5G network slice pre-arrangement method, device, equipment, medium and product

Abstract

The invention discloses a 5G network slice pre-arrangement method and device, equipment, a medium and a product, and the method comprises the steps: determining a predicted access base station track and a target service request type of each user terminal in a pre-arrangement time period based on the historical access base station track information of each user terminal; and performing network slice arrangement based on the predicted access base station trajectory and the target service request type to obtain at least one pre-arrangement network slice in the pre-arrangement time period. According to the invention, the preparation work of the 5G network slice can be completed in advance, the configuration efficiency of the 5G network slice and the adaptability under high load or burst flow are improved, the 5G network slice can be rapidly activated when the user demand changes, and different network resource configurations can be rapidly switched.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of wireless communication, in particular to a pre-orchestration method and device of 5G network slice, electronic equipment, computer readable storage medium and computer program product. BACKGROUND

[0002] Network slice is a technology of building multiple logically isolated end-to-end virtual networks on a shared network infrastructure. The current 5G network slice technology is mainly based on network function virtualization (NFV) and software defined network (SDN) technology, which divides the physical network into multiple independent virtual slices. Each network slice can be independently configured with bandwidth, delay and security according to specific quality of service requirements to support different business needs such as industrial internet, intelligent transportation and augmented reality, and provide customized network services.

[0003] In the prior art, the orchestration of 5G network slice usually relies on real-time monitoring data, which leads to the inability to adjust network slice strategy in time when user demand changes, and there is a lag phenomenon. When the network state changes, the network slice reacts slowly, thereby affecting the adaptability of the network slice under high load or burst traffic, and different network resource configurations cannot be quickly switched. SUMMARY

[0004] The present application provides a pre-orchestration method, device, equipment, medium and product of 5G network slice, which predicts the access base station trajectory and service request type of user terminals in the pre-orchestration time period based on the historical access base station trajectory information of user terminals, and performs network slice orchestration based on this, so as to obtain at least one pre-orchestration network slice in the pre-orchestration time period, thereby completing the preparation work of 5G network slice in advance, improving the configuration efficiency of 5G network slice and the adaptability under high load or burst traffic, and ensuring that the 5G network slice can be quickly activated when the user demand changes, and different network resource configurations can be quickly switched.

[0005] To solve the above technical problems, the first aspect of the embodiment of the present application provides a pre-orchestration method of 5G network slice, comprising: Based on the historical access base station trajectory information of each user terminal, the predicted access base station trajectory and the target service request type of each user terminal in the pre-orchestration time period are determined; Based on the predicted access base station trajectory and the target service request type, network slice orchestration is performed to obtain at least one pre-orchestration network slice in the pre-orchestration time period.

[0006] As a preferred solution, the historical access base station trajectory information of each user terminal is used to determine the predicted access base station trajectory and the target service request type of each user terminal in the pre-arrangement time period, specifically including: According to a preset prediction time length, the pre-arrangement time period is time-divided to obtain at least one prediction time period; Based on the start time and end time of each prediction time period, an initial prediction access trajectory template corresponding to each prediction time period is generated; The historical access base station trajectory information and the initial prediction access trajectory template are input into a transformer structure-based access trajectory prediction model for access trajectory prediction to obtain the predicted access trajectory information of each user terminal in each prediction time period; Based on each predicted access trajectory information, the predicted access base station trajectory and the target service request type of each user terminal in the pre-arrangement time period are determined.

[0007] As a preferred solution, the historical access base station trajectory information of each user terminal is used to determine the predicted access base station trajectory and the target service request type of each user terminal in the pre-arrangement time period, specifically including: Based on each predicted access trajectory information, the predicted access base station trajectory and the target service request type of each user terminal in the pre-arrangement time period are determined. When it is detected that the user terminal has at least two target prediction access trajectory information that are continuous in the prediction time period and consistent in the predicted access base station identifier, the predicted access application IP, and the target transmission channel slice type, each target prediction access trajectory information is merged to obtain prediction access trajectory merging information; Based on the predicted access base station identifier in the prediction access trajectory merging information and the predicted access application IP and / or the predicted access base station identifier and the predicted access application IP in the prediction access trajectory information that does not participate in merging, the predicted access base station trajectory and the target service request type are determined respectively.

[0008] As a preferred solution, the historical access base station trajectory information at least includes historical access start time, historical access end time, historical access base station identifier, historical access application IP, and historical transmission channel slice type.

[0009] As a preferred solution, the network slice pre-arrangement is performed based on the predicted access base station trajectory and the target service request type to obtain at least one pre-arranged network slice in the pre-arrangement time period, specifically including: determining at least one target service request link of each of the user terminals based on the predicted access base station trajectory and the target service request type; performing performance evaluation on the target service request link based on the link resource of the target service request link and the target service request type corresponding thereto, to obtain a performance score of each of the target service request link; constructing a network slice based on each of the target service request link with the performance score greater than the average performance score, to obtain at least one pre-arranged network slice in the pre-arranged time period; wherein the average performance score is the average of the performance scores of all the target service request links in the pre-arranged time period.

[0010] As a preferred solution, the method further comprises: obtaining slice link information and user behavior aggregation degree corresponding to each of the pre-arranged network slices; wherein the user behavior aggregation degree is determined based on the number of terminals with terminal access behavior information consistent with the pre-arranged network slice selected from each of the historical network slices; the terminal access behavior information includes access time, access base station identifier, access application IP and transmission channel slice type; adjusting each of the pre-arranged network slices according to the slice link information, the user behavior aggregation degree and the current pre-arrangement accuracy; wherein the pre-arrangement accuracy is determined based on the pre-arranged network slice update in a preset detection time period before the current time.

