Network slice management method and device, equipment, storage medium and product

Abstract

The invention discloses a network slice management method and device, equipment, a storage medium and a product, and relates to the technical field of network resource management, and the method comprises the steps: setting on-off states corresponding to a preset number of small base stations according to the slice energy consumption of a network slice; dividing the network slice into a plurality of time windows, and determining a slice configuration strategy of each small base station in each time window according to the on-off state; and dividing the time window into a plurality of time slots, and determining a resource allocation strategy corresponding to each time slot to perform slice management. According to the application, the switch is set based on the slice energy consumption, when the load is relatively low, the slices on part of the small base stations are closed to reduce the energy consumption, and the slices on the large base station are kept in the open state, so that the problem of service interruption is avoided; and a slice configuration strategy and a resource allocation strategy are respectively determined based on the long scale of the time window and the short scale of the divided time slot, so that the slice spectrum efficiency is ensured while the network slice energy consumption is reduced, and the long-term performance requirement of the slice is met.

Description

Technical Field

[0001] This application relates to the technical field of network resource management, and particularly to a network slicing management method, device, equipment, storage medium and product. Background Art

[0002] The 6G wireless access network supports multiple applications such as immersive communication, ultra-reliable low-latency communication, and massive communication. For applications such as data rate, latency, reliability, and connection density, different requirements for service quality and user experience are needed. To achieve this goal, network slicing provides flexible network management and supports various services in large-scale heterogeneous wireless networks, showing great application potential in 6G networks.

[0003] Existing solutions can adaptively allocate resources according to current network conditions and user needs by combining resource management and reinforcement learning methods, and have achieved good real-time monitoring and predictive analysis of Radio Access Network Slicing (RANSlicing), promoting the development of intelligent resource management for network slicing and improving network performance and resource utilization efficiency to a certain extent. In practical applications, energy consumption, as a key performance indicator of network operation, has received extensive attention from operators and slice customers, etc. However, most existing methods optimize energy consumption from the perspective of the system or device, rarely paying attention to the operating energy consumption of the slice itself, and ignoring the additional energy consumption brought by deploying a large number of slices in a large-scale network scenario to ensure service isolation and service quality. In addition, the real network environment is often highly complex and requires a hybrid network decision that includes both discrete variables and continuous variables. Most existing solutions are designed only for network decisions of a single variable type. These factors lead to room for optimization in slice energy consumption and hybrid action space algorithms in the existing technology. How to ensure service quality requirements while reducing slice energy consumption remains a key problem that urgently needs to be studied. Summary of the Invention

[0004] The main purpose of this application is to provide a network slicing management method, device, equipment, storage medium and product, aiming to solve the technical problem of how to reduce energy consumption while ensuring slice performance when deploying a large number of network slices.

[0005] To achieve the above purpose, this application proposes a network slicing management method, and the method includes:

[0006] Set the switch states of a preset number of small base stations according to the slice energy consumption of the network slice, where the slice energy consumption is the total energy consumption generated by all devices in the network slice, and the small base station is a base station covered by the network slice with a power less than the power threshold and a coverage range less than the range threshold;

[0007] Divide the network slice into multiple time windows, and determine the slice configuration strategy of each small base station in each time window according to the switch state;

[0008] Divide the time window corresponding to each network slice into multiple time slots, and determine the resource allocation strategy corresponding to each network slice in each time slot;

[0009] Perform slice management according to the slice configuration strategy and the resource allocation strategy.

[0010] In one embodiment, the step of setting the switch states of a preset number of small base stations according to the slice energy consumption of the network slice includes:

[0011] Sum the virtual device energy consumption and the remote radio head energy consumption in the time window of the network slice to obtain the network virtual energy consumption, where the virtual device energy consumption is the energy consumption generated by the virtualization server and the optical devices in the baseband unit pool;

[0012] When the switch state of the network slice in the current time window is switched to the on state, determine the slice restart energy consumption corresponding to restarting the virtual machine of the network slice;

[0013] Sum the network virtual energy consumption and the slice restart energy consumption to obtain the slice energy consumption of the network slice in each time window;

[0014] When the slice energy consumption of the network slice in the time window is lower than the preset energy consumption threshold, set the switch states of a preset number of small base stations to the off state.

[0015] In one embodiment, the step of dividing the network slice into multiple time windows and determining the slice configuration strategy of each small base station in each time window according to the switch state includes:

[0016] Divide the network slice into multiple time windows, and determine the bandwidth reservation ratio according to the total bandwidth of each small base station and the number of network slices with the switch state being on in each small base station;

[0017] Determine the slice configuration strategy of each small base station in each time window according to the bandwidth reservation ratio.

[0018] In one embodiment, the step of determining the slice configuration strategy of each small base station in each time window according to the bandwidth reservation ratio includes:

[0019] Determine the bandwidth reservation decision of each small base station in each time window according to the bandwidth reservation ratio;

[0020] Construct a Markov decision process based on at least one of the number of user equipments, the switch state, the bandwidth reservation decision, the preset slice cost, and the discount factor within the time window corresponding to the network slice, where the discount factor is used to adjust the reward of the Markov decision process;

[0021] Train the Markov decision process, store the historical data generated during the training process in the experience replay area, and randomly sample sample data from the experience replay area to iteratively train the Markov decision process to obtain an optimal policy;

[0022] Map the total number of optional actions in the optimal policy as an input to an output head with a preset number of dimensions, where the preset number of dimensions is the number of action dimensions of the optimal policy;

[0023] Determine the sub-actions of the overall action on each dimension according to each output head;

[0024] Combine the sub-actions into an optimal action, and use the optimal action as the optimal slice configuration strategy for each small base station within each time window.

[0025] In one embodiment, the step of dividing the time window corresponding to each network slice into multiple time slots and determining the resource allocation strategy corresponding to each network slice in each time slot includes:

[0026] Divide the time window corresponding to each network slice into multiple time slots;

[0027] Determine the associated users according to the switch state of each network slice within the time window, where the associated users represent the users whose user equipments are connected to the small base stations;

[0028] Determine the bandwidth allocation capacity of the associated users according to the reserved bandwidth capacity;

[0029] Determine the quality of service for each time slot of each network slice according to the minimum transmission rate;

[0030] Determine the resource allocation strategy corresponding to each network slice in each time slot according to the bandwidth allocation capacity and the quality of service.

