Channel isolation wireless resource scheduling method and system based on service trusted access

Abstract

The invention discloses a channel isolation wireless resource scheduling method and system based on service trusted access, and the method comprises the steps: constructing a wireless network according to a specific industry application scene, and carrying out the slicing hard isolation of a wireless network channel and the division of logic resources in a hard isolation channel in combination with the industrial service security isolation requirement; constructing an end-to-end quantum encryption channel; the method comprises the following steps: based on a quantum key wireless trusted access wireless network, constructing a graph neural network by combining network topology and channel service monitoring and channel resource real-time occupation monitoring, and establishing a network channel time delay prediction model; and presetting various service priorities, carrying out prediction planning on channel resources, and carrying out wireless resource scheduling based on a network channel time delay prediction result. According to the invention, network resource isolation and service real-time performance at a wireless resource level are realized, a broadband wireless network which is safe and can efficiently utilize network resources is constructed, service requirements of different security levels can be met at the same time, and the cost is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of communication system channel resource allocation and isolation, and in particular to a channel isolation wireless resource scheduling method and system based on service trusted access. Background Art

[0002] The allocation and isolation of communication system channel resources is an important concept in the field of wireless communication. Currently, wireless communication uses single-channel transmission, which is simple to set up and communicate. All data is transmitted serially in a single channel. The security is worse than that of a multi-channel transmission system. The reasons are mainly two-fold. On the one hand, the data transmission path is fixed, and the problem of transmission channel exposure cannot be solved, so the security of the channel cannot be guaranteed. On the other hand, a single-channel transmission system can only use a single channel to transmit data serially. If the channel is monitored by a malicious third party, it can monitor all the contents of the transmitted data, and the security of the data cannot be guaranteed. It is difficult to guarantee a single-channel wireless communication system that carries different isolation levels and different communication index requirements. A multi-channel transmission system can use a combination of wired and wireless channels to transmit data. Multiple channels can easily hide the channel that actually transmits data, solve the channel exposure problem of a single channel, and ensure the security of the transmission channel. A multi-channel transmission system uses a fragmentation mechanism to decompose data into fragments, and then transmits them in parallel on multiple channels. A malicious third party cannot monitor all channels, and naturally cannot obtain the complete ciphertext. In addition, a hopping mechanism can be used to randomly hash and transmit ciphertext fragments in the same transmission channel, making it more difficult for the monitor to steal data and further improving the security of the data.

[0003] In the context of the development and application of quantum encryption technology, the quantum immutability feature is used to meet the isolation requirements of different security levels of channels. In combination with the network communication computing power perception network at the end of the communication network, the scheduling of different business resource requirements in the channel is studied. The existing channel resource allocation and isolation are very challenging. Mathematically, the channel resource allocation problem can be modeled as the problem of optimizing a utility function under limited resources. The optimization of these functions is usually non-convex and NP-hard problems, and it is very difficult to find the optimal solution. The existing classical methods all find suboptimal solutions through multiple iterations and achieve good performance. However, multiple iterative calculations will lead to high computational complexity, which will bring two serious consequences: the device battery will be exhausted more quickly, and the channel resources need to be frequently recalculated in the time-varying channel, making the calculation results unable to adapt to the changed channel characteristics. Channel isolation currently mainly adopts secure channel isolation measures. Secure channel isolation adopts security mechanisms, such as dedicated hardware and unique security protocols, for internal and external network isolation and data transmission, effectively isolating internal and external network communications, achieving secure data interaction between internal and external networks, and supporting a variety of network applications. It is the main direction of current isolation technology research, but there is no effective means for channel isolation and wireless resource isolation within the same wireless communication system. Summary of the invention

[0004] The purpose of this section is to summarize some aspects of embodiments of the present invention and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section and the specification abstract and the invention title of this application to avoid blurring the purpose of this section, the specification abstract and the invention title, and such simplifications or omissions cannot be used to limit the scope of the present invention.

[0005] In view of the above existing problems, the present invention is proposed.

[0006] Therefore, the present invention provides a channel isolation wireless resource scheduling method and system based on service trusted access to solve the problems in existing communication systems that single-channel transmission is susceptible to monitoring, it is difficult to meet the requirements of different isolation levels, the resource allocation problem is complex and the calculation complexity is high, and effective channel isolation cannot be achieved within the same system.

[0007] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0008] In a first aspect, an embodiment of the present invention provides a channel isolation wireless resource scheduling method based on service trusted access, including: building a wireless network according to a specific industry application scenario, slicing and hard isolating the wireless network channel and dividing the logical resources in the hard isolation channel in combination with the industry service security isolation requirements;

[0009] Set up a quantum key distribution and authentication platform at the front end of the industry application platform, apply quantum communication terminals on the wireless network service side, and build an end-to-end quantum encryption channel;

[0010] Based on quantum key wireless trusted access to wireless networks, combined with network topology and channel service monitoring and real-time channel resource occupancy monitoring, a graph neural network is constructed to establish a network channel delay prediction model.

