for data content to be uploaded, the system allocates a computing power percentage coefficient to the user terminal. Using the user terminal's security detection-related computational load information and the computing power percentage coefficient, a current security detection task allocation strategy is determined. This strategy prevents the user terminal or a third party from predicting which edge server will be responsible for sensitive information detection, thereby improving privacy protection for sensitive information such as data content uploaded by the user terminal. Furthermore, by utilizing the user terminal's security detection-related computational load information and the computing power percentage coefficient to determine the current security detection task allocation strategy, the system can allocate the globally (i.e., the edge alliance composed of multiple edge servers) optimal edge server to perform the security detection task for the user terminal. This also improves the efficiency of sensitive information detection and avoids the problems of insufficient computing power and low detection efficiency caused by data concurrency during peak periods.

[0179] In some embodiments, the current security detection task allocation strategy is determined using the computational load information related to security detection of the user terminal and the computing power percentage coefficient, including:

[0180] Using the computational load information related to security detection of the user terminal and the computing power percentage coefficient, the sum of user service benefits generated by the local edge server for the user terminal that sends the security detection request to the local edge server is calculated, such as the aforementioned ∑β. j *(t j -t i );

[0181] Using a penalty function, iterate through combinations of multiple user terminals and multiple edge servers, and determine the sum of user service benefits corresponding to each combination. Then, use the sum of user service benefits corresponding to each combination to determine the set of computing power percentage coefficients that maximizes the current global revenue, such as the α set mentioned above.

[0182] Using the set of computing power percentage coefficients, the current security detection task allocation strategy is determined. The security detection task allocation strategy includes: the correspondence between edge servers and user terminals (i.e., which edge server performs sensitive information detection on the user terminal's data), and the computing power percentage coefficient allocated by the edge server to the corresponding user terminal.

[0183] In some embodiments, the method further includes:

[0184] According to the security detection task allocation strategy, determine the user terminal corresponding to the local edge server;

[0185] Send a security detection response to the user terminal corresponding to the local edge server, and obtain the data content uploaded by the user terminal corresponding to the local edge server;

[0186] The data uploaded by the user terminal corresponding to the local edge server is encoded using a preset method and then uploaded to the cloud server.

[0187] In some embodiments, the data content uploaded by the user terminal corresponding to the local edge server is encoded using a preset method, including:

[0188] Using the current local time and the number of members in the edge alliance, a binary key of a preset length is generated;

[0189] The binary encoding of the data content uploaded by the user terminal corresponding to the local edge server is XORed with the binary key bitwise to form the encrypted data content;

[0190] The decryption key, the encrypted data content, the preset identifier check character, and the first information encoding block, the second information encoding block, and the third information encoding block (such as the three sets of encoding blocks mentioned above) are used to generate encoded data according to a preset format.

[0191] In some embodiments, the first information encoding block (e.g., the detection result encoding block described above) is generated using the following information:

[0192] The edge servers in the security detection task allocation strategy use preset natural language processing, voice, and image sensitive information detection models to detect the text, audio, and video in the data content provided by the corresponding user terminals in the security detection task allocation strategy, and obtain the detection results.

[0193] The second information encoding block (such as the algorithm allocation scheme encoding block mentioned above) is generated using the following information:

[0194] The number of security detection requests processed by each user terminal in the security detection task allocation strategy;

[0195] In the security detection task allocation strategy, each user terminal corresponds to 100*α, where α represents the computing power percentage coefficient.

[0196] The user terminal's unique identifier in the security detection task allocation strategy;

[0197] The third information encoding block (such as the node information encoding block mentioned above) is generated using the following information:

[0198] The edge server's region encoding in the security detection task allocation strategy;

[0199] The average available computing power of edge servers in the security detection task allocation strategy;

[0200] The edge servers in the security detection task allocation strategy are numbered in the network node structure of the edge alliance.

[0201] Accordingly, see Figure 16 On the cloud server side, this application provides a method for processing data uploaded by a user terminal, including:

[0202] S201, Receive the encoded block sent by the edge server (e.g., Figure 9 The encoded block (shown in the format) is obtained by the edge server encoding the data content from the user terminal; wherein the data content is the data content that has passed the security test by the edge server; and the security test is performed according to a determined security test task allocation strategy, which is determined by the method on the edge server side.