[0031] In one embodiment, the step of determining the resource allocation strategy corresponding to each network slice in each time slot according to the bandwidth allocation capacity and the quality of service includes:

[0032] Construct a coalition game model corresponding to each network slice in each time slot according to the user equipment corresponding to the associated users, the small base stations covered by the network slice, the bandwidth allocation capacity, and the quality of service;

[0033] Determine the user association relationship according to the priority order of the user equipment in the coalition game model;

[0034] Perform iterative convex optimization on the user association relationship until Nash equilibrium is reached, and obtain the optimal resource allocation strategy corresponding to each network slice in each time slot.

[0035] In addition, to achieve the above object, the present application also proposes a network slice management device, which includes:

[0036] A slice switch state setting module, configured to set the switch states of a preset number of small base stations according to the slice energy consumption of the network slice, where the slice energy consumption is the total energy consumption generated by all devices in the network slice, and the small base station is a base station covered by the network slice with a power less than the power threshold and a coverage range less than the range threshold;

[0037] A slice configuration strategy determination module, configured to divide the network slice into multiple time windows, and determine the slice configuration strategies of each small base station in each time window according to the switch states;

[0038] A resource allocation strategy determination module, configured to divide the time window corresponding to each network slice into multiple time slots, and determine the resource allocation strategy corresponding to each network slice in each time slot;

[0039] A network slice management module, configured to perform slice management according to the slice configuration strategy and the resource allocation strategy.

[0040] In addition, to achieve the above object, the present application also proposes a network slice management device, which includes: a memory, a processor, and a computer program stored on the memory and executable on the processor, where the computer program is configured to implement the steps of the network slice management method as described above.

[0041] In addition, to achieve the above object, the present application also proposes a storage medium, which is a computer-readable storage medium, and a computer program is stored on the storage medium, and when the computer program is executed by a processor, it implements the steps of the network slice management method as described above.

[0042] In addition, to achieve the above object, the present application also provides a computer program product, which includes a computer program, and when the computer program is executed by a processor, it implements the steps of the network slice management method as described above.

[0043] The present application provides a network slice management method. By setting the switch states of a preset number of small base stations according to the slice energy consumption of the network slice, where the small base stations are the base stations with small power and coverage area covered by the network slice; dividing the network slice into multiple time windows, determining the slice configuration strategy of each small base station in each time window according to the switch states; dividing the time window into multiple time slots, determining the resource allocation strategy corresponding to each time slot; and performing slice management according to the slice configuration strategy and the resource allocation strategy. The present application designs switches based on the slice energy consumption load. When the slice load is low, slices on some small base stations are turned off to reduce energy consumption, while the slices on the large base stations remain on, thus avoiding the service interruption problem caused by turning off the small base station slices; determining the slice configuration strategy and the resource allocation strategy respectively based on the long scale of the time window and the short scale of the divided time slots, and jointly performing slice management on the two time scales, ensuring the slice spectrum efficiency while reducing the energy consumption of the network slice and meeting the long-term performance requirements of the slice. Description of the Drawings

[0044] The drawings herein are incorporated into and constitute a part of this specification, showing embodiments consistent with the present application, and are used together with the specification to explain the principles of the present application.

[0045] To more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, for those of ordinary skill in the art, other drawings can also be obtained based on these drawings without creative efforts.

[0046] Figure 1 It is a schematic flowchart provided for Embodiment 1 of the network slice management method of the present application;

[0047] Figure 2 It is a schematic diagram of the physical network scenario in the network slice management method of the present application;

[0048] Figure 3 It is a dual-time-scale management framework diagram in the network slice management method of the present application;

[0049] Figure 4 It is a schematic flowchart provided for Embodiment 2 of the network slice management method of the present application;

[0050] Figure 5 It is a parameterized deep Q network structure diagram in the network slice management method of the present application;

[0051] Figure 6 It is a schematic flowchart provided for Embodiment 3 of the network slice management method of the present application;

[0052] Figure 7It is a flowchart of a resource allocation method based on coalition game in the network slice management method of this application;

[0053] Figure 8 It is a schematic diagram of the overall process of slice configuration and resource allocation in the network slice management method of this application;

[0054] Figure 9 It is a schematic diagram of the module structure of the network slice management device in the embodiment of this application;

[0055] Figure 10 It is a schematic diagram of the device structure of the hardware operating environment involved in the network slice management method in the embodiment of this application.

[0056] The implementation, functional features and advantages of this application will be further described with reference to the embodiments and the accompanying drawings. Specific Embodiments

[0057] It should be understood that the specific embodiments described herein are only used to explain the technical solutions of this application and are not used to limit this application.

[0058] In order to better understand the technical solutions of this application, the following will be described in detail with reference to the accompanying drawings of the specification and specific embodiments.

[0059] The main solution of the embodiment of this application is: set the switch states of a preset number of small base stations according to the slice energy consumption of the network slice, where the slice energy consumption is the total energy consumption generated by all devices in the network slice, and the small base station is a base station covered by the network slice with a power less than the power threshold and a coverage range less than the range threshold; divide the network slice into multiple time windows, and determine the slice configuration strategy of each small base station in each time window according to the switch states; divide each time window corresponding to each network slice into multiple time slots, and determine the resource allocation strategy of each network slice in each time slot; perform slice management according to the slice configuration strategy and the resource allocation strategy.

[0060] Since the existing solution can adaptively allocate resources by combining resource management and reinforcement learning methods according to the current network conditions and user requirements, it has achieved real-time monitoring and predictive analysis of Radio Access Network Slicing (RANSlicing) well, promoted the development of intelligent resource management for network slicing, and improved network performance and resource utilization efficiency to a certain extent. In practical applications, energy consumption, as a key performance indicator of network operation, has received extensive attention from operators and slice customers, etc. However, most of the existing methods optimize energy consumption from the perspective of the system or device, pay little attention to the operating energy consumption of the slice itself, and ignore the additional energy consumption brought by deploying a large number of slices in a large-scale network scenario to ensure service isolation and quality of service. In addition, the real network environment is often highly complex and requires a hybrid network decision that includes both discrete variables and continuous variables. Most of the existing solutions are designed only for network decisions of a single variable type. These factors lead to room for optimization in slice energy consumption and hybrid action space algorithms in the existing technology. How to ensure the quality of service requirements while reducing slice energy consumption remains a key problem that urgently needs to be studied.