[0011] Preset the priorities of various services and predict and plan channel resources. Based on the network channel delay prediction results, if the predicted delay is greater than the preset limit of the service, the channel carrying service is automatically adjusted to perform channel switching. If it is a single channel, wireless resources are scheduled according to the service priority.

[0012] As a preferred solution of the channel isolation wireless resource scheduling method based on service trusted access described in the present invention, wherein: slicing and hard isolating the wireless network channel and dividing the logical resources in the hard isolation channel include: dividing the wireless resources into time slots for uplink and downlink, and dividing the resource blocks in each time slot;

[0013] The resource block division includes delay guarantee class division and resource competition class division, each type of resource block scheduling granularity is divided according to percentage, each type of resource block is distinguished from a wireless network identifier, and the terminal accesses the network by searching for a preset identifier;

[0014] If there is a service with guaranteed delay, the service data is bound to a fixed resource block in each time slot, and the resource block is mapped to different physical ports of the wireless base station / access point.

[0015] As a preferred solution of the channel isolation wireless resource scheduling method based on service trusted access described in the present invention, wherein: establishing a network channel delay prediction model includes: designing a wireless channel resource and computing power monitoring interface, and the wireless communication terminal, wireless access point / base station and time-sensitive switching network involved in the wireless link are equipped with the monitoring interface program;

[0016] Deploy the monitoring interface collection server on the wireless controller / core device side to collect real-time data with each network element of the wireless network through in-band and out-of-band channels. Build the industry wireless link / channel topology architecture based on the information collected by each network element of the wireless network.

[0017] Combined with network topology and channel service monitoring and real-time channel resource occupancy monitoring, wireless channel resources are predicted and a graph neural network is constructed;

[0018] Each pair of network elements is regarded as a node of the graph neural network, and the network elements that have an interference relationship with channel resource allocation such as burst business traffic are regarded as the edges of the graph neural network to establish a network channel delay prediction model.

[0019] As a preferred solution of the channel isolation wireless resource scheduling method based on service trusted access described in the present invention, the network channel delay prediction model also includes: when the predicted delay is greater than the service preset limit, the GraphSAGE algorithm enhanced by the multi-head attention mechanism is combined to analyze the adjacent available wireless channel resources and perform wireless terminal service migration scheduling.

[0020] As a preferred solution of the channel isolation wireless resource scheduling method based on service trusted access described in the present invention, wherein: combining the GraphSAGE algorithm enhanced by the multi-head attention mechanism, analyzing adjacent available wireless channel resources, and performing wireless terminal service migration scheduling also includes:

[0021] A two-dimensional random geometric graph with M network wireless access points / base stations is established, and each network communication terminal pair is used as a node of the graph neural network, where the network wireless access point / base station t i (1≤i≤M) are randomly placed at t according to the uniform distribution rule. i ∈[-E,E] 2 Within the range, the corresponding network communication terminal r i Place them randomly in r according to the uniform distribution rule i ∈[t i -E / 4,t i +E / 4] 2 within the scope;

[0022] All network wireless access point / base station channels are subject to Rayleigh fading. Under fading conditions, the channel matrix is ​​H M×M , where the element in row i and column j is For the transmitter t j Network communication terminal i The channel characteristic parameters between is the channel gain, is the fading coefficient that obeys the Rayleigh distribution;

[0023] Fixed channel gain in a network environment Change the channel fading coefficient for each training example in the training set It represents that the channel parameters change with the location of the network communication terminal, completing a sampling and training;

[0024] Build an overall network framework for network resource scheduling, including the UWMMSE power calculation iteration framework and the EGAT enhanced graph attention network training process embedded in each layer iteration.

[0025] As a preferred solution of the channel isolation wireless resource scheduling method based on service trusted access described in the present invention, the UWMMSE power calculation iteration framework and the EGAT enhanced graph attention network training process embedded in each layer iteration include:

[0026] The UWMMSE is composed of K WMMSE iterative layers, where the kth layer consists of It consists of three parts, which represent the decoding parameters, auxiliary parameters and encoding parameters of node i in this layer respectively; each layer of UWMMSE consists of an EGAT graph neural network as the learning part of the model, and the learning parameter is the slope and offset

[0027] The communication network is modeled as a graph and input into the graph neural network GNN. Indicates that the figure A collection of nodes It is composed of a network wireless access point / base station and a network communication terminal pair in wireless resource scheduling and delay guarantee, and a network communication terminal and an associated network wireless access point / base station pair represent a node;

[0028] The network wireless access point / base station to which you will associate j For network communication terminals i The channel quality is represented by the edge feature from node j to node i where f ξ Indicates the number of edge features, based on the type of edge E ij (1≤i,j≤M,i≠j) constitutes the edge set of the graph picture Each node Each has its own eigenvector where f η Represents the number of node features; in each UWMMSE iteration layer and Corresponding to an independent set of EGAT training networks and training parameters.