[0203] S202, Decode the encoded block.

[0204] The cloud server uses the decoding method corresponding to the encoding method used by the edge server to perform decoding, for example:

[0205] In some embodiments, the decoding process includes:

[0206] Determine the header encoding of the encoded block (e.g.) Figure 9 The code block of the detection result shown is a preset encoding value (e.g., "111") used to indicate that the data content has passed the security test by the edge server; if not, the code block can be discarded and the decoding process can be ended.

[0207] If so, then search the encoded block for the existence of a first preset checksum (e.g., at a preset position). Figure 9 The identifier check character F shown) and the second preset identifier check character (e.g. Figure 9If the first preset checksum and the second preset checksum do not exist, the encoded block can be discarded and the decoding process can end; if they exist, the encoded value between the first preset checksum and the second preset checksum is obtained; and the encoded value between the first preset checksum and the second preset checksum is determined (e.g., ...). Figure 9 Whether the last digit of the algorithm allocation scheme block shown is the number of the edge server in the preset edge alliance;

[0208] When it is confirmed that the last bit of the encoded value between the first preset identifier checksum and the second preset identifier checksum is the number of the edge server in the preset edge alliance, the encoded value of a preset number of bits is read starting from the header encoding (e.g., ...). Figure 9 The decryption key shown is used as the key to decrypt the data content from the user terminal (e.g., Figure 9 The encrypted video stream shown is decrypted to obtain the decrypted data content, wherein the encrypted data content from the user terminal is located between the key and the first preset identifier check character in the encoding block.

[0209] Accordingly, see Figure 17 On the terminal side, an embodiment of this application provides a method for a user terminal to upload data, including:

[0210] S301. When it is necessary to upload data content to the network side, a security detection request for the data content to be uploaded is sent to the edge server. This is used to instruct the edge server to determine the security detection task allocation strategy according to the security detection request and using the method described above on the edge server side. This includes information about the edge server that performs the security detection task for the local user terminal (e.g., the correspondence between the edge server number and the local user terminal number), and the percentage coefficient of computing power allocated by the edge server that performs the security detection task for the local user terminal to the local user terminal.

[0211] S302, Receive the response sent by the edge server that performs the security detection task for the local user terminal, and send the data content to the edge server that performs the security detection task for the local user terminal.

[0212] In some embodiments, before sending the security detection request, the method further includes:

[0213] The computational load information related to security detection of the local user terminal is sent to the edge server, wherein the computational load information includes: the computing power required to perform the security detection task (e.g., the aforementioned R). i ), and the size of the data content (e.g., the S mentioned above). i), and the computational latency required for the local user terminal to perform the security detection task (e.g., the aforementioned t). i ).

[0214] The following describes the device or apparatus provided in the embodiments of this application, and the explanations or examples of the same or corresponding technical features as those described in the above methods will not be repeated hereafter.

[0215] This application provides a computing device, such as a user terminal; the computing device may also be a network-side device, such as an edge server, a cloud server, etc.

[0216] The computing device may specifically be a desktop computer, a portable computer, a smartphone, a tablet computer, a personal digital assistant (PDA), etc.

[0217] See Figure 18 The computing device may include a memory 11 and a processor 12;

[0218] Processor 12 may be a central processing unit (CPU).

[0219] The computing device may also include input / output devices, such as keyboards, mice, and touch screens, and output devices may include display devices, such as liquid crystal displays (LCDs) and cathode ray tubes (CRTs).

[0220] The memory 11 may include read-only memory (ROM) and random access memory (RAM), and provides the processor 12 with program instructions and data stored in the memory 11. In this embodiment, the memory 11 may be used to store the program of any of the methods provided in this embodiment.

[0221] The processor 12 executes any of the methods described in the embodiments of this application according to the program instructions stored in the memory 11.

[0222] In some embodiments, a network-side device (which may be an edge server or a cloud server) provided in this application is described in [reference needed]. Figure 19 ,include:

[0223] The processor 500 is used to read the program in the memory 520 and execute any of the methods described in the embodiments of this application on the edge server side or the cloud server side.

[0224] Transceiver 510 is used to receive and send data under the control of processor 500.