[0061] This application provides a solution. By setting the switch states of a preset number of small base stations according to the slice energy consumption of the network slice, where the small base stations are the base stations with small power and coverage area covered by the network slice; dividing the network slice into multiple time windows, and determining the slice configuration strategy of each small base station in each time window according to the switch states; dividing the time window into multiple time slots, and determining the resource allocation strategy corresponding to each time slot; and performing slice management according to the slice configuration strategy and the resource allocation strategy. This application designs switches based on the slice energy consumption load. When the slice load is low, the slices on some small base stations are turned off to reduce energy consumption, while the slices on the large base stations remain on, thus avoiding the service interruption problem caused by turning off the small base station slices; determining the slice configuration strategy and the resource allocation strategy respectively based on the long scale of the time window and the short scale of the divided time slots, and jointly performing slice management on the two time scales to ensure the slice spectral efficiency while reducing the energy consumption of the network slice and meeting the long-term performance requirements of the slice.

[0062] It should be noted that the execution subject of the method in this embodiment can be a network slice management device with functions of network slice management, network communication, and program operation. The network slice management device is a device connected to the 6G radio access network. This embodiment and the following embodiments will be described by taking the network slice management device as an example.

[0063] Based on this, the embodiments of this application provide a network slice management method, referring to Figure 1 , Figure 1 is the flowchart of the first embodiment of the network slice management method of this application.

[0064] In this embodiment, the network slice management method includes steps S10 to S40:

[0065] Step S10, set the switch states of a preset number of small base stations according to the slice energy consumption of the network slice. The slice energy consumption is the total energy consumption generated by all devices in the network slice. The small base station is a base station covered by the network slice with a power less than the power threshold and a coverage range less than the range threshold.

[0066] It should be noted that the 6G network relies on large base stations to provide wide-area coverage and uses small base stations to achieve local capacity optimization and signal blind spot compensation, forming a "macro - micro - pico - femto" multi-tier heterogeneous network. Among them, large base stations (also known as macro base stations) are traditional base station types with high power (single-carrier transmission power above 10W) and wide coverage (hundreds of meters to several kilometers), suitable for large-scale user access in outdoor open areas (such as suburbs and urban main roads). Small base stations (also known as micro base stations / pico base stations / femto base stations) are base station types with smaller power and narrower coverage ranges. Each network slice can cover multiple small base stations or large base stations simultaneously. A schematic diagram of the physical network scenario can be referred to Figure 2 .

[0067] It can be understood that this embodiment proposes a slice switch scheme based on real-time traffic, which can solve the problem of excessive energy consumption caused by large-scale slice deployment. The basic idea of the slice switch scheme is to adaptively make slice switch decisions in each slice window to adapt to the dynamic changes of the service load and reduce the slice energy consumption. Specifically, in the case of low traffic load, if the slice always remains on the covered base station without user equipment access, it may cause unnecessary energy consumption. Therefore, two switch states are defined for each slice on the small base stations it covers: on and off. The specific decision is based on the slice traffic load (i.e., the slice energy consumption generated by all devices in the network slice), and the switch states of a preset number of network slices on the covered small base stations are switched according to the slice energy consumption of each slice.

[0068] In a feasible implementation manner, step S10 may include steps S101 to S103:

[0069] Step S101, sum the virtual device energy consumption and the remote radio head energy consumption in the time window of the network slice to obtain the network virtual energy consumption. Among them, the virtual device energy consumption is the energy consumption generated by virtualized servers and optical devices in the baseband unit pool.

[0070] It is understandable that in a large-scale sliced network, in order to support diverse emerging services, the deployment of a large number of slices will lead to serious energy consumption problems. From an energy perspective, the energy consumption performance of network slices is a key indicator reflecting their overall operation. System-level energy consumption includes multiple components, and it is impossible to separately evaluate the impact of slices on it. By establishing a network slice energy consumption model, when an operator provides slices to tenants, their performance can be monitored in real time. If the expected requirements cannot be met, the operator can directly adjust specific slices, thus avoiding additional overhead. Therefore, in this embodiment, a network slice energy consumption model is established to characterize the energy required for slice operation in each time window, including network virtual energy consumption and slice restart energy consumption.

[0071] Among them, for network virtual energy consumption, through network virtualization, a Software-Defined Networking (SDN) controller can use virtual resource blocks and separated network functions to achieve flexible network management. The energy consumption of network virtualization is modeled as the sum of the energy consumption of virtualized cloud baseband unit combination (Base Band Units, BBU) servers and the energy consumption of remote radio heads (Remote Radio Heads, RRH).

[0072] It should be noted that the power consumption of the virtualized cloud baseband unit combination server (i.e., the energy consumption of virtual devices) is related to the power consumption of the slice virtualization server, the power consumption of optical devices in the baseband unit pool, and the loss factors approximately such as AC-DC, DC-DC, and cooling losses. The power consumption of the remote radio head is closely related to the power consumption of optical devices, the DC loss factor and MS loss factor of the remote radio head, and the power consumption of radio frequency units, etc.

[0073] It is worth noting that the power consumption of the slice virtualization server is mainly related to the basic power consumption, the number of running virtual servers, and the number of processing bandwidth resources. The power consumption of network virtual machines mainly consists of four modules: Random Access Memory (RAM), Central Processing Unit (CPU), Network Interface Card (NIC), and Hard Drive (HD). Therefore, the power consumption of the slice virtualization server can be expressed as the sum of the running power consumption of the server's random access storage, central processor, network card, and hard drive, the basic power consumption of the server, and the power consumption of processing bandwidth resources.

[0074] Step S102, when the switch state of the network slice in the current time window is switched to the on state, determine the slice restart energy consumption corresponding to restarting the virtual machine of the network slice.

[0075] It should be noted that since the system adjusts the slice switch decision according to the dynamic change of the traffic state at the beginning of each slice window. If the slice is closed on the base station in the previous time window and opened in the current time window, the relevant virtual machine will be restarted, resulting in additional energy consumption, which is the slice restart energy consumption. Here, it is assumed that the time required for the base station to restart the slice is controlled within one time slot.

[0076] Step S103: Sum the network virtual energy consumption and the slice restart energy consumption to obtain the slice energy consumption of the network slice under each time window.

[0077] It can be understood that the slice energy consumption of the slice under the current time window is obtained by summing the network virtual energy consumption and the slice restart energy consumption.

[0078] Step S104: When the slice energy consumption of the network slice in the time window is lower than the preset energy consumption threshold, set the switch states of the corresponding switches of a preset number of small base stations to the off state.