[0029] As a preferred solution of the channel isolation wireless resource scheduling method based on service trusted access described in the present invention, it also includes: sending the collected training set channel matrix to EGAT for training;

[0030] The calculation formula of EGAT's attention coefficient is:

[0031] e ij =σ((w α ) Τ [W v η i ||W vη j ||W ρ ξ ij ])

[0032] in, and is a pre-set parameter, || represents the aggregation operation, σ o (·) is a nonlinear activation function;

[0033] E ij Normalized with the softmax function, we get

[0034]

[0035] in, is the set of adjacent nodes of node i, including node i itself;

[0036] The edge aggregation calculation formula is:

[0037]

[0038] in, is a learnable parameter;

[0039] The aggregated features are sent to the fully connected layer and then pass through the nonlinear activation function σ o (·) After processing, the final training output is obtained, which is expressed as:

[0040]

[0041] in, σ o (·) is the Sigmoid function.

[0042] In a second aspect, the present invention provides a channel isolation wireless resource scheduling system based on service trusted access, comprising:

[0043] A network construction module is used to build a wireless network according to specific industry application scenarios, and to perform hard isolation of wireless network channels and logical resource division within hard isolation channels in accordance with the industry business security isolation requirements;

[0044] Quantum key distribution and authentication module, which is used to set up a quantum key distribution and authentication platform at the front end of the industry application platform, apply quantum communication terminals on the wireless network service side, and build an end-to-end quantum encryption channel;

[0045] The network delay prediction module is used to build a graph neural network and establish a network channel delay prediction model based on quantum key wireless trusted access to wireless networks, combined with network topology and channel service monitoring and real-time channel resource occupancy monitoring;

[0046] The wireless channel resource scheduling module is used to preset the priorities of various services and predict and plan channel resources. Based on the network channel delay prediction results, if the predicted delay is greater than the preset limit of the service, the channel carrying service is automatically adjusted to perform channel switching. If it is a single channel, wireless resource scheduling is performed according to the service priority.

[0047] In a third aspect, the present invention provides a computing device, comprising:

[0048] Memory and processor;

[0049] The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions. When the computer-executable instructions are executed by the processor, the steps of the channel isolation wireless resource scheduling method based on service trusted access are implemented.

[0050] In a fourth aspect, the present invention provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the steps of the channel-isolated wireless resource scheduling method based on service trusted access.

[0051] Compared with the prior art, the present invention has the following beneficial effects: the present invention uses wireless resource blocks as scheduling objects to achieve isolation at the wireless resource level; based on the business model and channel time slots, the GraphSAGE algorithm enhanced by the multi-head attention mechanism is used to predict the network channel delay, thereby achieving network resource isolation and ensuring business real-time performance. On the basis of wireless resource block isolation, combined with the high security of quantum encryption, a quantum key wireless trusted access wireless network is constructed to meet the needs of business isolation at different security levels; by establishing a network channel resource monitoring system that goes deep into the network extremities and business terminals, real-time monitoring of business application requirements, network resource status and trends is achieved, and efficient allocation of network resources on demand is achieved based on demand trends, thus building a broadband wireless network that is both secure and can efficiently utilize network resources, which can simultaneously meet business requirements at different security levels and help reduce costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative labor. Among them:

[0053] Figure 1 A method flow chart of a channel isolation wireless resource scheduling method and system based on service trusted access according to an embodiment of the present invention;

[0054] Figure 2 A GNN-based wireless resource scheduling and delay guarantee topology structure model diagram of a channel isolation wireless resource scheduling method and system based on service trusted access according to an embodiment of the present invention;

[0055] Figure 3 A UWMMSE iterative algorithm flow chart of a channel isolation wireless resource scheduling method and system based on service trusted access according to an embodiment of the present invention;

[0056] Figure 4 An EGAT training flow chart of a channel isolation wireless resource scheduling method and system based on service trusted access according to an embodiment of the present invention;

[0057] Figure 5 A wireless communication system diagram of a channel isolation wireless resource scheduling method and system based on service trusted access according to an embodiment of the present invention;

[0058] Figure 6 This is a simulation result diagram of a channel isolation wireless resource scheduling method and system based on service trusted access described in an embodiment of the present invention. DETAILED DESCRIPTION

[0059] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0060] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0061] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0062] The present invention is described in detail with reference to schematic diagrams. When describing the embodiments of the present invention, for the sake of convenience, the cross-sectional diagrams showing the device structure will not be partially enlarged according to the general scale, and the schematic diagrams are only examples, which should not limit the scope of protection of the present invention. In addition, in actual production, the three-dimensional dimensions of length, width and depth should be included.