[0225] Among them, Figure 19 In this context, the bus architecture may include any number of interconnected buses and bridges, specifically linking various circuits together, represented by one or more processors (processor 500) and memory (memory 520). The bus architecture may also link various other circuits such as peripheral devices, voltage regulators, and power management circuits, which are well known in the art and therefore will not be described further herein. The bus interface provides an interface. The transceiver 510 may be multiple elements, including transmitters and receivers, providing a unit for communicating with various other devices over a transmission medium. The processor 500 is responsible for managing the bus architecture and general processing, and the memory 520 may store data used by the processor 500 during operation.

[0226] The processor 500 can be a central processing unit (CPU), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or a complex programmable logic device (CPLD).

[0227] In some embodiments, a user terminal is provided in this application, see [link to relevant documentation]. Figure 20 ,include:

[0228] The processor 600 is used to read the program in the memory 620 and execute the method provided on the user terminal side in the embodiments of this application described above.

[0229] In some embodiments, the user terminal further includes a transceiver 610 for receiving and sending data under the control of the processor 600.

[0230] Among them, Figure 20 In this context, the bus architecture can include any number of interconnected buses and bridges, specifically linking various circuits together, represented by one or more processors (processor 600) and memory (memory 620). The bus architecture can also link together various other circuits such as peripheral devices, voltage regulators, and power management circuits, which are well known in the art and therefore will not be described further herein. The bus interface provides an interface. The transceiver 610 can be multiple elements, including a transmitter and a receiver, providing a unit for communicating with various other devices over a transmission medium.

[0231] In some embodiments, a user interface 630 is also included. The user interface 630 may be an interface that can connect to external or internal devices, including but not limited to keypads, displays, speakers, microphones, joysticks, etc.

[0232] The processor 600 is responsible for managing the bus architecture and general processing, while the memory 620 can store the data used by the processor 600 when performing operations.

[0233] In some embodiments, the processor 600 may be a CPU (Central Processing Unit), an ASIC (Application Specific Integrated Circuit), an FPGA (Field-Programmable Gate Array), or a CPLD (Complex Programmable Logic Device).

[0234] This application also provides a computer program product or computer program that includes computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform any of the methods described in the above embodiments. The program product may employ any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of readable storage media (a non-exhaustive list) include: an electrical connection having one or more wires, a portable disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof.

[0235] This application provides a computer-readable storage medium for storing computer program instructions used in the apparatus provided in the above-described embodiments, including a program for performing any of the methods provided in the above-described embodiments. The computer-readable storage medium may be a non-transitory computer-readable medium.

[0236] The computer-readable storage medium can be any available medium or data storage device that a computer can access, including but not limited to magnetic storage (e.g., floppy disks, hard disks, magnetic tapes, magneto-optical disks (MOs), etc.), optical storage (e.g., CDs, DVDs, BDs, HVDs, etc.), and semiconductor storage (e.g., ROMs, EPROMs, EEPROMs, non-volatile memory (NAND flash), solid-state drives (SSDs)).

[0237] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage and optical storage) containing computer-usable program code.

[0238] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0239] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0240] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0241] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A method for processing data uploaded by a user terminal, characterized in that, Applied to edge servers, the method includes: When a security detection request is received from a user terminal for the data content to be uploaded, a computing power percentage coefficient is allocated to the user terminal. The computing power percentage coefficient is the percentage of the total computing power of the edge server that is allocated to the user terminal for security detection of the data content to be uploaded. Using the computational load information related to security detection of the user terminal and the computing power percentage coefficient, the current security detection task allocation strategy is determined. The security detection task allocation strategy is used to instruct the edge server to perform security detection tasks on the data content to be uploaded by the user terminal. According to the security detection task allocation strategy, determine the user terminal corresponding to the local edge server; Send a security detection response to the user terminal corresponding to the local edge server, and obtain the data content uploaded by the user terminal corresponding to the local edge server; Based on the data content uploaded by the user terminal corresponding to the local edge server, it is encoded using a preset method and then uploaded to the cloud server; wherein, the preset method includes an encryption encoding method and a network encoding method, the encryption encoding method has a verification function, and the network encoding method ensures that the results are traceable and queryable by incorporating a node allocation scheme.