[0079] It should be understood that in a slice window, if the total traffic volume of the slice is low, the slice will choose to turn off on some small base stations to reduce energy consumption; on the contrary, if the total traffic volume of the slice is high, it will choose to turn on on the small base stations to meet the service quality requirements. At the same time, in order to avoid service interruption that may be caused by turning off the slice on the small base station, the proposed slice switch scheme is only applied to the small base stations covered by the slice, while all slices deployed on the large base stations always remain on in each slice window.

[0080] In this embodiment, by summing the network virtual energy consumption and the slice restart energy consumption to determine the slice energy consumption of the network slice under each time window, in a slice window, when the slice energy consumption of the slice is lower than the preset energy consumption threshold, the slice is turned off on some small base stations to reduce energy consumption. The slice is switched on and off in units of slice time windows. When the traffic load is high, in order to avoid affecting the service quality, the slice is turned on on the base stations it covers to ensure that the user equipment can perform task transmission. When the traffic load is low, in order to reduce the energy consumption of slice operation, through intelligent decision-making, the slice is selected to be turned off on some base stations, achieving energy saving while ensuring the service quality. Through the intelligent slice switch scheme, controlling the slice operation according to the actual situation of the network can significantly reduce the slice energy consumption and operation cost.

[0081] Step S20: Divide the network slice into multiple time windows, and determine the slice configuration strategy of each small base station in each time window according to the switch state.

[0082] It can be understood that since the existing solutions are only designed for network decision-making of a single variable type, these factors still leave room for optimization in terms of slice energy consumption and the hybrid action space algorithm. Therefore, in this embodiment, a dual-time-scale management solution is proposed. Due to the interdependence between slice configuration and resource allocation, the dual-time-scale management framework is designed for the goal of efficient slice management. The dual-time-scale management framework diagram can be referred to Figure 3 . By utilizing the characteristics of the proposed management solution, the optimization problem can be naturally decoupled into two sub-problems on different time scales, namely, the long-time-scale slice configuration sub-problem and the short-time-scale resource allocation sub-problem, and then iterative solutions are carried out.

[0083] Specifically, the network slice is divided into multiple time windows as the long time scale. On the long time scale, the slice configuration strategy is determined by the on-off states of each small base station and the bandwidth resources reserved for it by the corresponding base station. The long time scale can be set as the length of the time window for each slice, and only one slice on-off decision and bandwidth reservation decision are made at the beginning of each slice window. The slice configuration strategy remains unchanged within a given time window.

[0084] Step S30: Divide the time window corresponding to each network slice into multiple time slots, and determine the resource allocation strategy corresponding to each network slice in each time slot.

[0085] It should be understood that the time window corresponding to each network slice is divided into multiple time slots as the short time scale. On the short time scale, the time window corresponding to each network slice can be divided or equally divided into a preset number T of time slots. The long time scale decision remains unchanged, and the resource allocation strategy corresponding to each network slice in each time slot is determined according to the user association and bandwidth allocation problems.

[0086] Step S40: Perform slice management according to the slice configuration strategy and the resource allocation strategy.

[0087] This embodiment provides a network slice management method, which sets the switch states of a preset number of small base stations according to the slice energy consumption of the network slice. The small base stations are the base stations covered by the network slice with low power and small coverage range; the network slice is divided into multiple time windows, and the slice configuration strategy of each small base station in each time window is determined according to the switch states; the time window is divided into multiple time slots, and the resource allocation strategy corresponding to each time slot is determined; slice management is performed according to the slice configuration strategy and the resource allocation strategy. In this embodiment, by designing switches based on slice energy consumption load, when the slice load is low, the slices on some small base stations are turned off to reduce energy consumption, while the slices on large base stations remain on, thus avoiding service interruption problems caused by turning off small base station slices; the slice configuration strategy and the resource allocation strategy are determined respectively based on the long scale of the time window and the short scale of the divided time slots, and slice management is jointly performed on the two time scales, ensuring the slice spectrum efficiency while reducing the energy consumption of the network slice and meeting the long-term performance requirements of the slice.

[0088] Based on the first embodiment of this application, in the second embodiment of this application, the same or similar content as in the above-mentioned first embodiment can be referred to the above introduction and will not be repeated hereinafter. On this basis, please refer to Figure 4 , step S20, the network slice management method includes steps S201 to S202:

[0089] Step S201, divide the network slice into multiple time windows, and determine the bandwidth reservation ratio according to the total bandwidth of each small base station and the number of network slices with the switch state being on in each small base station.

[0090] It can be understood that dividing the network slice into multiple time windows, on a long time scale (i.e., within the time window corresponding to the network slice), the slice configuration strategy is completed by determining the switch state of the slice on each small base station and reserving bandwidth resources for each slice, aiming to optimize the slice cost within the slice window, minimizing the energy consumption at the slice level while ensuring the spectrum efficiency. Frequent slice switch operations according to the immediate service demand may cause unnecessary service interruptions and additional restart overheads.

[0091] It is understandable that, in order to adapt to the long-term traffic changes of the network and avoid the additional energy consumption caused by frequent operations, the slice switching decision and bandwidth reservation decision are made only once at the beginning of each slice window. The slice configuration remains unchanged within a given time window. A binary variable is used to represent whether the slice is turned on at the base station, and the bandwidth reservation ratio is represented as a continuous variable between 0 and 1. Considering that the bandwidth reservation decision is affected by the slice switching decision, that is, only when the slice is turned on at the base station, the base station will reserve bandwidth resources for it, otherwise it will not reserve. The slice configuration sub-problem is formulated as minimizing the long-term cost function weighted by slice energy consumption and spectral efficiency. During the optimization process, the following constraints need to be satisfied. First, the slice switching variable is a binary integer variable of 0 and 1, with a value of 1 indicating that the slice is turned on at the base station and a value of 0 indicating that the slice is turned off at the base station. In addition, the bandwidth reservation ratio needs to satisfy relevant constraints. Specifically, the bandwidth reservation ratio cannot exceed the total bandwidth of the base station, and bandwidth resources are reserved only for the slices that are turned on at the base station.

[0092] Step S202: Determine the slice configuration strategy of each small base station within each of the time windows according to the bandwidth reservation ratio.

[0093] It is understandable that after determining the bandwidth reservation ratio, the reserved bandwidth amount of each small base station within each of the time windows can be determined according to this ratio, and the slice configuration strategy can be obtained.

[0094] In a feasible implementation manner, step S202 may include steps S2021 to S2026:

[0095] Step S2021: Determine the bandwidth reservation decision of each small base station within each of the time windows according to the bandwidth reservation ratio.

