Network service management method and system for office area

Abstract

The invention relates to the technical field of network services, in particular to a network service management method and system for an office area, and the method comprises the steps: a preset request node issues a request task set to a preset management node, and the management node carries out the clustering of the request task set, and dividing the request task set into different types of sub-request task sets. According to the method, the request task set is subjected to clustering analysis through the management node, the similar tasks are divided into the same subset, and the position of the task clustering center is determined, so that the task scheduling complexity can be remarkably reduced, repeated resource allocation is avoided, and the overall task processing efficiency is improved; and the resources of the edge nodes are dynamically allocated by utilizing an optimization algorithm, the optimal configuration of calculation, storage and bandwidth resources is ensured, and through mathematical modeling and solution, the resource waste can be minimized and the resource utilization rate can be improved on the premise of meeting task requirements.

Description

Technical Field

[0001] The present invention relates to the technical field of network services, and in particular to a network service management method and system for an office area. Background Art

[0002] Traditional methods usually do not use task clustering analysis, and task scheduling is based on a single task, lacking analysis of the similarity between tasks, which leads to high complexity of task scheduling and may cause repeated allocation and waste of resources; and when traditional methods process a large number of tasks, the scheduling efficiency is low, and it is difficult to obtain good performance in a large-scale environment; and the resource allocation of traditional methods is usually static or based on preset strategies, and lacks a dynamic adjustment mechanism. This method cannot be optimized in real time according to the actual load and resource status of the edge nodes, which may lead to uneven allocation of computing, storage and bandwidth resources, resulting in resource waste or processing bottlenecks, affecting the efficiency and performance of the overall system; and in traditional methods, the location data of task scheduling and task processing nodes are usually transmitted in plain text, which is vulnerable to the risk of man-in-the-middle attacks or data leakage. Especially when processing sensitive information, traditional methods cannot guarantee the privacy of location data and user data, which can easily lead to information theft or tampering; and traditional methods lack the support of edge computing, and tasks need to be transmitted to a central server far away from the user for processing, resulting in a long response time and low efficiency. Moreover, due to the lack of optimization of dynamic resource allocation and scheduling, tasks may be delayed due to insufficient computing and storage resources or bandwidth limitations. Summary of the invention

[0003] The technical problem to be solved by the present invention is to overcome the disadvantages of the above-mentioned prior art and provide a network service management method and system for an office area.

[0004] The technical solution adopted to solve the above technical problems is: a network service management method for an office area, comprising: The preset request node publishes the request task set to the preset management node, and the management node clusters the request task set to divide the request task set into different types of sub-request task sets, and obtains the positions of the task clustering centers of the different types of sub-request task sets; Encrypting the position of the task cluster center to obtain a first encrypted position, and transmitting the task cluster center and the first encrypted position to the edge node set; Constructing a resource allocation optimization model between the edge node sets, and solving the resource allocation optimization model based on an ant algorithm to obtain an optimal resource allocation between the edge node sets; The edge node set based on the optimal resource allocation sends the task cluster center and the first encrypted position to a preset task processing node. If the task processing node decides to execute the task type corresponding to the task cluster center, the position of the task processing node is encrypted to obtain a second encrypted position, and the second encrypted position is transmitted to the corresponding edge node. The edge node calculates the distance between the first encrypted position and the second encrypted position, encrypts the distance to obtain an encrypted distance, transmits the encrypted distance to the management node, and the management node selects a task processing node according to the distance and the reliability of all task processing nodes: Preferably, the management node clusters the request task set to divide the request task set into a plurality of sub-request task sets, including: Extracting features of each request task in the request task set, wherein the features of the request task include a request task location, a request task quality requirement, and a request task budget; Calculating the Euclidean distance between the features of each request task in the request task set, and calculating the local density of the features of the request task based on the Euclidean distance; Calculating relative distances of features of each request task based on local densities of features of the request task; Calculating a decision value of an initial cluster based on the local density and relative distance of the features of the requested task, and selecting a cluster center based on the decision value; The remaining features of the requested task are assigned one by one in descending order of local density to the cluster to which the features of the requested task with the highest local density belong, so as to complete clustering.