[0063] At the same time, in the description of the present invention, it should be noted that the directions or positional relationships indicated by the terms "upper, lower, inner and outer" are based on the directions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore cannot be understood as limiting the present invention. In addition, the terms "first, second or third" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0064] In the present invention, unless otherwise clearly specified and limited, the terms "install, connect, connect" should be understood in a broad sense, for example: it can be a fixed connection, a detachable connection or an integral connection; it can also be a mechanical connection, an electrical connection or a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0065] Example 1

[0066] Reference Figure 1 to Figure 5 , is an embodiment of the present invention, which provides a channel isolation wireless resource scheduling method based on service trusted access, including:

[0067] S1: Build a wireless network according to specific industry application scenarios, and perform hard isolation of wireless network channels and logical resource division within hard isolation channels in combination with the industry service security isolation requirements;

[0068] In an embodiment of the present application, slicing and hard isolating a wireless network channel and dividing logical resources within the hard isolated channel include: dividing the wireless resources into time slots for uplink and downlink, and dividing resource blocks within each time slot;

[0069] The resource block division includes delay guarantee class division and resource competition class division, each type of resource block scheduling granularity is divided according to percentage, each type of resource block is distinguished from a wireless network identifier, and the terminal accesses the network by searching for a preset identifier;

[0070] If there is a service with guaranteed delay, the service data is bound to a fixed resource block in each time slot, and the resource block is mapped to different physical ports of the wireless base station / access point.

[0071] In an optional embodiment, the wireless resources are divided into time slots, and each time slot is further divided into resource blocks, with a latency guarantee mechanism. Wireless resources can be divided into time slots of several ms according to the wireless technology system, and the time slots of several ms are further divided into fine-grained granularity to form resource blocks. Resource blocks are installed to arrange services with different isolation requirements. At the same time, combined with service scheduling requirements, resource blocks are exclusive, services with the same isolation level are preemptively shared, and flexibly scheduled, etc., to meet both service isolation requirements and latency requirements. Wireless resource scheduling is combined with the northbound interface of the wireless network system, linked with the wireless scheduling platform, pre-arranged in combination with network planning and division, and network resource scheduling and monitoring.

[0072] It should be noted that based on the priority guarantee of latency services, wireless network resources are divided into two categories: latency guarantee and resource competition. This mechanism gives priority to latency guarantee service requirements to a certain extent, and other types of services need to compete for the remaining resource blocks. It is necessary to realize real-time monitoring of various types of current wireless service terminals and wireless base stations / access points based on wireless channel resources and computing power monitoring, and predict the use of wireless resources in combination with the wireless network channel latency prediction model. For latency predictions that exceed the preset value, the GraphSAGE algorithm enhanced by the multi-head attention mechanism is combined to analyze adjacent available wireless channel resources, perform wireless terminal service migration scheduling, and further ensure performance indicators such as service latency.

[0073] It should also be noted that in the embodiment of the present application, the wireless communication terminal is connected to the wireless access point / base station through wireless air interface resources, and the wireless access point / base station is connected to the wireless controller / core device through a time-sensitive switching network. In order to achieve full service delay guarantee, it is necessary to design in combination with the wired communication network. For a simple hierarchical point-to-point network, the optical cable can be directly pulled by the physical interface of the wireless base station / access point. For networks with 2 layers or above, a TSN delay guarantee switching network can be built between the wireless base station / access point and the data aggregation node or service master station.

[0074] S2: Set up a quantum key distribution and authentication platform at the front end of the industry application platform, apply quantum communication terminals on the wireless network service side, and build an end-to-end quantum encryption channel;

[0075] In an optional embodiment, a quantum key trusted certificate issuance and authentication platform is deployed on the wireless dispatching platform side, and the quantum key trusted certificate issuance and authentication platform is interconnected through a data interface to obtain quantum key certificates and authentication results. Quantum key certificates are used for various communication terminals (including those integrated with business terminals), and the quantum key trusted certificate authentication platform realizes terminal trusted access. Various services use quantum keys to encrypt data, achieve high-security isolation of services, and build quantum key wireless trusted access wireless networks.

[0076] It should be noted that the embodiment of the present application establishes an independent quantum key trusted certificate issuance and authentication platform in the wireless communication network. Based on the frequency isolation and time slot isolation of the wireless channel, combined with the high security and unbreakable characteristics of quantum communication, an end-to-end quantum encryption channel is constructed to enable the same wireless network to carry data with different security levels, further enhancing the security of wireless channel data applications.

[0077] S3: Based on quantum key wireless trusted access to wireless networks, combined with network topology and channel service monitoring and real-time channel resource occupancy monitoring, a graph neural network is constructed to establish a network channel delay prediction model;

[0078] In the embodiment of the present application, establishing a network channel delay prediction model includes: designing a wireless channel resource and computing power monitoring interface, and the wireless communication terminal, wireless access point / base station and time-sensitive switching network involved in the wireless link are equipped with the monitoring interface program;

[0079] Deploy the monitoring interface collection server on the wireless controller / core device side to collect real-time data with each network element of the wireless network through in-band and out-of-band channels. Build the industry wireless link / channel topology architecture based on the information collected by each network element of the wireless network.

[0080] Combined with network topology and channel service monitoring and real-time channel resource occupancy monitoring, wireless channel resources are predicted and a graph neural network is constructed;

[0081] Each pair of network elements is regarded as a node of the graph neural network, and the network elements that have an interference relationship with channel resource allocation such as burst business traffic are regarded as the edges of the graph neural network to establish a network channel delay prediction model.