[0096] It is understandable that in this implementation manner, a management scheme based on a parameterized reinforcement learning architecture and optimization method is proposed. By optimizing the slice configuration and resource allocation on a double time scale, the overall cost of slice operation is reduced. Through the proposed slice configuration and resource allocation algorithm, the iterative optimization of the intelligent slice management strategy is completed. According to the problem characteristics of two different time scales, the slice configuration and resource management scheme can be divided into two parts: the PDQN (Parametrized Deep Q-Network) algorithm and the coalition game algorithm. The former is used to learn the best action as the slice configuration decision on a long time scale, and the structure of the parametrized deep Q network can refer to Figure 5 The latter designs the resource allocation strategy on a short time scale based on coalition game and convex optimization theory to achieve double-layer closed-loop solution.

[0097] It should be understood that, first, the bandwidth reservation decisions of each small base station within each of the time windows are determined according to the bandwidth reservation ratio, that is, the bandwidth reserved for the small base stations covered by each slice.

[0098] Step S2022, construct a Markov decision process according to at least one of the number of user equipments within the time window corresponding to the network slice, the switch state, the bandwidth reservation decision, the preset slice cost, and the discount factor, where the discount factor is used to adjust the reward of the Markov decision process.

[0099] It can be understood that, in order to minimize the long-term network slice cost, the problem can be reformulated as a Markov decision process. The agent collects the current state information and inputs it into the neural network, determines the slice switch and bandwidth reservation actions, obtains the reward after execution in the environment, and transfers to the next state. Here, the state, action, and reward of the Markov decision process are defined as follows: State: The state information includes the number of user equipments of each slice within the current time window and the slice switch decision within the previous time window. Action: In the current time window, the action corresponds to the slice switch decision and the bandwidth reservation decision, that is, the slice switch of each slice on the base stations it covers and the bandwidth reserved for it by the base stations. Reward: The definition of the reward is consistent with the optimization objective and is represented by the preset slice cost of the current time window.

[0100] It should be noted that at the beginning of each time window, the slice configuration strategy is used to guide the agent to determine the slice switch and bandwidth reservation according to the current state information. Let Π represent the set of slice configuration strategies, and the optimization objective can be transformed into determining the optimal slice configuration strategy π * ∈∏ to maximize the cumulative reward. The problem can be reformulated as:

[0101] max π∈∏ [∑ w∈W β w R w (S w ,A w )|π],

[0102] where β ∈ (0,1) represents the discount factor. When the discount factor is close to 1, the weight value of future rewards increases. When it is close to 0, more attention is paid to immediate rewards. When the discount factor is close to 1, this problem can effectively approximate the original problem.

[0103] Step S2023, train the Markov decision process, store the historical data generated during the training process in the experience replay area, and randomly sample the data in the experience replay area to perform iterative training on the Markov decision process to obtain the optimal strategy.

[0104] It should be noted that in order to overcome the instability of network learning during the training process, an experience replay strategy is adopted for training. First, the Markov decision process is trained, and then the historical data during the training process is stored in the experience replay area, rather than only using the current data for training. By randomly extracting a small batch of data from the experience replay area each time, the correlation between samples is reduced, ensuring the global optimality of the strategy. At the same time, a greedy strategy is adopted during the exploration process to ensure that the agent fully explores the environment during the learning process and obtains the optimal strategy.

[0105] Step S2024, map the total number of optional actions in the optimal strategy as input to an output head with a preset number of dimensions, where the preset number of dimensions is the number of action dimensions of the optimal strategy.

[0106] It is worth noting that considering that the slicing switch involves multi-dimensional discrete action decisions, the action dimension is I×J, and the total number of available actions is 2 I×J , as the number of slices and the number of base stations increase, the total number of available actions grows exponentially. The huge action space will affect the training performance of the network. To overcome this problem, the idea of a multi-discrete actor network is adopted to design the network structure. Specifically, the total number of available actions in the optimal strategy is mapped as input to I×J output heads through a shared intermediate layer of the neural network, and the number is equal to the action dimension of the optimal strategy.

[0107] It should be noted that the algorithm based on parameterized reinforcement learning adopts a fully connected neural network architecture, and the computational complexity is mainly related to the dimensions of the input and output layers, the number of hidden layers, the number of neurons in these layers, and the size of the training dataset. Therefore, for a fully connected neural network with a fixed structure of hidden layers and neuron counts, the computational complexity is mainly related to the dimensions of the input and output layers. Based on our previous investigation of the state space and action space, the input dimension and output dimension of the parameter network are I×J + 1 and 2×I×J respectively. Here, the input corresponds to the number of user devices and the slicing switch decision in the previous time window, and the output corresponds to the continuous action parameters, that is, the bandwidth reservation decision corresponding to each discrete switch action taken for the slice. At the same time, the input and output dimensions of the action network are 3×I×J + 1 and 2×I×J respectively. In this case, the input includes the state and continuous action parameters corresponding to each discrete action, and the output is the Q value of each discrete action.

[0108] Step S2025, determine the sub-actions of the overall action in each dimension according to each output head.

[0109] It can be understood that since the switch decisions of each slice on a single base station are independent of each other, each output head can independently determine the sub-actions of the overall action in each dimension, that is, the switch decision of a single slice on a single base station.

[0110] Step S2026, combine the sub-actions into an optimal action, and use the optimal action as the optimal slice configuration strategy for each small base station within each time window.

[0111] It should be understood that all sub-actions are combined into an optimal overall action, and the overall slice switching action is executed in the environment. Through the above design, the action dimension of each output head becomes 1, and the number of actions is reduced to 2, effectively reducing the difficulty of action selection and network training. Use the optimal action as the optimal slice configuration strategy for network slices on each small base station, and the slice configuration strategy will not be updated until the next slice window.

[0112] In this embodiment, since the bandwidth reservation ratio needs to meet relevant constraints, the bandwidth reservation ratio cannot exceed the total bandwidth of the base station, and bandwidth resources are reserved only for the slices in the active state on the base station. Then use the PDQN algorithm to solve the slice configuration sub-problem on the long time scale to minimize the long-term network slice cost, reformulate the problem as a Markov decision process, and introduce a discount factor for training to maximize the cumulative reward. The experience replay strategy is adopted to overcome the instability of network learning during the training process, and the input is mapped to multiple output heads, effectively reducing the difficulty of action selection and network training.