[0005] Preferably, the calculation formula of the local density is as follows: ; in, Indicates the first The local density of the features of the requested task, Indicates the first and The Euclidean distance between the features of the requested tasks, Indicates the first The nearest neighbors of the requested task’s features, Indicates the first and Whether the requested task is characterized by a special neighbor, and ,when , then the request task set and The request task is a special neighbor. , then the request task set and The request task is not a special neighbor, and Indicates the first and The characteristics of the requested task; The calculation formula of the relative distance is as follows: ; in, Indicates the first The relative distance of the features of the requested task, Indicates the first The relative distance between the feature of the requested task and the feature of the requested task corresponding to the highest local density, Indicates the first The relative distance between the feature of the requested task and the feature of the requested task corresponding to the lowest local density, and ,and ; The calculation formula of the decision value is as follows: ; in, Represents the decision value of the initial cluster.

[0006] Preferably, encrypting the position of the task cluster center to obtain a first encrypted position includes: The position of the task cluster center is encrypted based on the public key of the management node to obtain an encrypted position, wherein the encryption formula of the position of the task cluster center is as follows: ; in, and represents the longitude and latitude of the encrypted location, represents the encryption function, and express, represents the longitude and latitude of the location of the task cluster center, Indicates the public key of the management node; The encrypted position is encrypted based on the public key of the edge node to obtain a first encrypted position, wherein the encryption formula of the encrypted position is as follows: ; in, and represents the longitude and latitude of the first encrypted position, Indicates the public key of the edge node.

[0007] Preferably, the task cluster center and the first encrypted position are transmitted to the edge node set, and then the method further includes: After receiving the task cluster center and the first encrypted position, the edge node set decrypts the first encrypted position based on the private key of the edge node to obtain an encrypted position, wherein the decryption formula of the first encrypted position is as follows: ; in, represents the decryption function, Represents the private key of the edge node; The edge node broadcasts the public key of the management node and the content of the task cluster center to the task processing nodes within its coverage.

[0008] Preferably, constructing a resource allocation optimization model between the edge node sets includes: Calculate the response time of each edge node in the edge node set, wherein the calculation formula of the response time of the edge node is as follows: ; in, Indicates the response time of the edge node, Indicates the data size corresponding to the requested task. Represents an edge node Request task Send to task processing node The emission rate is , Indicates bandwidth resources. represents Gaussian white noise, Represents an edge node Request task Send to task processing node The transmission power, Represents an edge node and task processing nodes channel gain; calculating the processing time of each edge node in the edge node set, wherein the calculation formula of the processing time of the edge node is as follows: ; in, represents the processing time of the edge node, represents the workload of the requested task, Represents an edge node Assign to request task computing resources.

[0009] Preferably, constructing the resource allocation optimization model between the edge node sets further includes: The total energy consumption of the processing task is calculated based on the response time and processing time of the edge node, wherein the calculation formula of the total energy consumption of the processing task is as follows: An objective function is constructed based on the total energy consumption of the processing task, a constraint condition is constructed based on the bandwidth resources and computing resources of the edge node, and a resource allocation optimization model is constructed based on the objective function and the constraint condition, wherein the constraint condition is expressed as follows: ; in, represents the total bandwidth of the edge nodes, represents the total computing resources of the edge nodes, Indicates the maximum transmit power.

[0010] Preferably, solving the resource allocation optimization model based on an ant algorithm to obtain an optimal resource allocation between the edge node sets includes: Initializing the parameters of the ant algorithm, wherein the parameters of the ant algorithm include the number of ants, the maximum number of iterations, the initial pheromone, the global attenuation coefficient and the local attenuation coefficient, scattering the ants at the positions of the edge nodes, and setting the pheromone at the positions of the task processing nodes; Calculating the objective function value assigned by each edge node to each task processing node, and calculating the allocation probability of each task processing node based on the objective function value; The task allocation node is determined based on the allocation probability of the task processing node.