[0082] In an embodiment of the present application, the network channel delay prediction model also includes: when the predicted delay is greater than the preset limit of the service, the GraphSAGE algorithm enhanced by the multi-head attention mechanism is combined to analyze the adjacent available wireless channel resources and perform wireless terminal service migration scheduling.

[0083] like Figure 2 As shown, in the embodiment of the present application, the GraphSAGE algorithm enhanced by the multi-head attention mechanism is combined to analyze the adjacent available wireless channel resources, and the wireless terminal service migration scheduling also includes:

[0084] A two-dimensional random geometric graph with M network wireless access points / base stations is established, and each network communication terminal pair is used as a node of the graph neural network, where the network wireless access point / base station t i (1≤i≤M) are randomly placed at t according to the uniform distribution rule. i ∈[-E,E] 2 Within the range, the corresponding network communication terminal r iPlace them randomly in r according to the uniform distribution rule i ∈[t i -E / 4,t i +E / 4] 2 within the scope;

[0085] All network wireless access point / base station channels are subject to Rayleigh fading. Under fading conditions, the channel matrix is ​​H M×M , where the element in row i and column j is For the transmitter t j Network communication terminal i The channel characteristic parameters between is the channel gain, is the fading coefficient that obeys the Rayleigh distribution;

[0086] It should be noted that the underlying topology of the network wireless access point / base station is fixed, while the underlying topology of the network communication terminal is subject to change.

[0087] Fixed channel gain in a network environment Change the channel fading coefficient for each training example in the training set It represents that the channel parameters change with the location of the network communication terminal, completing a sampling and training;

[0088] In an optional embodiment, the training set may be sampled 640,000 times.

[0089] Build an overall network framework for network resource scheduling, including the UWMMSE power calculation iteration framework and the EGAT enhanced graph attention network training process embedded in each layer iteration.

[0090] like Figure 3 As shown, in an embodiment of the present application, the UWMMSE power calculation iteration framework and the EGAT enhanced graph attention network training process embedded in each layer iteration include:

[0091] The UWMMSE is composed of K WMMSE iterative layers, where the kth layer consists of It consists of three parts, which represent the decoding parameters, auxiliary parameters and encoding parameters of node i in this layer respectively; each layer of UWMMSE consists of an EGAT graph neural network as the learning part of the model, and the learning parameter is the slope and offset

[0092] Specifically, the slope and offset are expressed as:

[0093]

[0094] in, is the slope of node M at layer k, is the offset of node M at layer k, ψ(·) is a function, and is the weight parameter of the function ψ.

[0095] The decoding parameters, auxiliary parameters and encoding parameters are expressed as:

[0096]

[0097] Among them, α(·) is the windowing function, which is used to Limit to the maximum available value In the first iteration, the initial value is After the last round of iteration, t i The transmit power is set to

[0098] In an optional embodiment, the EGAT training network is used as an auxiliary parameter Calculating the slope in the formula and offset Design a smarter update strategy to accelerate its convergence, such as Figure 4 As shown in the figure, after using this update strategy, only K≤5 UWMMSE iterations are needed to achieve good performance, while the traditional WMMSE algorithm without GNN embedding requires hundreds of iterations to achieve similar performance.

[0099] The communication network is modeled as a graph and input into the graph neural network GNN. Indicates that the figure A collection of nodes It is composed of a network wireless access point / base station and a network communication terminal pair in wireless resource scheduling and delay guarantee, and a network communication terminal and an associated network wireless access point / base station pair represent a node;

[0100] It should be noted that since each iterative layer inputs a GNN with the same structure, the number of iterative layers k is ignored in the following EGAT parameters.

[0101] The network wireless access point / base station to which you will associate j For network communication terminals i The channel quality is represented by the edge feature from node j to node i where f ξ Indicates the number of edge features, based on the type of edge E ij (1≤i,j≤M,i≠j) constitutes the edge set of the graph picture Each node Each has its own eigenvector where f ηRepresents the number of node features; in each UWMMSE iteration layer and Corresponding to an independent set of EGAT training networks and training parameters.

[0102] It should be noted that in order to shorten the calculation time, EGAT only contains one layer of graph attention mechanism. Unlike ordinary GAT, it adds one-dimensional edge features when calculating the attention mechanism coefficient and the aggregated edge features, so that this method can better distinguish the channel characteristics of the network communication terminal and the associated network wireless access point / base station by extracting edge information.

[0103] In the embodiment of the present application, it also includes: sending the acquired training set channel matrix to EGAT for training;

[0104] The calculation formula of EGAT's attention coefficient is:

[0105] e ij =σ((w α ) Τ [W v η i ||W v η j ||W ρ ξ ij ])

[0106] in, and is a pre-set parameter, || represents the aggregation operation, σ o (·) is a nonlinear activation function;

[0107] In an optional embodiment, the nonlinear activation function may be a Leaky-ReLU function, which is defined as:

[0108]

[0109] Among them, α L is a small positive real number. In the embodiment of the present application, α is taken L =0.01.