[0113] Based on the first embodiment of the present application, in the third embodiment of the present application, the same or similar content as the above-mentioned embodiment 1 can be referred to the above introduction and will not be repeated hereinafter. On this basis, please refer to Figure 6 Step S30, the network slice management method includes steps S301 to S305:

[0114] Step S301, divide the time window corresponding to each network slice into multiple time slots.

[0115] It can be understood that on the short time scale, each slice window can be divided into T time slots. Considering that in reality, service requests and channel conditions change faster, bandwidth resources are allocated to user equipment within the short time scale. According to the established slice energy consumption model, the energy consumption of the slice is determined based on the slice configuration at the beginning of the time window, and the user association and bandwidth allocation decisions on the short time scale do not affect the calculation of slice energy consumption. Therefore, the overall goal of this sub-problem is to make an optimal resource allocation decision for the transmission tasks of user equipment within each scheduling time slot to maximize the slice spectrum efficiency.

[0116] Step S302, determine the associated users according to the switch state of each network slice within the time window, and the associated users represent the users whose user equipment is connected to the small base station.

[0117] It is understandable that during the optimization process, the following constraint conditions need to be satisfied. First, the user association variable is a binary integer variable of 0 and 1. A value of 1 indicates that the user equipment is associated with the base station, otherwise it indicates that the user equipment is not associated with the base station. The user equipment can only be connected to one base station within the same time slot to access the corresponding slice. Considering that the slice makes on-off decisions on a long time scale, the user equipment can only access the slice through the associated base station when the slice is turned on at the base station.

[0118] Step S303, determine the bandwidth allocation capacity of the associated users according to the reserved bandwidth capacity.

[0119] Step S304, determine the quality of service for each time slot of each network slice according to the minimum transmission rate.

[0120] Step S305, determine the resource allocation strategy corresponding to each network slice in each time slot according to the bandwidth allocation capacity and the quality of service.

[0121] It should be understood that bandwidth allocation needs to meet the resource capacity constraint and the quality of service requirement constraint. The bandwidth allocated to all associated devices cannot exceed the bandwidth capacity reserved by the base station for the slice. The quality of service for each time slot of each network slice is determined according to the minimum transmission rate.

[0122] In a feasible implementation manner, step S305 may include steps S3051 to S3053:

[0123] Step S3051, construct a coalition game model corresponding to each network slice in each time slot according to the user equipment corresponding to the associated users, the small base stations covered by the network slice, the bandwidth allocation capacity, and the quality of service.

[0124] It should be noted that for the resource allocation sub-problem on a short time scale, first, a convex optimization method is used to solve the bandwidth allocation under the given user association relationship to obtain the revenue of the corresponding coalition partition. Then, based on the idea of coalition game, iteration is carried out until the Nash equilibrium is reached to obtain the final user association and resource allocation decisions. The flowchart of the resource allocation method based on coalition game can be referred to Figure 7 .

[0125] It is understandable that the user association problem of the slice can be expressed as a coalition game:

[0126] G i =(U i ,C i ,D i ),

[0127] where U i is the set of user equipment in this game, C iFor the alliance partition, D i is the set of utilities. In the user association problem, each user device accesses the corresponding slice by associating with a base station. After determining the association relationship, the base station allocates the reserved bandwidth resources to the user device. Define the base stations covered by the slice as the minimum unit of the alliance, and the alliance partition is denoted as:

[0128] C i ={C i,1 ,C i,2 ,···,C i,j},

[0129] where C i,j represents the alliance composed of user devices. The alliance needs to satisfy: and C i,j＇ ∪C i,j =U i . The utility of the alliance partition at a certain time slot in the time window can be expressed as

[0130] Step S3052, determine the user association relationship according to the priority order of the user devices in the alliance game model.

[0131] Next, define the priority order of the user devices to determine the user association relationship, expressed as For any user device, when given two alliance partitions C i and C′ i , means that the user device is more inclined to associate with a certain base station to form the partition C i , rather than associating with another base station to form the partition C′ i . The mathematical expression is:

[0132]

[0133] When given the partition C′ i ={C i,1 ,C i,2 ,···,C i,j′}, if and only if the user device chooses to leave the base station alliance C i,j′ it is associated with, and associates with another base station C i,j (j≠j′) to form the partition C i . The mathematical representation of the user device handover operation:

[0134] {C i,j ,C i,j′}\rightarrow{C i,j′ \ {k},C i,j ∪{k}}.

[0135] Step S3053: Perform iterative convex optimization on the user association relationship until Nash equilibrium is reached, and obtain the optimal resource allocation strategy for each network slice corresponding to each time slot.

[0136] It should be understood that after determining the user association relationship, the bandwidth resources reserved by the base station are allocated to the associated user equipment to maximize the utility. Given the slice configuration decision and user association decision, since the objective function is a convex function and the constraints are all linear, the simplified resource allocation sub-problem is a convex problem. The optimal bandwidth allocation decision can be directly obtained using a convex optimization solver. The overall process of the slice configuration and resource allocation method can be referred to Figure 8 as shown.

[0137] It can be understood that the resource allocation algorithm based on coalition game can ensure that in any initial partition case, it converges to the final stable partition after a finite number of iterations. Since the number of user equipment and base stations is finite, the coalitions that user equipment can form are also finite. For any given coalition partition, if the user equipment chooses to switch operations to change its coalition, the total utility of the partition will increase. Then, after a finite number of iterations, the above algorithm will reach Nash equilibrium and finally converge to a stable partition.

[0138] It should be noted that the computational complexity of the resource allocation algorithm based on coalition game mainly depends on the number of iterations. In each iteration, a user equipment and its current coalition are randomly selected, the spectral efficiency of the slice is calculated as the coalition utility, and the preferences of the user equipment for the current and potential coalitions are evaluated. If the user equipment is more willing to associate with a potential coalition, it will be removed from the current coalition and added to the new coalition. The above process is completed by randomly selecting a user equipment in each iteration, and then the next iteration is carried out. Assuming that the total number of iterations of the algorithm is M, the computational complexity of the resource allocation algorithm can be expressed as O(M).

[0139] In this embodiment, the associated users are determined according to the switch state, and the bandwidth allocation that meets the resource capacity constraint and service quality requirement constraint is determined. The overall goal of this resource allocation problem is to make the optimal resource allocation decision for the user equipment transmission tasks in each scheduling time slot and maximize the slice spectral efficiency. For the resource allocation sub-problem on a short time scale, the convex optimization method is used to solve the bandwidth allocation given the user association relationship, and the revenue of the corresponding coalition partition is obtained; then, based on the idea of coalition game, iterative operations are carried out until Nash equilibrium is reached, and the final user association and resource allocation decisions are obtained.