[0011] Preferably, the calculation formula of the allocation probability is as follows: ; in, represents the probability of allocating edge nodes to task processing nodes, Indicates pheromone, represents heuristic information, represents the relative importance between pheromone and heuristic information, Indicates multiple task processing nodes that can be assigned to edge nodes, represents a subchannel in an edge node, and ; The determination formula of the task allocation node is as follows: ; in, Represents an edge node Determine the assigned task processing node , Indicates that the distribution Random numbers on represents the preset threshold parameter, Represents an edge node The last allocated processing node .

[0012] The technical solution adopted to solve the above technical problem is: a network service management system for an office area, which is applicable to the network service management method for an office area, comprising: A task classification unit, wherein the task classification unit is used for a preset request node to publish a request task set to a preset management node, and the management node clusters the request task set to divide the request task set into different types of sub-request task sets, and obtains the positions of task cluster centers of the different types of sub-request task sets; A position encryption unit, the position encryption unit is used to encrypt the position of the task cluster center to obtain a first encrypted position, and transmit the task cluster center and the first encrypted position to the edge node set; A resource allocation unit, the resource allocation unit is used to construct a resource allocation optimization model between the edge node sets, and solve the resource allocation optimization model based on an ant algorithm to obtain an optimal resource allocation between the edge node sets; A task transmission unit, the task transmission unit is used to send the task cluster center and the first encrypted position to a preset task processing node based on the edge node set with the best resource allocation, if the task processing node decides to execute the task type corresponding to the task cluster center, then encrypt the position of the task processing node to obtain a second encrypted position, and transmit the second encrypted position to the corresponding edge node; A task selection unit, wherein the task selection unit is used for the edge node to calculate the distance between the first encrypted position and the second encrypted position, encrypt the distance to obtain an encrypted distance, and transmit the encrypted distance to the management node, and the management node selects a task processing node based on the distance and the reliability of all task processing nodes.

[0013] The beneficial effects of the present invention are as follows: (1) The present invention performs cluster analysis on the request task set through the management node, divides similar tasks into the same subset, and determines the location of the task cluster center, thereby significantly reducing the complexity of task scheduling, avoiding repeated resource allocation, and improving the overall efficiency of task processing. The optimization algorithm is used to dynamically allocate the resources of the edge nodes to ensure the optimal configuration of computing, storage and bandwidth resources. Through mathematical modeling and solution, it can minimize resource waste and improve resource utilization while meeting task requirements; (2) The present invention encrypts the location of the task cluster center and the location of the task processing node to ensure sensitive The sensing information will not be stolen or tampered with during the transmission process, and the edge node will calculate the distance of the encrypted location information and transmit the result to the management node in encrypted form. This mechanism avoids the plaintext exposure of the location data and effectively prevents man-in-the-middle attacks or location privacy leaks. (3) The present invention reduces the delay in data transmission to the central server by sinking task scheduling and resource allocation to the edge node. The edge node processes tasks nearby and can respond to requests quickly. It is especially suitable for office scenarios with high real-time requirements. From task release to final execution, all links involving location and user data are encrypted for transmission to meet the needs of enterprise data privacy protection. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 A schematic flow chart of the steps of an overall method in an embodiment of the present invention; Figure 2 A schematic diagram of the system architecture of an overall system in an embodiment of the present invention.