[0011] As a preferred solution, the adjusting each of the pre-arranged network slices according to the slice link information, the user behavior aggregation degree and the current pre-arrangement accuracy specifically comprises: calculating the accuracy of the pre-arranged network slice according to the slice link information and the user behavior aggregation degree; deleting the pre-arranged network slice with the accuracy less than the pre-arrangement accuracy.

[0012] As a preferred solution, the method specifically determines the current pre-arrangement accuracy by the following steps: obtaining the pre-arranged network slice update in the preset detection time period from the 5G base station controller; determining the number of pre-arranged network slice updates corresponding to each of the pre-arranged time periods in the preset detection time period based on the pre-arranged network slice update; determining the current pre-arrangement accuracy based on the preset weight coefficient, the number of pre-arranged network slices and the number of pre-arranged network slice updates corresponding to each of the pre-arranged time periods in the preset detection time period.

[0013] As a preferred solution, the slice link information comprises channel quality, user experience score and slice link complexity; wherein the user experience score is determined based on the ratio between the user complaint rate of the historical network slice consistent with the terminal access behavior information corresponding to the pre-arranged network slice and the average user complaint rate of each historical network slice; and the slice link complexity is obtained based on the product between the slice link length and the non-link transmission algorithm power consumption.

[0014] As a preferred solution, the method further comprises: sending each of the pre-arranged network slices to a 5G base station controller; distributing each of the pre-arranged network slices to each of the 5G base stations through the 5G base station controller according to the floating resource shortage degree of each 5G base station within the management range of the 5G base station controller and the preset resource support force saturation degree; wherein the floating resource shortage degree is determined based on the running state information of the 5G base station after executing the pre-arranged network slice.

[0015] As a preferred solution, the distributing each of the pre-arranged network slices to each of the 5G base stations through the 5G base station controller according to the floating resource shortage degree of each 5G base station within the management range of the 5G base station controller and the preset resource support force saturation degree, specifically comprises: evaluating the floating resource shortage degree of each of the 5G base stations after executing the pre-arranged network slice through the 5G base station controller based on the current each of the pre-arranged network slices; when the floating resource shortage degree of any one 5G base station is greater than or equal to the resource support force saturation degree, deleting the pre-arranged network slice with the lowest performance score among each of the pre-arranged network slices required to be executed by the any one 5G base station through the 5G base station controller; and repeating the above steps until the floating resource shortage degree of each of the 5G base stations is less than the resource support force saturation degree; obtaining the target network slice of each of the 5G base stations through the 5G base station controller based on the predicted access base station identifier corresponding to the current each of the pre-arranged network slices, and distributing each of the target network slices to each of the 5G base stations to instruct the 5G base station to configure the target network slice, and activate the target network slice for data transmission when detecting that the terminal corresponding to the target network slice requests.

[0016] As a preferred solution, the running state information comprises hardware resource usage rate, hardware resource maximum usage rate and hardware resource minimum usage rate.

[0017] The second aspect of the embodiment of the present application provides a 5G network slice pre-arrangement device, comprising: a terminal access prediction module configured to determine predicted access base station trajectories and target service request types of each user terminal in a pre-arrangement time period based on historical access base station trajectory information of each user terminal; a network slice pre-arrangement module configured to perform network slice arrangement based on the predicted access base station trajectories and the target service request types, and obtain at least one pre-arranged network slice in the pre-arrangement time period.

[0018] As a preferred solution, the terminal access prediction module is configured to determine predicted access base station trajectories and target service request types of each user terminal in a pre-arrangement time period based on historical access base station trajectory information of each user terminal, and specifically comprises: dividing the pre-arrangement time period into at least one prediction time period according to a preset prediction time length; generating an initial prediction access trajectory template corresponding to each prediction time period based on the start time and end time of each prediction time period; inputting the historical access base station trajectory information and the initial prediction access trajectory template into an access trajectory prediction model based on a transformer structure to perform access trajectory prediction, and obtaining predicted access trajectory information of each user terminal in each prediction time period; determining the predicted access base station trajectories and the target service request types of each user terminal in the pre-arrangement time period based on each predicted access trajectory information.

[0019] As a preferred solution, the terminal access prediction module is configured to determine predicted access base station trajectories and target service request types of each user terminal in the pre-arrangement time period based on each predicted access trajectory information, and specifically comprises: determining predicted access base station identifiers, predicted access application IPs and target transmission channel slice types of each user terminal in each prediction time period based on each predicted access trajectory information; when it is detected that the user terminal has at least two target predicted access trajectory information that are continuous in the prediction time period and consistent in the predicted access base station identifier, the predicted access application IP and the target transmission channel slice type, merging each target predicted access trajectory information to obtain predicted access trajectory merging information; determine the predicted access base station trajectory and the target service request type based on the predicted access base station identifier and the predicted access application IP in the predicted access trajectory merging information, and the predicted access base station identifier and the predicted access application IP in the predicted access trajectory information that is not involved in the merging.

[0020] As a preferred solution, the historical access base station trajectory information at least includes historical access start time, historical access end time, historical access base station identifier, historical access application IP and historical transmission channel slice type.