[0110] E ij Normalized with the softmax function, we get

[0111]

[0112] in, is the set of adjacent nodes of node i, including node i itself;

[0113] The edge aggregation calculation formula is:

[0114]

[0115] in, is a learnable parameter;

[0116] The aggregated features are sent to the fully connected layer and then pass through the nonlinear activation function σ o (·) After processing, the final training output is obtained, which is expressed as:

[0117]

[0118] in, σ o (·) is the Sigmoid function.

[0119] In an optional embodiment, the Sigmoid function is defined as:

[0120]

[0121] o i The output results will enter the UWMMSE backbone structure to participate in subsequent calculations; the collected training set channel matrix will be sent to EGAT for training. Figure 3 As shown, for node i, each layer of UWMMSE embeds two EGATs, whose inputs are channel matrices and outputs are o i Auxiliary parameters The slope With offset All embedded EGATs are trained simultaneously, and the loss function is obtained after obtaining the network resource scheduling vector.

[0122] S4: Preset the priorities of various services and predict and plan channel resources. Based on the network channel delay prediction results, if the predicted delay is greater than the preset limit of the service, the channel carrying service is automatically adjusted to perform channel switching. If it is a single channel, wireless resource scheduling is performed according to the service priority.

[0123] In the embodiment of the present application, a monitoring interface covering all terminals and networks is established in the wireless resource scheduling network. From the original passive monitoring of only monitoring network resources, it is actively extended to the wireless terminal measurement required by the business. The wireless channel resource and computing power monitoring interface is composed of the terminal side interface program and the platform side / network scheduling program, which regularly or actively reports business characteristic information such as business type, transmission frequency, bandwidth, etc., to realize real-time monitoring of resources and computing power network.

[0124] It should be noted that by combining the wireless network channel delay prediction model, the wireless channel resource and computing power monitoring interface, and the wireless resource scheduling interface, the wireless resource scheduling requirements are analyzed, and the wireless resource scheduling interface is called to schedule the wireless resources.

[0125] It should also be noted that the embodiments of the present application combine quantum key wireless trusted access, wireless resource block division, and a network channel delay prediction model established based on the GraphSAGE algorithm enhanced by a multi-head attention mechanism for different isolation levels of wireless bearers and different communication indicator requirements, and propose a wireless channel isolation wireless resource scheduling method and system, which provides a solution for on-demand resource and security assurance for different services on the same network. Based on this method, a wireless network with trusted access and service time-sensitive guarantees can be constructed.

[0126] The above is a schematic scheme of a channel isolation wireless resource scheduling method based on business trusted access in this embodiment. It should be noted that the technical scheme of the channel isolation wireless resource scheduling system based on business trusted access and the technical scheme of the channel isolation wireless resource scheduling method based on business trusted access belong to the same concept, and the details of the technical scheme of the channel isolation wireless resource scheduling system based on business trusted access in this embodiment that are not described in detail can all be referred to the description of the technical scheme of the channel isolation wireless resource scheduling method based on business trusted access.

[0127] In this embodiment, a channel isolation wireless resource scheduling system based on service trusted access includes:

[0128] A network construction module is used to build a wireless network according to specific industry application scenarios, and to perform hard isolation of wireless network channels and logical resource division within hard isolation channels in accordance with the industry business security isolation requirements;

[0129] Quantum key distribution and authentication module, which is used to set up a quantum key distribution and authentication platform at the front end of the industry application platform, apply quantum communication terminals on the wireless network service side, and build an end-to-end quantum encryption channel;

[0130] The network delay prediction module is used to build a graph neural network and establish a network channel delay prediction model based on quantum key wireless trusted access to wireless networks, combined with network topology and channel service monitoring and real-time channel resource occupancy monitoring;

[0131] The wireless channel resource scheduling module is used to preset the priorities of various services and predict and plan channel resources. Based on the network channel delay prediction results, if the predicted delay is greater than the preset limit of the service, the channel carrying service is automatically adjusted to perform channel switching. If it is a single channel, wireless resource scheduling is performed according to the service priority.

[0132] In an optional embodiment, the quantum key distribution authentication module, the network delay prediction module, the service feature monitoring module and the wireless channel resource scheduling module provide the following functions: through the combination of quantum + wireless, quantum encryption communication channels are provided for different types of services to improve wireless isolation and security; the wireless channel delay is predicted through the network channel delay prediction model established based on the GraphSAGE algorithm enhanced by the multi-head attention mechanism; based on the network delay prediction, combined with the time-sensitive network design, the delay is further guaranteed; the service communication frequency, the number of communication bytes and the service priority are monitored, and the wireless channel resource scheduling is performed in combination with the network channel delay prediction.

[0133] This embodiment further provides a computing device, which is applicable to the case of a channel isolation wireless resource scheduling method based on service trusted access, including:

[0134] A memory and a processor; the memory is used to store computer executable instructions, and the processor is used to execute computer executable instructions to implement the channel isolation wireless resource scheduling method based on service trusted access as proposed in the above embodiment.