[0015] Figure numerals: 1. Task classification unit; 2. Position encryption unit; 3. Resource allocation unit; 4. Task transmission unit; 5. Task selection unit. DETAILED DESCRIPTION

[0016] Embodiment 1, as Figure 1 As shown, the present invention proposes a network service management method for an office area, comprising: S1. The preset request node publishes the request task set to the preset management node, and the management node clusters the request task set to divide the request task set into different types of sub-request task sets, and obtains the positions of the task clustering centers of the different types of sub-request task sets; S2, encrypting the position of the task cluster center to obtain a first encrypted position, and transmitting the task cluster center and the first encrypted position to the edge node set; S3, a resource allocation optimization model between edge node sets, solving the resource allocation optimization model based on the ant algorithm to obtain the optimal resource allocation between edge node sets; S4, the edge node set based on the optimal resource allocation sends the task cluster center and the first encrypted position to the preset task processing node, if the task processing node decides to execute the task type corresponding to the task cluster center, the position of the task processing node is encrypted to obtain a second encrypted position, and the second encrypted position is transmitted to the corresponding edge node; S5. The edge node calculates the distance between the first encrypted position and the second encrypted position, encrypts the distance to obtain the encrypted distance, and transmits the encrypted distance to the management node. The management node selects a task processing node based on the distance and the reliability of all task processing nodes.

[0017] In the present invention, a request node refers to a node that initiates a request task in a network system. These request tasks are usually tasks that need to be processed by a task processing node in the network; a management node is a node in the entire system that is responsible for receiving and managing a set of request tasks. The main task of the management node is to classify, allocate, and schedule the request tasks, and make relevant decisions; a task cluster center refers to the central position of each type of task after being processed by a clustering algorithm in task clustering analysis. Clustering refers to dividing tasks into different subtask sets based on certain features. The cluster center usually represents the "representative" task or core task of this type of task; an edge node set refers to a group of nodes located at the edge of the network, which usually have the function of processing, storing or transmitting information. Edge computing places data processing on a device close to the data source. This reduces latency and bandwidth requirements. The edge node collection is used to store and forward encrypted task information. The resource allocation optimization model refers to a mathematical model that describes how to reasonably allocate computing, storage, bandwidth and other resources among multiple nodes or systems to achieve optimal system performance or benefits. A task processing node refers to a node in the network that is responsible for executing tasks. It receives tasks assigned by the management node and processes them according to the task type. During this process, the task processing node may choose to perform specific tasks based on the task type and workload. Reliability generally refers to the stability and fault tolerance of a system, node or network. The management node selects the most suitable node to process the task based on the reliability of the task processing node. Reliability may include the availability, performance, and historical performance of the node.

[0018] Embodiment 2, a network service management method for an office area proposed by the present invention, compared with embodiment 1, this embodiment further includes: the management node clusters the request task set to divide the request task set into multiple sub-request task sets, including: A1. extracting the features of each request task in the request task set, wherein the features of the request task include the request task location, the request task quality requirement, and the request task budget; A2. Calculate the Euclidean distance between the features of each request task in the request task set, and calculate the local density of the features of the request task based on the Euclidean distance; A3. Calculate the relative distance of the features of each request task based on the local density of the features of the request task; A4. Calculate the decision value of the initial cluster based on the local density and relative distance of the features of the requested task, and select the cluster center based on the decision value; A5. Assign the features of the remaining request tasks one by one in descending order of local density to the cluster to which the features of the request task with the highest local density belong to, so as to complete clustering.

[0019] In this embodiment, the requested task location refers to the geographical location or spatial location of the requested task, which is usually expressed in coordinates or the like; the requested task quality requirement refers to the quality standard or requirement of the requested task, which may be the demand for task completion effect, time, etc.; the requested task budget refers to the cost or budget that the requested task can bear.

[0020] In an optional embodiment, the calculation formula of the local density is as follows: ; in, Indicates the first The local density of the features of the requested task, Indicates the first and The Euclidean distance between the features of the requested tasks, Indicates the first The nearest neighbors of the requested task’s features, Indicates the first and Whether the requested task is characterized by a special neighbor, and ,when , then the request task set and The request task is a special neighbor. , then the request task set and The request task is not a special neighbor, and Indicates the first and The characteristics of the requested task; The calculation formula of relative distance is as follows: ; in, Indicates the first The relative distance between the features of the requested task, Indicates the first The relative distance between the feature of the requested task and the feature of the requested task corresponding to the highest local density, Indicates the first The relative distance between the feature of the requested task and the feature of the requested task corresponding to the lowest local density, and ,and ; The calculation formula of decision value is as follows: ; in, Represents the decision value of the initial cluster.