[0021] As a preferred solution, the network slice pre-arrangement module is configured to perform network slice arrangement based on the predicted access base station trajectory and the target service request type, to obtain at least one pre-arranged network slice within the pre-arrangement time period, specifically including: determine at least one target service request link for each of the user terminals based on the predicted access base station trajectory and the target service request type; perform performance evaluation on the target service request link based on the link resource of the target service request link and the target service request type corresponding thereto, to obtain a performance score of each of the target service request links; perform network slice construction based on each of the target service request links with a performance score greater than a performance score average, to obtain at least one pre-arranged network slice within the pre-arrangement time period; wherein the performance score average is an average of performance scores of all the target service request links within the pre-arrangement time period.

[0022] As a preferred solution, the apparatus is further configured to: obtain slice link information and user behavior aggregation degree corresponding to each of the pre-arranged network slices; wherein the user behavior aggregation degree is determined based on a number of terminals with terminal access behavior information consistent with the pre-arranged network slices, which is filtered from each of the historical network slices; the terminal access behavior information includes access time, access base station identifier, access application IP and transmission channel slice type; adjust each of the pre-arranged network slices based on the slice link information, the user behavior aggregation degree and current pre-arrangement accuracy; wherein the pre-arrangement accuracy is determined based on pre-arranged network slice update within a preset detection time period before the current time.

[0023] As a preferred solution, the apparatus is configured to adjust each of the pre-arranged network slices based on the slice link information, the user behavior aggregation degree and current pre-arrangement accuracy, specifically including: According to the slice link information and the user behavior aggregation degree, the accuracy of the pre-arranged network slice is calculated; The pre-arranged network slice with an accuracy less than the pre-arranged accuracy is deleted.

[0024] As a preferred solution, the device specifically determines the current pre-arranged accuracy by the following steps: Obtain the pre-arranged network slice update situation in the preset detection time period from the 5G base station controller; Based on the pre-arranged network slice update situation, determine the pre-arranged network slice update quantity corresponding to each pre-arranged time period in the preset detection time period; Based on the preset weight coefficient, the pre-arranged network slice quantity and the pre-arranged network slice update quantity corresponding to each pre-arranged time period in the preset detection time period, determine the current pre-arranged accuracy.

[0025] As a preferred solution, the slice link information includes channel quality, user experience score and slice link complexity; wherein the user experience score is determined based on the ratio between the user complaint rate of the historical network slice consistent with the terminal access behavior information corresponding to the pre-arranged network slice and the average user complaint rate of each historical network slice; the slice link complexity is obtained based on the product between the slice link length and the non-link transmission algorithm power consumption.

[0026] As a preferred solution, the device is further used for: Send each pre-arranged network slice to the 5G base station controller; Through the 5G base station controller, distribute each pre-arranged network slice to each 5G base station according to the floating resource shortage degree of each 5G base station within the management range of the 5G base station controller and the preset resource support force degree saturation; wherein the floating resource shortage degree is determined based on the running state information of the 5G base station after executing the pre-arranged network slice.

[0027] As a preferred solution, the device is used to distribute each pre-arranged network slice to each 5G base station through the 5G base station controller according to the floating resource shortage degree of each 5G base station within the management range of the 5G base station controller and the preset resource support force degree saturation, specifically including: Based on the current each pre-arranged network slice, evaluate the floating resource shortage degree of each 5G base station after executing the pre-arranged network slice through the 5G base station controller; When the floating resource scarcity of any 5G base station is greater than or equal to the resource support saturation, the 5G base station controller deletes the pre-arranged network slice with the lowest performance score among the pre-arranged network slices to be executed by any 5G base station; repeat the above steps until the floating resource scarcity of each 5G base station is less than the resource support saturation. Based on the predicted access base station identifiers corresponding to each of the current pre-arranged network slices, the target network slices of each of the 5G base stations are obtained through the 5G base station controller, and each of the target network slices is distributed to each of the 5G base stations to instruct the 5G base stations to configure the target network slices, and when a terminal request corresponding to the target network slice is detected, the target network slice is activated to perform data transmission.

[0028] As a preferred embodiment, the operating status information includes hardware resource utilization rate, maximum hardware resource utilization rate, and minimum hardware resource utilization rate.

[0029] A third aspect of the present invention provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the 5G network slicing pre-arrangement method described in any one of the first aspects.

[0030] A fourth aspect of the present invention provides a computer-readable storage medium comprising a stored computer program, wherein, when the computer program is executed, it controls the device where the computer-readable storage medium is located to perform the 5G network slicing pre-arrangement method described in any one of the first aspects.

[0031] A fifth aspect of the present invention provides a computer program product, including a computer program / instructions, wherein when the computer program / instructions are executed by a processor, the steps of the 5G network slicing pre-arrangement method described in any one of the first aspects are implemented.

[0032] Compared with the prior art, the beneficial effects of the embodiments of the present invention are that, by predicting the access base station trajectory and service request type of the user terminal within a pre-arranged time period based on the historical access base station trajectory information of the user terminal, and performing network slicing in this way, at least one pre-arranged network slice can be obtained within the pre-arranged time period. This allows for the early completion of 5G network slice preparation, improves the configuration efficiency of 5G network slices and their adaptability under high load or burst traffic, and ensures that 5G network slices can be quickly activated when user needs change, enabling rapid switching between different network resource configurations. Attached Figure Description

[0033] Figure 1is a flowchart of a 5G network slice pre-arrangement method in an embodiment of the present application; Figure 2 is a structural diagram of a 5G network slice pre-arrangement device in an embodiment of the present application; Figure 3 is a structural diagram of an electronic device in an embodiment of the present application. DETAILED DESCRIPTION

[0034] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative labor fall within the scope of the present application.