2. The method according to claim 1, characterized in that, Using the computational load information related to security detection of the user terminal and the computing power percentage coefficient, the current security detection task allocation strategy is determined, including: Using the computational load information related to the security detection of the user terminal and the computing power percentage coefficient, calculate the user service benefits generated by the local edge server for the user terminal that sends the security detection request to the local edge server; Using a penalty function, iterate through combinations of multiple user terminals and multiple edge servers, and determine the sum of user service benefits for each combination. Then, use the sum of user service benefits for each combination to determine the set of computing power percentage coefficients that maximize the current global revenue. Using the set of computing power percentage coefficients, the current security detection task allocation strategy is determined. The security detection task allocation strategy includes: the correspondence between edge servers and user terminals, and the computing power percentage coefficient allocated by the edge server to the corresponding user terminal.

3. The method according to claim 1, characterized in that, The data content uploaded by the user terminal corresponding to the local edge server is encoded using a preset method, including: Using the current local time and the number of members in the edge alliance, a binary key of a preset length is generated; The binary encoding of the data content uploaded by the user terminal corresponding to the local edge server is XORed with the binary key bitwise to form the encrypted data content; The decryption key, the encrypted data content, the preset identifier check character, and the first information encoding block, the second information encoding block, and the third information encoding block are used to generate encoded data according to a preset format.

4. The method according to claim 3, characterized in that, The first information encoding block is generated using the following information: The edge servers in the security detection task allocation strategy use preset natural language processing, voice, and image sensitive information detection models to detect the text, audio, and video in the data content provided by the corresponding user terminals in the security detection task allocation strategy, and obtain the detection results. The second information encoding block is generated using the following information: The number of security detection requests processed by each user terminal in the security detection task allocation strategy; The security detection task allocation strategy corresponds to 100 user terminals. α, where α represents the computing power percentage coefficient; The user terminal's number in the security detection task allocation strategy; The third information encoding block is generated using the following information: The edge server's region encoding in the security detection task allocation strategy; The average available computing power of edge servers in the security detection task allocation strategy; The edge servers in the security detection task allocation strategy are numbered in the network node structure of the edge alliance.

5. A method for processing data uploaded by a user terminal, characterized in that, Applied to a cloud server, the method includes: The system receives an encoded block sent by an edge server, the encoded block being obtained by the edge server encoding data content from a user terminal; wherein the data content is data content that has passed security detection by the edge server; and the security detection is performed according to a determined security detection task allocation strategy, the security detection task allocation strategy being determined using the method described in claim 1. The encoded block is then decoded.

6. The method according to claim 5, characterized in that, The decoding process includes: Determine whether the header encoding of the encoded block is a preset encoding value used to indicate that the data content has passed the security check by the edge server; If so, then search the encoding block for the existence of a first preset identifier check character and a second preset identifier check character according to the preset position; if so, then obtain the encoded value between the first preset identifier check character and the second preset identifier check character. When it is confirmed that the last bit of the encoded value between the first preset identifier checksum and the second preset identifier checksum is the number of the edge server in the preset edge alliance, a preset number of encoded values ​​are read from the header encoding as a key to decrypt the encrypted data content from the user terminal, thereby obtaining the decrypted data content. The encrypted data content from the user terminal is located between the key and the first preset identifier checksum in the encoding block.

7. A method for uploading data using a user terminal, characterized in that, Applied to a user terminal, the method includes: When data content needs to be uploaded to the network side, a security detection request for the data content to be uploaded is sent to the edge server, which is used to instruct the edge server to determine the security detection task allocation strategy according to the security detection request using the method of claim 1, including information on the edge server that performs the security detection task for the local user terminal, and the percentage coefficient of computing power allocated by the edge server that performs the security detection task for the local user terminal to the local user terminal. The system receives a response from the edge server that performs the security detection task for the local user terminal, and sends the data content to the edge server that performs the security detection task for the local user terminal.

8. The method according to claim 7, characterized in that, Before sending the security detection request, the method further includes: The computational load information related to security detection of the local user terminal is sent to the edge server. The computational load information includes: the computing power required to perform the security detection task, the size of the data content, and the computational latency required for the local user terminal to perform the security detection task.

9. A computing device, characterized in that, include: Memory, used to store program instructions; A processor is configured to invoke program instructions stored in the memory and execute the method according to any one of claims 1 to 8.

Images