[0021] It should be noted that local density is an indicator that describes the density of distribution of data points in their neighborhood. In clustering algorithms, local density is often used to measure the relationship between data points and their neighboring points.

[0022] In an optional embodiment, encrypting the position of the task cluster center to obtain a first encrypted position includes: B1. Encrypt the location of the task cluster center based on the public key of the management node to obtain an encrypted location, wherein the encryption formula of the location of the task cluster center is as follows: ; in, and represents the longitude and latitude of the encrypted location, represents the encryption function, and express, represents the longitude and latitude of the location of the task cluster center, Indicates the public key of the management node; B2. Encrypt the encrypted position based on the public key of the edge node to obtain a first encrypted position, wherein the encryption formula of the encrypted position is as follows: ; in, and represents the longitude and latitude of the first encrypted position, Indicates the public key of the edge node.

[0023] It should be noted that the encryption function is a mathematical algorithm that performs encryption operations. It accepts original plaintext data (such as longitude and latitude) and a key as input, and outputs encrypted data after certain transformations.

[0024] In an optional embodiment, the task cluster center and the first encrypted position are transmitted to the edge node set, and then the method further includes: C1. After receiving the task cluster center and the first encrypted position, the edge node set decrypts the first encrypted position based on the private key of the edge node to obtain the encrypted position, wherein the decryption formula of the first encrypted position is as follows: ; in, represents the decryption function, Represents the private key of the edge node; C2. The edge node broadcasts the public key of the management node and the content of the task cluster center to the task processing nodes within its coverage area.

[0025] It should be noted that the first encrypted position refers to the longitude and latitude of the task cluster center encrypted by the public key of the management node, and then the result of secondary encryption by the public key of the edge node. Through this double encryption method, the location information is protected, and only the node with the corresponding private key can decrypt and restore the original information; broadcasting refers to sending information to all nodes or devices in a network. The edge node will broadcast the public key of the management node and the content of the task cluster center to the task processing nodes within its coverage area, which means that all nodes receiving the broadcast can use the broadcast information for data encryption or task processing.

[0026] In an optional embodiment, building a resource allocation optimization model between edge node sets includes: D1. Calculate the response time of each edge node in the edge node set, where the calculation formula for the response time of the edge node is as follows: ; in, Indicates the response time of the edge node, Indicates the data size corresponding to the requested task. Represents an edge node Request task Send to task processing node The emission rate is , Indicates bandwidth resources. represents Gaussian white noise, Represents an edge node Request task Send to task processing node The transmission power, Represents an edge node and task processing nodes The channel gain of D2. Calculate the processing time of each edge node in the edge node set, where the calculation formula for the processing time of the edge node is as follows: ; in, represents the processing time of the edge node, represents the workload of the requested task, Represents an edge node Assign to request task computing resources.

[0027] It should be noted that the transmission rate refers to the transmission rate when the edge node sends data to the task processing node, usually expressed in bits per second (bps). It determines the speed of data transmission. A higher transmission rate means that more data can be transmitted in a shorter time; bandwidth is the amount of data that the network can transmit per unit time, usually expressed in bits per second (bps). Bandwidth resources refer to the network capacity that can be used for data transmission. The size of the bandwidth will affect the data transmission rate. The larger the bandwidth, the faster the data transmission; Gaussian white noise is a signal noise model in which the amplitude of the noise follows a Gaussian distribution; channel gain refers to the gain or attenuation of the transmission signal caused by the propagation environment (such as distance, obstacles, reflections, etc.) during the transmission process.