[0035] Please refer to Figure 1 The first aspect of the embodiments of the present application provides a 5G network slice pre-arrangement method, comprising the following steps S1 and S2: Step S1, based on the historical access base station trajectory information of each user terminal, determining the predicted access base station trajectory and the target service request type of each user terminal in a pre-arrangement time period.

[0036] Specifically, since the historical access base station trajectory information of each user terminal can reflect the access behavior habits of each user terminal, thereby providing a reference basis for the decision of the access base station trajectory and the service request type of different user terminals in the future pre-arrangement time period, therefore, based on the historical access base station trajectory information, the predicted access base station trajectory and the target service request type of each user terminal in the pre-arrangement time period are inferred to determine the access behavior of each user terminal in the pre-arrangement time period. It can be understood that the predicted access base station trajectory is the access order of each access base station from small to large according to time inferred in the pre-arrangement time period, and the target service request type is the application type to which the target access application IP corresponding to each access base station in the pre-arrangement time period belongs.

[0037] Step S2, based on the predicted access base station trajectory and the target service request type, performing network slice arrangement to obtain at least one pre-arrangement network slice in the pre-arrangement time period.

[0038] Specifically, after the predicted access base station trajectory and the target service request type of each user terminal in the pre-arrangement time period are determined, each service request link of each user terminal in the pre-arrangement time period can be determined, a network slice is constructed by the service request link to realize network slice arrangement, at least one pre-arranged network slice in the pre-arrangement time period is obtained, and the network slice strategy is predicted in advance, network delay is effectively reduced, and user experience is optimized.

[0039] The pre-arrangement method of the 5G network slice provided by the embodiment of the application can predict the access base station trajectory and the service request type of the user terminal in the pre-arrangement time period based on the historical access base station trajectory information of the user terminal, and arrange the network slice based on the same, so that at least one pre-arranged network slice in the pre-arrangement time period can be obtained, and the preparation work of the 5G network slice can be completed in advance, the configuration efficiency of the 5G network slice and the adaptability under high load or burst traffic are improved, and the 5G network slice can be quickly activated when the user demand changes, and different network resource configurations can be quickly switched.

[0040] As a preferred solution, the historical access base station trajectory information of each user terminal is used to determine the predicted access base station trajectory and the target service request type of each user terminal in the pre-arrangement time period, and specifically includes: The pre-arrangement time period is time-divided according to a preset prediction time length, and at least one prediction time period is obtained; Based on the start time and the end time of each prediction time period, an initial prediction access trajectory template corresponding to each prediction time period is generated; The historical access base station trajectory information and the initial prediction access trajectory template are input into an access trajectory prediction model based on a transformer structure to perform access trajectory prediction, and prediction access trajectory information of each user terminal in each prediction time period is obtained; Based on each prediction access trajectory information, the predicted access base station trajectory and the target service request type of each user terminal in the pre-arrangement time period are determined.

[0041] Specifically, the embodiment pre-sets a prediction time length as a minimum prediction time unit of access trajectory prediction, and then generates an initial prediction access trajectory template corresponding to each prediction time period based on the start time and the end time of each prediction time period obtained by division, wherein the initial prediction access trajectory template records the start time, the end time, the access application IP (Internet Protocol Address), the transmission channel slice type used at the time of access, and the access base station identifier of the prediction time period, wherein the access application IP, the transmission channel slice type used at the time of access, and the access base station identifier are all empty data, and are used to be filled in completely after the corresponding prediction result is obtained. For example, assuming that the jthinitial prediction access trajectory template is For wherein, represents the start time of the jthprediction time period, is a pre-arranged time period, is a prediction time length; represents the end time of the jthprediction time period; is an access application IP; is a transmission channel slice type used at the time of access; is an access base station identifier.

[0042] Further, the embodiment adopts an artificial intelligence model based on a transformer structure as an access trajectory prediction model, which can capture the long-term dependence relationship of the access behavior of the user terminal in the time dimension by using the self-attention mechanism of the transformer, and can simultaneously model the complex space-time association between the access application IP, the transmission channel slice type, and the access base station identifier. Therefore, the historical access base station trajectory information and each initial prediction access trajectory template are input into the access trajectory prediction model for access trajectory prediction, so as to predict the access application IP, the transmission channel slice type used at the time of access, and the access base station identifier in each initial prediction access trajectory template, and determine the prediction access trajectory information of each user terminal in each prediction time period. Based on the predicted access base station identifier, the predicted access base station trajectory of the user terminal in the pre-arranged time period can be determined, and based on the predicted access application IP, the target service request type of the user terminal in the pre-arranged time period can be determined.

[0043] The embodiment of the application can accurately capture the long-term dependence relationship of the access behavior of the user terminal in the time dimension by adopting the access trajectory prediction model based on the transformer structure for access trajectory prediction, so as to ensure the accuracy of the obtained prediction access trajectory information and the pre-arrangement accuracy of the 5G network slice.

[0044] As a preferred solution, the determining the predicted access base station trajectory and the target service request type of each user terminal in the pre-arranged time period based on each predicted access trajectory information specifically comprises: determining the predicted access base station identifier, the predicted access application IP and the target transmission channel slice type of the user terminal in each predicted time period based on each predicted access trajectory information; when detecting that the user terminal has at least two target predicted access trajectory information which are continuous in the predicted time period and consistent in the predicted access base station identifier, the predicted access application IP and the target transmission channel slice type, merging each target predicted access trajectory information to obtain predicted access trajectory merging information; determining the predicted access base station trajectory and the target service request type based on the predicted access base station identifier and the predicted access application IP in the predicted access trajectory merging information and / or the predicted access base station identifier and the predicted access application IP in the predicted access trajectory information which is not involved in the merging, respectively.