[0135] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, the method for implementing channel isolation wireless resource scheduling based on service trusted access as proposed in the above embodiment is implemented.

[0136] The storage medium proposed in this embodiment and the method for implementing channel isolation wireless resource scheduling based on service trusted access proposed in the above embodiment belong to the same inventive concept. The technical details not described in detail in this embodiment can be referred to the above embodiment, and this embodiment has the same beneficial effects as the above embodiment.

[0137] Through the above description of the implementation methods, the technicians in the relevant field can clearly understand that the present invention can be implemented by means of software and necessary general hardware, and of course can also be implemented by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ReadOnly, Memory, ROM), random access memory (RandomAccess Memory, RAM), flash memory (FLASH), hard disk or optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform the methods of various embodiments of the present invention.

[0138] Example 2

[0139] Reference Figure 6 , is an embodiment of the present invention. This embodiment is different from the first embodiment in that this embodiment verifies the beneficial effects of the present invention through simulation experiments.

[0140] In the embodiment of the present application, a network communication terminal and an associated network wireless access point / base station pair (r i ,t i ) between the channel characteristics h ii is the node feature of node i (i.e., η i =h ii ), connect the network wireless access point / base station to j To network communication terminal i The channel characteristics h ij As the edge feature from node j to node i (i.e., ξ ij =h ij ), represents the interference between nodes. According to the feature definition, the channel features and the interference features between nodes are used to speed up the auxiliary parameter w i The iterative convergence speed is improved to achieve the optimization of network resource scheduling of network wireless access points / base station channels associated with network communication terminals.

[0141] according to Figure 6 The simulation results shown show the data distribution of the total system throughput of four different categories including Tr, WMMSE, GCN, and EGAT. The data distribution of the Tr category is narrower and the median is lower, which means that the system throughput of the Tr method changes less and the overall performance is not as good as other categories. There is room for optimization to improve the system throughput. The data distribution of the two categories of WMMSE and GCN is relatively wide, and the median position is close. This shows that the system throughput of the two methods varies greatly, but their medians are close, indicating that the overall performance is similar, and one of the methods can be selected according to application requirements. The data distribution of the EGAT category is the widest and the median is the highest, which means that the system throughput of the EGAT method varies the most, but also shows the best performance. Although the data volatility is large, in most cases, the EGAT method performs better than other categories. In summary, the UWMMSE wireless network data scheduling method based on the EGAT training mechanism proposed in the present invention has a better application prospect in improving network resource scheduling, whether compared with the traditional iterative optimization method or the UWMMSE optimization method based on the GCN training mechanism.

[0142] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A channel isolation wireless resource scheduling method based on service trusted access, characterized in that: include: Build a wireless network based on specific industry application scenarios, and perform hard isolation of wireless network channels and logical resource division within hard isolation channels in accordance with the industry service security isolation requirements; Set up a quantum key distribution and authentication platform at the front end of the industry application platform, apply quantum communication terminals on the wireless network service side, and build an end-to-end quantum encryption channel; Based on quantum key wireless trusted access to wireless networks, combined with network topology and channel service monitoring and real-time channel resource occupancy monitoring, a graph neural network is constructed to establish a network channel delay prediction model. Preset the priorities of various services and predict and plan channel resources. Based on the network channel delay prediction results, if the predicted delay is greater than the preset limit of the service, the channel carrying service is automatically adjusted to perform channel switching. If it is a single channel, wireless resources are scheduled according to the service priority.

2. The channel isolation wireless resource scheduling method based on service trusted access according to claim 1, characterized in that: Slicing and hard isolating wireless network channels and dividing logical resources within the hard isolated channels include: dividing wireless resources into time slots for uplink and downlink, and dividing resource blocks within each time slot; The resource block division includes delay guarantee class division and resource competition class division, each type of resource block scheduling granularity is divided according to percentage, each type of resource block is distinguished from a wireless network identifier, and the terminal accesses the network by searching for a preset identifier; If there is a service with guaranteed delay, the service data is bound to a fixed resource block in each time slot, and the resource block is mapped to different physical ports of the wireless base station / access point.

3. The channel isolation wireless resource scheduling method based on service trusted access according to claim 2, characterized in that: Establishing a network channel delay prediction model includes: designing a wireless channel resource and computing power monitoring interface, and equipping the wireless communication terminal, wireless access point / base station, and time-sensitive switching network involved in the wireless link with the monitoring interface program; Deploy the monitoring interface collection server on the wireless controller / core device side to collect real-time data with each network element of the wireless network through in-band and out-of-band channels. Build the industry wireless link / channel topology architecture based on the information collected by each network element of the wireless network. Combined with network topology and channel service monitoring and real-time channel resource occupancy monitoring, wireless channel resources are predicted and a graph neural network is constructed; Each pair of network elements is regarded as a node of the graph neural network, and the network elements that have an interference relationship with channel resource allocation such as burst business traffic are regarded as the edges of the graph neural network to establish a network channel delay prediction model.