[0028] In an optional embodiment, building a resource allocation optimization model between edge node sets further includes: D3. Calculate the total energy consumption of the processing task based on the response time and processing time of the edge node. The calculation formula for the total energy consumption of the processing task is as follows: ; in, represents the total energy consumption of processing tasks, It represents the average power consumption of edge nodes in processing request tasks; D4. Construct an objective function based on the total energy consumption of the processing task, construct constraints based on the bandwidth resources and computing resources of the edge nodes, and construct a resource allocation optimization model based on the objective function and constraints. The constraint expression is as follows: ; in, represents the total bandwidth of the edge nodes, represents the total computing resources of the edge nodes, Indicates the maximum transmit power.

[0029] In an optional embodiment, solving the resource allocation optimization model based on the ant algorithm to obtain the optimal resource allocation between the edge node sets includes: E1. Initialize the parameters of the ant algorithm, where the parameters of the ant algorithm include the number of ants, the maximum number of iterations, the initial pheromone, the global attenuation coefficient, and the local attenuation coefficient. Spread the ants at the positions of the edge nodes and set the pheromone at the positions of the task processing nodes. E2. Calculate the objective function value assigned to each task processing node by each edge node, and calculate the allocation probability of each task processing node based on the objective function value; E3. Determine the task allocation node based on the allocation probability of the task processing node.

[0030] It should be noted that the ant algorithm is a heuristic algorithm that simulates the foraging behavior of ants in nature and is mainly used to solve combinatorial optimization problems; the number of ants refers to the total number of simulated ant individuals in the ant algorithm. Each ant represents a possible solution. In each iteration, all ants will traverse the problem space and look for solutions. The setting of the number of ants affects the breadth and convergence speed of the search; the global attenuation coefficient represents the attenuation rate of pheromones during global updates. Global attenuation is usually applied to all paths passed by ants, affecting the pheromone concentrations of these paths; the allocation probability refers to the probability that a task is assigned to a task processing node. This probability is usually calculated based on the pheromone concentrations of each path and the objective function value of the path.

[0031] In an optional embodiment, the calculation formula of the allocation probability is as follows: ; in, represents the probability of allocating edge nodes to task processing nodes, Indicates pheromone, represents heuristic information, represents the relative importance between pheromone and heuristic information, Indicates multiple task processing nodes that can be assigned to edge nodes, represents a subchannel in an edge node, and ; The formula for determining the task allocation node is as follows: ; in, Represents an edge node Determine the assigned task processing node , Indicates that the distribution Random numbers on represents the preset threshold parameter, Represents an edge node The last allocated processing node .

[0032] Embodiment three, as Figure 2 As shown, the present invention proposes a network service management system for an office area, which is applicable to a network service management method for an office area, comprising: Task classification unit 1, task classification unit 1 is used for a preset request node to publish a request task set to a preset management node, the management node clusters the request task set to divide the request task set into different types of sub-request task sets, and obtains the location of the task clustering center of different types of sub-request task sets; Position encryption unit 2, the position encryption unit 2 is used to encrypt the position of the task cluster center to obtain a first encrypted position, and transmit the task cluster center and the first encrypted position to the edge node set; Resource allocation unit 3, resource allocation unit 3 is used to build a resource allocation optimization model between edge node sets, and solve the resource allocation optimization model based on the ant algorithm to obtain the optimal resource allocation between edge node sets; The task transmission unit 4 is used to send the task cluster center and the first encrypted position to a preset task processing node based on the edge node set with the best resource allocation. If the task processing node decides to execute the task type corresponding to the task cluster center, the position of the task processing node is encrypted to obtain a second encrypted position, and the second encrypted position is transmitted to the corresponding edge node; The task selection unit 5 is used for the edge node to calculate the distance between the first encrypted position and the second encrypted position, encrypt the distance to obtain the encrypted distance, and transmit the encrypted distance to the management node. The management node selects the task processing node according to the distance and the reliability of all task processing nodes.

[0033] The embodiments of the present invention are described in detail above with reference to the accompanying drawings, but the present invention is not limited thereto, and various changes can be made within the knowledge scope of technicians in the relevant technical field without departing from the purpose of the present invention.