[0045] Specifically, after obtaining the predicted access trajectory information, the embodiment can directly and explicitly determine the predicted access base station identifier, the predicted access application IP and the target transmission channel slice type of the user terminal in each predicted time period. If the user terminal has at least two target predicted access trajectory information which are continuous in the predicted time period, i.e., the end time of one of the two predicted time periods corresponding to the adjacent two predicted access trajectory information is continuous with the start time of the other predicted time period, and the predicted access base station identifier, the predicted access application IP and the target transmission channel slice type are consistent, it indicates that these target predicted access trajectory information represent the same terminal access behavior, which needs to be merged to make the terminal access behavior and the predicted access trajectory information complete.

[0046] Further, if all the current predicted access trajectory information is merged into the same information, the predicted access base station trajectory and the target service request type are directly determined based on the predicted access base station identifier and the predicted access application IP in the predicted access trajectory merging information, respectively. If all the current predicted access trajectory information is not involved in the merging, the predicted access base station trajectory and the target service request type are determined based on the predicted access base station identifier and the predicted access application IP in each predicted access trajectory information, respectively. If there are both part of the predicted access trajectory information involved in the merging and part of the predicted access trajectory information not involved in the merging, the predicted access base station trajectory and the target service request type are determined based on the predicted access base station identifier and the predicted access application IP in the predicted access trajectory merging information and the predicted access trajectory information not involved in the merging, respectively.

[0047] The embodiment of the present application can ensure the integrity of the terminal access behavior and the predicted access trajectory information, and ensure the pre-arrangement accuracy of the 5G network slice, by merging each target predicted access trajectory information when it is detected that the user terminal exists for a predicted time period, and the predicted access base station identifier, the predicted access application IP and the target transmission channel slice type of at least two target predicted access trajectory information are consistent.

[0048] As a preferred solution, the historical access base station trajectory information at least includes a historical access start time, a historical access end time, a historical access base station identifier, a historical access application IP and a historical transmission channel slice type.

[0049] Specifically, the historical access base station trajectory information in the embodiment may be expressed as: , wherein, is the historical access start time of the jth historical access base station trajectory information; is the historical access end time; is the historical access application IP; is the historical transmission channel slice type; is the historical access base station identifier, and through the content recorded in the historical access base station trajectory information, the access behavior habit of each user terminal can be accurately reflected, so that the accuracy of the prediction of the access base station trajectory and the service request type of the user terminal in the pre-arrangement time period can be improved, and the pre-arrangement accuracy of the 5G network slice is ensured.

[0050] As a preferred solution, the network slice arrangement based on the predicted access base station trajectory and the target service request type obtains at least one pre-arranged network slice in the pre-arrangement time period, and specifically includes: determining at least one target service request link of each user terminal based on the predicted access base station trajectory and the target service request type; performing performance evaluation on the target service request link according to the link resource of the target service request link and the target service request type corresponding thereto, to obtain a performance score of each target service request link; constructing a network slice based on each target service request link with a performance score greater than the average performance score, to obtain at least one pre-arranged network slice in the pre-arrangement time period; wherein the average performance score is the average value of the performance scores of all target service request links in the pre-arrangement time period.

[0051] Specifically, for each predicted access base station in the predicted access base station trajectory and the target service request type corresponding thereto, at least one target service request link of each user terminal can be determined based on a target access application IP corresponding to the target service request type.

[0052] Further, the embodiment obtains link resources of the target service request link, including computing resources, bandwidth, etc., and performs performance evaluation on each target service request link in combination with the target service request type. It can be understood that for each item of link resource, a corresponding performance score can be set, and the target service request type is used to determine the value of the business scenario, for example, the scenario priority of service request types such as emergency communication, automatic driving and industrial control is high, so the scenario value is high, and the corresponding performance score can be set to be high, while the scenario priority of service request types such as mobile short video and web browsing is low, so the scenario value is low, and the corresponding performance score can be set to be low. The embodiment is not specifically limited herein. When performing performance evaluation, the performance score of each target service request link can be obtained by weighted summation of each item of performance score.

[0053] Further, for the target service request links of all user terminals in the current pre-arrangement time period, high-value links are selected based on the screening standard that the performance score is greater than the average performance score, so as to avoid low-value network slices from occupying the limited computing resources and bandwidth of the 5G base station, thereby realizing dynamic planning of network slices for different time periods and improving the utilization rate of pre-slice resources.

[0054] As a preferred solution, the method further comprises: obtaining slice link information corresponding to each pre-arranged network slice and user behavior aggregation degree; wherein the user behavior aggregation degree is determined based on the number of terminals whose access behavior information is consistent with the terminal access behavior information corresponding to the pre-arranged network slice, which is selected from each historical network slice; the terminal access behavior information includes access time, access base station identifier, access application IP and transmission channel slice type; adjusting each pre-arranged network slice according to the slice link information, the user behavior aggregation degree and the current pre-arrangement accuracy; wherein the pre-arrangement accuracy is determined based on the pre-arranged network slice update situation in a preset detection time period before the current time.