[0140] It should be noted that the above examples are only for understanding this application and do not constitute a limitation on the network slice management method of this application. Based on this technical concept, more forms of simple transformations are within the protection scope of this application.

[0141] The present application also provides a network slice management device. Please refer to Figure 9 , the network slice management device includes:

[0142] A slice switch state setting module 10, configured to set the switch states of a preset number of small base stations according to the slice energy consumption of a network slice, where the slice energy consumption is the total energy consumption generated by all devices in the network slice, and the small base station is a base station covered by the network slice with a power less than a power threshold and a coverage range less than a range threshold;

[0143] A slice configuration policy determination module 20, configured to divide the network slice into multiple time windows, and determine the slice configuration policies of each small base station in each time window according to the switch states;

[0144] A resource allocation policy determination module 30, configured to divide the time window corresponding to each network slice into multiple time slots, and determine the resource allocation policies corresponding to each network slice in each time slot;

[0145] A network slice management module 40, configured to perform slice management according to the slice configuration policy and the resource allocation policy.

[0146] The network slice management device provided by the present application adopts the network slice management method in the above-mentioned embodiment, and can solve the technical problems. Compared with the prior art, the beneficial effects of the network slice management device provided by the present application are the same as those of the network slice management method provided by the above-mentioned embodiment, and other technical features in the network slice management device are the same as the features disclosed in the method of the above-mentioned embodiment, and will not be elaborated here.

[0147] The present application provides a network slice management device. The network slice management device includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein, the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the network slice management method in the first embodiment above.

[0148] Next, refer to Figure 10, which shows a schematic structural diagram of a network slice management device suitable for implementing the embodiments of the present application. The network slice management device in the embodiments of the present application may include, but is not limited to, mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistant), PADs (Portable Application Description: tablet computers), PMPs (Portable Media Player), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 10 The shown network slice management device is merely an example and should not impose any limitation on the functions and scope of use of the embodiments of the present application.

[0149] As Figure 10 shown, the network slice management device may include a processing device 1001 (such as a central processing unit, a graphics processing unit, etc.), which may perform various appropriate actions and processes according to the program stored in the read-only memory 1002 or the program loaded from the storage device 1003 into the random access memory 1004. In the random access memory 1004, various programs and data required for the operation of the network slice management device are also stored. The processing device 1001, the read-only memory 1002, and the random access memory 1004 are connected to each other through a bus 1005. The input / output interface 1006 is also connected to the bus. Generally, the following systems may be connected to the input / output interface 1006: an input device 1007 including, for example, a touch screen, a touchpad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; an output device 1008 including, for example, a liquid crystal display (LCD: Liquid Crystal Display), a speaker, a vibrator, etc.; a storage device 1003 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 1009. The communication device 1009 may allow the network slice management device to communicate with other devices wirelessly or wiredly to exchange data. Although the figure shows a network slice management device having various systems, it should be understood that it is not required to implement or have all the shown systems. More or fewer systems may be alternatively implemented or had.

[0150] In particular, according to the embodiments disclosed in the present application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, the embodiments disclosed in the present application include a computer program product that includes a computer program carried on a computer-readable medium, and the computer program contains program codes for executing the methods shown in the flowcharts. In such an embodiment, the computer program can be downloaded and installed from a network through a communication device, or installed from a storage device 1003, or installed from a ROM 1002. When the computer program is executed by a processing device 1001, the above functions defined in the methods of the embodiments disclosed in the present application are executed.

[0151] The network slice management device provided in the present application adopts the network slice management method in the above embodiment, and can solve the technical problems of network slice management. Compared with the prior art, the beneficial effects of the network slice management device provided in the present application are the same as those of the network slice management method provided in the above embodiment, and other technical features in the network slice management device are the same as the features disclosed in the method of the previous embodiment, which will not be elaborated here.

[0152] It should be understood that each part disclosed in the present application can be implemented by hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in a suitable manner in any one or more embodiments or examples.

[0153] As described above, only the specific embodiments of the present application are provided, but the protection scope of the present application is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed in the present application, and all of them should be covered by the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

[0154] The present application provides a computer-readable storage medium having computer-readable program instructions (i.e., computer programs) stored thereon, and the computer-readable program instructions are used to execute the network slice management method in the above embodiment.

[0155] The computer-readable storage medium provided by the present application may, for example, be a USB flash drive, but is not limited to electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or components, or any combination of the above. More specific examples of the computer-readable storage medium may include, but are not limited to: electrical connections with one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM) or flash memory, optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the above. In this embodiment, the computer-readable storage medium may be any tangible medium that contains or stores a program, and this program can be used by or in combination with an instruction execution system, device, or component. The program code contained on the computer-readable storage medium can be transmitted by any suitable medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination of the above.

[0156] The above computer-readable storage medium may be included in a network slice management device; or it may exist independently and not be assembled into the network slice management device.

[0157] The above computer-readable storage medium carries one or more programs. When the above one or more programs are executed by a network slice management device, the network slice management device is caused to: set the switch states of a preset number of small base stations according to the slice energy consumption of a network slice, where the slice energy consumption is the total energy consumption generated by all devices in the network slice, and the small base stations are base stations covered by the network slice with a power less than a power threshold and a coverage range less than a range threshold; divide the network slice into multiple time windows, and determine the slice configuration strategy of each small base station in each time window according to the switch states; divide each time window corresponding to each network slice into multiple time slots, and determine the resource allocation strategy corresponding to each network slice in each time slot; and perform slice management according to the slice configuration strategy and the resource allocation strategy.

[0158] Computer program code for performing the operations of this application can be written in one or more programming languages or combinations thereof. The above-mentioned programming languages include object-oriented programming languages such as Java, Smalltalk, C++, and also include conventional procedural programming languages such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, executed as a stand-alone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any kind of network, including a local area network (LAN: Local Area Network) or a wide area network (WAN: Wide Area Network), or it can be connected to an external computer (for example, by using an Internet service provider to connect through the Internet).

[0159] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in the flowchart or block diagram may represent a module, a program segment, or a part of code that contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than that marked in the accompanying drawings. For example, two consecutive blocks shown may actually be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagram and / or flowchart, and the combination of blocks in the block diagram and / or flowchart, can be implemented by a dedicated hardware-based system for performing the specified functions or operations, or can be implemented by a combination of dedicated hardware and computer instructions.