Claims

1. A network service management method for an office area, characterized in that: include: The preset request node publishes the request task set to the preset management node, and the management node clusters the request task set to divide the request task set into different types of sub-request task sets, and obtains the positions of the task clustering centers of the different types of sub-request task sets; Encrypting the position of the task cluster center to obtain a first encrypted position, and transmitting the task cluster center and the first encrypted position to the edge node set; Constructing a resource allocation optimization model between the edge node sets, and solving the resource allocation optimization model based on an ant algorithm to obtain an optimal resource allocation between the edge node sets; The edge node set based on the optimal resource allocation sends the task cluster center and the first encrypted position to a preset task processing node. If the task processing node decides to execute the task type corresponding to the task cluster center, the position of the task processing node is encrypted to obtain a second encrypted position, and the second encrypted position is transmitted to the corresponding edge node. The edge node calculates the distance between the first encrypted position and the second encrypted position, encrypts the distance to obtain an encrypted distance, transmits the encrypted distance to the management node, and the management node selects a task processing node based on the distance and the reliability of all task processing nodes.

2. A network service management method for an office area according to claim 1, characterized in that: The management node clusters the request task set to divide the request task set into a plurality of sub-request task sets, including: Extracting features of each request task in the request task set, wherein the features of the request task include a request task location, a request task quality requirement, and a request task budget; Calculating the Euclidean distance between the features of each request task in the request task set, and calculating the local density of the features of the request task based on the Euclidean distance; Calculating relative distances of features of each request task based on local densities of features of the request task; Calculating a decision value of an initial cluster based on the local density and relative distance of the features of the requested task, and selecting a cluster center based on the decision value; The remaining features of the requested task are assigned one by one in descending order of local density to the cluster to which the features of the requested task with the highest local density belong, so as to complete clustering.

3. A network service management method for an office area according to claim 2, characterized in that: The calculation formula of the local density is as follows: ； in, Indicates the first The local density of the features of the requested task, Indicates the first and The Euclidean distance between the features of the requested tasks, Indicates the first The nearest neighbors of the requested task’s features, Indicates the first and Whether the requested task is characterized by a special neighbor, and ,when , then the request task set and The request task is a special neighbor. , then the request task set and The request task is not a special neighbor, and Indicates the first and The characteristics of the requested task; The calculation formula of the relative distance is as follows: ； in, Indicates the first The relative distance between the features of the requested task, Indicates the first The relative distance between the feature of the requested task and the feature of the requested task corresponding to the highest local density, Indicates the first The relative distance between the feature of the requested task and the feature of the requested task corresponding to the lowest local density, and ,and ; The calculation formula of the decision value is as follows: ； in, Represents the decision value of the initial cluster.

4. A method for managing network services in an office area according to claim 1, characterized in that: Encrypting the position of the task cluster center to obtain a first encrypted position includes: The position of the task cluster center is encrypted based on the public key of the management node to obtain an encrypted position, wherein the encryption formula of the position of the task cluster center is as follows: ； in, and represents the longitude and latitude of the encrypted location, represents the encryption function, and express, represents the longitude and latitude of the location of the task cluster center, represents a public key of a management node; and encrypting the encrypted position based on the public key of the edge node to obtain a first encrypted position, wherein the encryption formula of the encrypted position is as follows: ； in, and represents the longitude and latitude of the first encrypted position, Indicates the public key of the edge node.

5. A method for managing network services in an office area according to claim 4, characterized in that: The task cluster center and the first encrypted position are transmitted to the edge node set, and then the method further includes: After receiving the task cluster center and the first encrypted position, the edge node set decrypts the first encrypted position based on the private key of the edge node to obtain an encrypted position, wherein the decryption formula of the first encrypted position is as follows: ； in, represents the decryption function, Represents the private key of the edge node; The edge node broadcasts the public key of the management node and the content of the task cluster center to the task processing nodes within its coverage.