[0055] Specifically, in order to further ensure the accuracy of the pre-arranged network slice, the embodiment also adjusts the pre-arranged network slice, first acquires slice link information corresponding to each pre-arranged network slice and user behavior aggregation degree, wherein the user behavior aggregation degree can reflect the number of terminals in the historical network slice that are consistent with the terminal access behavior information of the pre-arranged network slice, and if the number of terminals is large, it indicates that the accuracy of the pre-arranged network slice is higher. In order to be able to acquire the user behavior aggregation degree, the embodiment adopts a collaborative filtering manner based on the terminal access behavior information of the past historical network slice, including historical access time, historical access base station identifier, historical access application IP and historical transmission channel slice type, to filter out the historical network slice that is consistent with the terminal access behavior information of the pre-arranged network slice, thereby determining the number of terminals. The embodiment can set a corresponding user behavior aggregation degree for each terminal quantity range, or determine the user behavior aggregation degree based on the ratio of the number of terminals that are consistent with the terminal access behavior information to the total number of terminals corresponding to all historical network slices, which is not limited in the embodiment.

[0056] Further, according to the slice link information and the user behavior aggregation degree, the accuracy of each pre-arranged network slice is evaluated, and the pre-arranged network slice with higher accuracy is retained based on the current pre-arrangement accuracy, so as to ensure the pre-arrangement accuracy of the final 5G network slice, and avoid triggering the update of the pre-arranged network slice due to the inability to match too many pre-arranged network slices when actually receiving the access request, thereby affecting the configuration efficiency of the 5G network slice and the adaptability under high load or burst traffic.

[0057] As a preferred solution, the adjusting of each pre-arranged network slice according to the slice link information, the user behavior aggregation degree and the current pre-arrangement accuracy specifically includes: calculating the accuracy of the pre-arranged network slice according to the slice link information and the user behavior aggregation degree; deleting the pre-arranged network slice with accuracy less than the pre-arrangement accuracy.

[0058] Specifically, according to the slice link information and the user behavior aggregation degree, the embodiment first calculates the accuracy of each pre-arranged network slice, and for the pre-arranged network slice with accuracy less than the current pre-arrangement accuracy, the embodiment directly deletes it, so as to avoid the pre-arranged network slice with lower accuracy from occupying limited slice resources, improve the utilization rate of the slice resources, and ensure the reliability of the pre-arranged network slice sent to the 5G base station controller subsequently.

[0059] As a preferred solution, the method specifically determines the current pre-arrangement accuracy through the following steps: obtain the pre-arranged network slice update situation in the preset detection time period from the 5G base station controller; determine the pre-arranged network slice update quantity corresponding to each pre-arranged time period in the preset detection time period based on the pre-arranged network slice update situation; determine the current pre-arrangement accuracy based on the preset weight coefficient, the pre-arranged network slice quantity and the pre-arranged network slice update quantity corresponding to each pre-arranged time period in the preset detection time period.

[0060] Specifically, in order to be able to determine the current pre-arrangement accuracy, the embodiment first obtains the pre-arranged network slice update situation in the preset detection time period from the 5G base station controller, which reflects the difference quantity between the pre-arranged network slice set sent in the preset detection time period and the actual network slice set executed by the 5G base station. In the statistics of the difference quantity, the predicted access base station identifier and the predicted access application IP corresponding to the pre-arranged network slice are taken as the unique association basis, and the remaining elements are taken as the difference judgment basis. If any element is not completely the same, it is determined that the pre-arranged network slice is different from the corresponding actual network slice.

[0061] Further, since the preset detection time period may contain pre-arranged network slices sent in different pre-arranged time periods, and the update situation of the pre-arranged network slices with closer time distance can reflect the current pre-arrangement accuracy, the embodiment obtains the pre-arranged network slice update quantity corresponding to each pre-arranged time period in the preset detection time period. The pre-arranged network slice update quantity is the difference quantity between the pre-arranged network slice set corresponding to a single pre-arranged time period and the actual network slice set executed by the 5G base station.

[0062] Further, according to the preset weight coefficient corresponding to each pre-arranged time period in the preset detection time period, preferably, the closer the time distance of the pre-arranged time period, the greater the preset weight coefficient corresponding to the pre-arranged time period. The ratio of the pre-arranged network slice update quantity and the pre-arranged network slice quantity is weighted and summed to obtain the update rate of the recent pre-arranged network slice, and then by using 1 minus the update rate, the pre-arrangement accuracy is obtained.

[0063] The embodiment of the application can timely adjust the current 5G network slice pre-arrangement by obtaining the recent pre-arranged network slice update situation from the 5G base station controller and determining the current pre-arrangement accuracy, and can continuously improve the pre-arrangement accuracy of the 5G network slice by retaining the pre-arranged network slice with accuracy greater than the current pre-arrangement accuracy.

[0064] Preferably, the slice link information comprises channel quality, user experience score and slice link complexity; the user experience score is determined based on a ratio between a user complaint rate of a historical network slice consistent with terminal access behavior information corresponding to the pre-arranged network slice and an average user complaint rate of each historical network slice; and the slice link complexity is obtained based on a product of slice link length and non-link transmission computing power consumption.