[0160] The modules described in the embodiments of this application can be implemented in software or in hardware. Among them, the name of the module does not constitute a limitation to the unit itself in some cases.

[0161] The readable storage medium provided by this application is a computer-readable storage medium. The computer-readable storage medium stores computer-readable program instructions (i.e., computer programs) for performing the above-mentioned network slice management method and can solve technical problems. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided by this application are the same as those of the network slice management method provided by the above embodiments and will not be elaborated here.

[0162] The present application also provides a computer program product, including a computer program which, when executed by a processor, implements the steps of the network slice management method as described above.

[0163] The computer program product provided by the present application can solve technical problems. Compared with the prior art, the beneficial effects of the computer program product provided by the present application are the same as those of the network slice management method provided by the above embodiments, and will not be elaborated herein.

[0164] The foregoing are only partial embodiments of the present application, and thus do not limit the patent scope of the present application. Any equivalent structural transformation made by using the content of the specification and drawings of the present application under the technical concept of the present application, or any direct / indirect application in other related technical fields shall be included within the patent protection scope of the present application.

Claims

1. A network slice management method, characterized in that The method described includes: Setting the switch states corresponding to a preset number of small base stations according to the slice energy consumption of a network slice, where the slice energy consumption is the total energy consumption generated by all devices in the network slice, and the small base stations are base stations covered by the network slice with a power less than a power threshold and a coverage range less than a range threshold; Dividing the network slice into multiple time windows, and determining the slice configuration strategy of each small base station in each time window according to the switch states; Dividing the time window corresponding to each network slice into multiple time slots, and determining the resource allocation strategy corresponding to each network slice in each time slot; Performing slice management according to the slice configuration strategy and the resource allocation strategy.

2. The method according to claim 1, characterized in that, The step of setting the switch states corresponding to a preset number of small base stations according to the slice energy consumption of the network slice includes: Summing the virtual device energy consumption and the remote radio head energy consumption in the time window of the network slice to obtain the network virtual energy consumption, where the virtual device energy consumption is the energy consumption generated by the virtualization server and the optical devices in the baseband unit pool; When the switch state of the network slice in the current time window is switched to the on state, determining the slice restart energy consumption for restarting the virtual machine corresponding to the network slice; Summing the network virtual energy consumption and the slice restart energy consumption to obtain the slice energy consumption of the network slice in each time window; When the slice energy consumption of the network slice in the time window is lower than a preset energy consumption threshold, setting the switch states corresponding to a preset number of small base stations to the off state.

3. The method according to claim 1, characterized in that, The step of dividing the network slice into multiple time windows and determining the slice configuration strategy of each small base station in each time window according to the switch states includes: Dividing the network slice into multiple time windows, and determining the bandwidth reservation ratio according to the total bandwidth of each small base station and the number of network slices with the switch state being on in each small base station; Determining the slice configuration strategy of each small base station in each time window according to the bandwidth reservation ratio.

4. The method according to claim 3, characterized in that, The step of determining the slice configuration strategy of each small base station in each time window according to the bandwidth reservation ratio includes: Determining the bandwidth reservation decision of each small base station in each time window according to the bandwidth reservation ratio; Constructing a Markov decision process according to at least one of the number of user equipment, the switch state, the bandwidth reservation decision, a preset slice cost, and a discount factor in the time window corresponding to the network slice, where the discount factor is used to adjust the reward of the Markov decision process; Training the Markov decision process, storing the historical data generated during the training process in an experience replay area, and randomly sampling sample data from the experience replay area to iteratively train the Markov decision process to obtain an optimal strategy; Mapping the total number of optional actions in the optimal strategy as an input to an output head with a preset dimension number, where the preset dimension number is the number of action dimensions of the optimal strategy; Determining the sub-actions of the overall action in each dimension according to each output head; Combine the sub-actions into an optimal action, and use the optimal action as the optimal slice configuration strategy for each small cell within each of the time windows.

5. The method according to claim 1, characterized in that, The step of dividing the time window corresponding to each network slice into multiple time slots and determining the resource allocation strategy corresponding to each network slice in each time slot includes: Divide the time window corresponding to each network slice into multiple time slots; Determine the associated users according to the on-off state of each network slice within the time window, where the associated users represent the users whose user equipment is connected to the small cells; Determine the bandwidth allocation capacity of the associated users according to the reserved bandwidth capacity; Determine the quality of service for each time slot of each network slice according to the minimum transmission rate; Determine the resource allocation strategy corresponding to each network slice in each time slot according to the bandwidth allocation capacity and the quality of service.

6. The method according to claim 5, wherein The step of determining the resource allocation strategy corresponding to each network slice in each time slot according to the bandwidth allocation capacity and the quality of service includes: Construct a coalition game model corresponding to each network slice in each time slot according to the user equipment corresponding to the associated users, the small cells covered by the network slice, the bandwidth allocation capacity, and the quality of service; Determine the user association relationship according to the priority order of the user equipment in the coalition game model; Perform iterative convex optimization on the user association relationship until reaching the Nash equilibrium, and obtain the optimal resource allocation strategy corresponding to each network slice in each time slot.

7. A network slice management device, characterized in that, The network slice management device includes: A slice on-off state setting module, configured to set the on-off state of a preset number of small cells according to the slice energy consumption of the network slice, where the slice energy consumption is the total energy consumption generated by all devices in the network slice, and the small cell is a base station covered by the network slice with a power less than a power threshold and a coverage range less than a range threshold; A slice configuration strategy determination module, configured to divide the network slice into multiple time windows and determine the slice configuration strategy of each small cell within each of the time windows according to the on-off state; A resource allocation strategy determination module, configured to divide the time window corresponding to each network slice into multiple time slots and determine the resource allocation strategy corresponding to each network slice in each time slot; A network slice management module, configured to perform slice management according to the slice configuration strategy and the resource allocation strategy.

8. A network slice management device, characterized in that, The device includes: a memory, a processor, and a computer program stored on the memory and executable on the processor, where the computer program is configured to implement the steps of the network slice management method according to any one of claims 1 to 6.

9. A storage medium, characterized in that, The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium, and when the computer program is executed by the processor, it implements the steps of the network slice management method according to any one of claims 1 to 6.

10. A computer program product, characterized in that, The computer program product includes a computer program, and when the computer program is executed by the processor, it implements the steps of the network slice management method according to any one of claims 1 to 6.

Images