6. A method for managing network services in an office area according to claim 1, characterized in that: Constructing a resource allocation optimization model between the edge node sets, including: Calculate the response time of each edge node in the edge node set, wherein the calculation formula of the response time of the edge node is as follows: ； in, Indicates the response time of the edge node, Indicates the data size corresponding to the requested task. Represents an edge node Request task Send to task processing node The emission rate is , Indicates bandwidth resources. represents Gaussian white noise, Represents an edge node Request task Send to task processing node The transmission power, Represents an edge node and task processing nodes channel gain; calculating the processing time of each edge node in the edge node set, wherein the calculation formula of the processing time of the edge node is as follows: ； in, represents the processing time of the edge node, represents the workload of the requested task, Represents an edge node Assign to request task computing resources.

7. A method for managing network services in an office area according to claim 6, characterized in that: Constructing a resource allocation optimization model between the edge node sets also includes: The total energy consumption of the processing task is calculated based on the response time and processing time of the edge node, wherein the calculation formula of the total energy consumption of the processing task is as follows: ； in, represents the total energy consumption of processing tasks, It represents the average power consumption of edge nodes in processing request tasks; An objective function is constructed based on the total energy consumption of the processing task, a constraint condition is constructed based on the bandwidth resources and computing resources of the edge node, and a resource allocation optimization model is constructed based on the objective function and the constraint condition, wherein the constraint condition is expressed as follows: ； in, represents the total bandwidth of the edge nodes, represents the total computing resources of the edge nodes, Indicates the maximum transmit power.

8. A method for managing network services in an office area according to claim 7, characterized in that: Solving the resource allocation optimization model based on the ant algorithm to obtain the optimal resource allocation between the edge node sets includes: Initializing the parameters of the ant algorithm, wherein the parameters of the ant algorithm include the number of ants, the maximum number of iterations, the initial pheromone, the global attenuation coefficient and the local attenuation coefficient, scattering the ants at the positions of the edge nodes, and setting the pheromone at the positions of the task processing nodes; Calculating the objective function value assigned by each edge node to each task processing node, and calculating the allocation probability of each task processing node based on the objective function value; The task allocation node is determined based on the allocation probability of the task processing node.

9. A method for managing network services in an office area according to claim 8, characterized in that: The calculation formula of the allocation probability is as follows: ； in, represents the probability of allocating edge nodes to task processing nodes, Indicates pheromone, represents heuristic information, represents the relative importance between pheromone and heuristic information, Indicates multiple task processing nodes that can be assigned to edge nodes, represents a subchannel in an edge node, and ; The determination formula of the task allocation node is as follows: ； in, Represents an edge node Determine the assigned task processing node , Indicates that the distribution Random numbers on represents the preset threshold parameter, Represents an edge node The last allocated processing node .

10. A network service management system for an office area, which is applicable to a network service management method for an office area as claimed in any one of claims 1 to 9, characterized in that: include: A task classification unit (1), wherein the task classification unit (1) is used for a preset request node to publish a request task set to a preset management node, wherein the management node clusters the request task set to divide the request task set into different types of sub-request task sets, and obtains the positions of task cluster centers of the different types of sub-request task sets; A position encryption unit (2), the position encryption unit (2) being used to encrypt the position of the task cluster center to obtain a first encrypted position, and transmit the task cluster center and the first encrypted position to the edge node set; A resource allocation unit (3), the resource allocation unit (3) being used to construct a resource allocation optimization model between the edge node sets, and to solve the resource allocation optimization model based on an ant algorithm to obtain an optimal resource allocation between the edge node sets; A task transmission unit (4), the task transmission unit (4) is used to send the task cluster center and the first encrypted position to a preset task processing node based on the edge node set with the best resource allocation, if the task processing node decides to execute the task type corresponding to the task cluster center, then encrypt the position of the task processing node to obtain a second encrypted position, and transmit the second encrypted position to the corresponding edge node; A task selection unit (5), the task selection unit (5) is used for the edge node to calculate the distance between the first encrypted position and the second encrypted position, encrypt the distance to obtain an encrypted distance, transmit the encrypted distance to the management node, and the management node selects a task processing node based on the distance and the reliability of all task processing nodes.

Images