[0065] Specifically, the slice link information in the embodiment further comprises channel quality, user experience score and slice link complexity, wherein the channel quality can be directly obtained from the 5G base station controller, and the user experience score is determined based on a ratio between a user complaint rate of a historical network slice consistent with terminal access behavior information corresponding to the pre-arranged network slice and an average user complaint rate of each historical network slice; the higher the ratio, the worse the user experience of the pre-arranged network slice, and accordingly the lower the user experience score and the lower the accuracy; for different ratios or ratio ranges, corresponding user experience scores can be set, which are not limited in the embodiment. The slice link complexity is obtained based on a product of slice link length and non-link transmission computing power consumption; the non-link transmission computing power consumption is the computing power consumption other than link transmission; the greater the computing power consumption, the higher the slice link complexity, and the higher the probability of network instability. Based on this, the calculation method of the accuracy of the pre-arranged network slice in the embodiment is as follows: multiplying the channel quality by the user experience score to obtain a first value, then dividing the first value by the slice link complexity to obtain a second value, and finally multiplying the second value by the user behavior aggregation degree to obtain the accuracy of the pre-arranged network slice.

[0066] The embodiment of the application can improve the rationality and accuracy of obtaining the accuracy of the pre-arranged network slice by fully considering the channel quality, user experience score and slice link complexity when determining the accuracy of the pre-arranged network slice, thereby further ensuring the reliability of the pre-arranged network slice sent to the 5G base station controller.

[0067] Preferably, the method further comprises: sending each pre-arranged network slice to a 5G base station controller; distributing each pre-arranged network slice to each 5G base station in the management range of the 5G base station controller according to the floating resource shortage degree of each 5G base station and the preset resource support degree saturation of the 5G base station controller; and

[0068] Specifically, after determining the current various pre-arranged network slices, it is needed to send them to the 5G base station controller, so that the 5G base station controller performs distribution of the pre-arranged network slices based on the floating resource shortage and the preset resource support saturation of each 5G base station within the management range of the 5G base station controller, wherein the preset resource support saturation reflects the safety upper limit requirement of the 5G base station resource utilization rate, and the floating resource shortage is a dynamic quantitative value of the actual resource shortage degree of the current 5G base station, and the resource support saturation and the floating resource shortage determine the execution amount of the pre-arranged network slice of the current 5G base station.

[0069] The embodiment of the application can distribute the pre-arranged network slice by considering the actual resource utilization of each 5G base station through the 5G base station controller, thereby avoiding 5G base station resource overload under the premise of ensuring high slice resource utilization, leading to network service quality decline and affecting user experience.

[0070] As a preferred scheme, the 5G base station controller distributes each of the pre-arranged network slices to each of the 5G base stations according to the floating resource shortage and the preset resource support saturation of each of the 5G base stations within the management range of the 5G base station controller, specifically including: Based on the current various pre-arranged network slices, the 5G base station controller evaluates the floating resource shortage of each of the 5G base stations after executing the pre-arranged network slice; When the floating resource shortage of any one of the 5G base stations is greater than or equal to the resource support saturation, the 5G base station controller deletes the pre-arranged network slice with the lowest performance score required to be executed by the any one of the 5G base stations from each of the pre-arranged network slices; and the above steps are repeatedly executed until the floating resource shortage of each of the 5G base stations is less than the resource support saturation; Based on the predicted access base station identifier corresponding to the current various pre-arranged network slices, the 5G base station controller obtains a target network slice of each of the 5G base stations, and distributes each of the target network slices to each of the 5G base stations to instruct the 5G base station to configure the target network slice, and activate the target network slice for data transmission when detecting that the terminal corresponding to the target network slice requests.

[0071] Specifically, since each pre-arranged network slice has a corresponding predicted access base station identifier, the embodiment first needs to evaluate the floating resource shortage degree of each 5G base station after it executes the corresponding pre-arranged network slice through the 5G base station controller, to check whether it will be overloaded. If the floating resource shortage degree of any 5G base station is greater than or equal to the resource support degree saturation, it means that the 5G base station will be overloaded after it executes the corresponding pre-arranged network slice, so the pre-arranged network slice with the lowest performance score that needs to be executed by the 5G base station needs to be deleted to reduce the execution amount of the pre-arranged network slice and alleviate the bearing pressure. Repeat the above step process until each 5G base station will not be overloaded after it executes the corresponding pre-arranged network slice.

[0072] Further, according to each predicted access base station identifier, the current each pre-arranged network slice is distributed to each 5G base station. Each 5G base station performs target network slice pre-configuration after receiving the target network slice, so that it can be quickly activated when needed. Specifically, the 5G base station will judge whether there is an interaction link of the target network slice based on the current 5G network slice connection, if not, create an interaction link based on the target link corresponding to the target network slice, and save it in the cache, then negotiate with other nodes in the slice link to complete the encryption key exchange between nodes. After the target network slice is configured, the 5G base station activates the target network slice when it detects a terminal request matching the target network slice, and uses the activated target network slice for data transmission. For the target network slice that has not been activated for more than the current pre-arrangement period, it is deleted from the cache.

[0073] The embodiment of the application can ensure user experience while avoiding the situation that the 5G base station will be overloaded after it executes the corresponding pre-arranged network slice by deleting the pre-arranged network slice with the lowest performance score that needs to be executed by the 5G base station when the floating resource shortage degree of any 5G base station is greater than or equal to the resource support degree saturation.

[0074] As a preferred solution, the running state information includes hardware resource usage, hardware resource maximum usage and hardware resource minimum usage.

[0075] Specifically, the hardware resource usage is recorded as , the hardware resource maximum usage is recorded as , and the hardware resource minimum usage is recorded as , so that the floating resource shortage degree of the 5G base station after executing the pre-arranged network slice is: To accurately reflect the resource usage of the 5G base station after performing the pre-arrangement network slice.