4. The channel isolation wireless resource scheduling method based on service trusted access according to claim 3, characterized in that: The network channel delay prediction model also includes: when the predicted delay is greater than the preset limit of the service, the GraphSAGE algorithm enhanced by the multi-head attention mechanism is combined to analyze the adjacent available wireless channel resources and perform wireless terminal service migration scheduling.

5. The channel isolation wireless resource scheduling method based on service trusted access according to claim 4, characterized in that: In combination with the GraphSAGE algorithm enhanced by the multi-head attention mechanism, the adjacent available wireless channel resources are analyzed, and the wireless terminal service migration scheduling also includes: A two-dimensional random geometric graph with M network wireless access points / base stations is established, and each network communication terminal pair is used as a node of the graph neural network, where the network wireless access point / base station t i (1≤i≤M) are randomly placed at t according to the uniform distribution rule. i ∈[-E,E] 2 Within the range, the corresponding network communication terminal r i Place them randomly in r according to the uniform distribution rule i ∈[t i -E / 4,t i +E / 4] 2 within the scope; All network wireless access point / base station channels are subject to Rayleigh fading. Under fading conditions, the channel matrix is ​​H M×M , where the element in row i and column j is For the transmitter t j Network communication terminal i The channel characteristic parameters between is the channel gain, is the fading coefficient that obeys the Rayleigh distribution; Fixed channel gain in a network environment Change the channel fading coefficient for each training example in the training set It represents that the channel parameters change with the location of the network communication terminal, completing a sampling and training; Build an overall network framework for network resource scheduling, including the UWMMSE power calculation iteration framework and the EGAT enhanced graph attention network training process embedded in each layer iteration.

6. The channel isolation wireless resource scheduling method based on service trusted access according to claim 5, characterized in that: The UWMMSE power calculation iteration framework and the EGAT enhanced graph attention network training process embedded in each layer iteration include: The UWMMSE is composed of K WMMSE iterative layers, where the kth layer consists of It consists of three parts, which represent the decoding parameters, auxiliary parameters and encoding parameters of node i in this layer respectively; each layer of UWMMSE consists of an EGAT graph neural network as the learning part of the model, and the learning parameter is the slope and offset The communication network is modeled as a graph and input into the graph neural network GNN. Indicates that the figure A collection of nodes It is composed of a network wireless access point / base station and a network communication terminal pair in wireless resource scheduling and delay guarantee, and a network communication terminal and an associated network wireless access point / base station pair represent a node; The network wireless access point / base station to which you will associate j For network communication terminals i The channel quality is represented by the edge feature from node j to node i where f ξ Indicates the number of edge features, based on the type of edge E ij (1≤i,j≤M,i≠j) constitutes the edge set of the graph picture Each node Each has its own eigenvector where f η Represents the number of node features; in each UWMMSE iteration layer and Corresponding to an independent set of EGAT training networks and training parameters.

7. The channel isolation wireless resource scheduling method based on service trusted access according to claim 6, characterized in that: Also includes: Send the acquired training set channel matrix to EGAT for training; The calculation formula of EGAT's attention coefficient is: e ij =σ((w α ) Τ [W v or i ||W v or j ||W ρ x ij ]) in, and is a pre-set parameter, || represents the aggregation operation, σ o (·) is a nonlinear activation function; E ij Normalized with the softmax function, we get in, is the set of adjacent nodes of node i, including node i itself; The edge aggregation calculation formula is: in, is a learnable parameter; The aggregated features are sent to the fully connected layer and then pass through the nonlinear activation function σ o (·) After processing, the final training output is obtained, which is expressed as: in, σ o (·) is the Sigmoid function.

8. A channel isolation wireless resource scheduling system based on service trusted access, characterized in that: include: A network construction module is used to build a wireless network according to specific industry application scenarios, and to perform hard isolation of wireless network channels and logical resource division within hard isolation channels in accordance with the industry business security isolation requirements; Quantum key distribution and authentication module, which is used to set up a quantum key distribution and authentication platform at the front end of the industry application platform, apply quantum communication terminals on the wireless network service side, and build an end-to-end quantum encryption channel; The network delay prediction module is used to build a graph neural network and establish a network channel delay prediction model based on quantum key wireless trusted access to wireless networks, combined with network topology and channel service monitoring and real-time channel resource occupancy monitoring; The wireless channel resource scheduling module is used to preset the priorities of various services and predict and plan channel resources. Based on the network channel delay prediction results, if the predicted delay is greater than the preset limit of the service, the channel carrying service is automatically adjusted to perform channel switching. If it is a single channel, wireless resource scheduling is performed according to the service priority.

9. An electronic device, comprising: Memory and processor; The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions. When the computer-executable instructions are executed by the processor, the steps of the channel isolation wireless resource scheduling method based on service trusted access as described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the steps of the channel-isolated wireless resource scheduling method based on service trusted access as described in any one of claims 1 to 7.

Images