[0076] See Figure 2 The second aspect of the embodiment of the present application provides a 5G network slice pre-arrangement device 100, which comprises: The terminal access prediction module 11 is configured to determine the predicted access base station trajectory and the target service request type of each user terminal in the pre-arrangement time period based on the historical access base station trajectory information of each user terminal. The network slice pre-arrangement module 12 is configured to perform network slice arrangement based on the predicted access base station trajectory and the target service request type, and obtain at least one pre-arranged network slice in the pre-arrangement time period.

[0077] As a preferred solution, the terminal access prediction module 11 is configured to determine the predicted access base station trajectory and the target service request type of each user terminal in the pre-arrangement time period based on the historical access base station trajectory information of each user terminal, and specifically comprises: dividing the pre-arrangement time period into at least one prediction time period according to a preset prediction time length; generating an initial prediction access trajectory template corresponding to each prediction time period based on the start time and end time of each prediction time period; inputting the historical access base station trajectory information and the initial prediction access trajectory template into a transformer structure-based access trajectory prediction model to perform access trajectory prediction, and obtaining the prediction access trajectory information of each user terminal in each prediction time period; determining the predicted access base station trajectory and the target service request type of each user terminal in the pre-arrangement time period based on each prediction access trajectory information.

[0078] As a preferred solution, the terminal access prediction module 11 is configured to determine the predicted access base station trajectory and the target service request type of each user terminal in the pre-arrangement time period based on each prediction access trajectory information, and specifically comprises: determining the predicted access base station identifier, the predicted access application IP and the target transmission channel slice type of each prediction time period of the user terminal based on each prediction access trajectory information; when it is detected that the user terminal has at least two target prediction access trajectory information which are continuous in the prediction time period and consistent in the predicted access base station identifier, the predicted access application IP and the target transmission channel slice type, merging each target prediction access trajectory information to obtain prediction access trajectory merging information; Determine the predicted access base station trajectory and the target service request type based on the predicted access base station identifier and the predicted access application IP in the predicted access trajectory merging information, and the predicted access base station identifier and the predicted access application IP in the predicted access trajectory information that is not involved in the merging.

[0079] As a preferred solution, the historical access base station trajectory information at least includes historical access start time, historical access end time, historical access base station identifier, historical access application IP and historical transmission channel slice type.

[0080] As a preferred solution, the network slice pre-arrangement module 12 is configured to perform network slice arrangement based on the predicted access base station trajectory and the target service request type, to obtain at least one pre-arranged network slice within the pre-arrangement time period, specifically including: Determine at least one target service request link for each user terminal based on the predicted access base station trajectory and the target service request type; Perform performance evaluation on the target service request link based on the link resource of the target service request link and the target service request type corresponding thereto, to obtain a performance score of each target service request link; Perform network slice construction based on each target service request link with a performance score greater than the average performance score, to obtain at least one pre-arranged network slice within the pre-arrangement time period; wherein the average performance score is the average value of the performance scores of all target service request links within the pre-arrangement time period.

[0081] As a preferred solution, the device is further configured to: Obtain slice link information and user behavior aggregation degree corresponding to each pre-arranged network slice; wherein the user behavior aggregation degree is determined based on the number of terminals with consistent terminal access behavior information as that of the pre-arranged network slice selected from each historical network slice; the terminal access behavior information includes access time, access base station identifier, access application IP and transmission channel slice type; Adjust each pre-arranged network slice based on the slice link information, the user behavior aggregation degree and the current pre-arrangement accuracy; wherein the pre-arrangement accuracy is determined based on the pre-arranged network slice update within a preset detection time period before the current time.

[0082] As a preferred solution, the device is configured to adjust each pre-arranged network slice based on the slice link information, the user behavior aggregation degree and the current pre-arrangement accuracy, specifically including: According to the slice link information and the user behavior aggregation degree, the accuracy of the pre-arranged network slice is calculated; The pre-arranged network slice with an accuracy less than the pre-arranged accuracy is deleted.

[0083] As a preferred solution, the device specifically determines the current pre-arranged accuracy by the following steps: Obtain the pre-arranged network slice update situation in the preset detection time period from the 5G base station controller; Based on the pre-arranged network slice update situation, determine the pre-arranged network slice update quantity corresponding to each pre-arranged time period in the preset detection time period; Based on the preset weight coefficient, the pre-arranged network slice quantity and the pre-arranged network slice update quantity corresponding to each pre-arranged time period in the preset detection time period, determine the current pre-arranged accuracy.

[0084] As a preferred solution, the slice link information includes channel quality, user experience score and slice link complexity; wherein the user experience score is determined based on the ratio between the user complaint rate of the historical network slice consistent with the terminal access behavior information corresponding to the pre-arranged network slice and the average user complaint rate of each historical network slice; the slice link complexity is obtained based on the product of the slice link length and the non-link transmission algorithm power consumption.

[0085] As a preferred solution, the device is further used for: Send each pre-arranged network slice to the 5G base station controller; Through the 5G base station controller, distribute each pre-arranged network slice to each 5G base station according to the floating resource shortage degree of each 5G base station within the management range of the 5G base station controller and the preset resource support force degree saturation; wherein the floating resource shortage degree is determined based on the running state information of the 5G base station after executing the pre-arranged network slice.

[0086] As a preferred solution, the device is used to distribute each pre-arranged network slice to each 5G base station through the 5G base station controller according to the floating resource shortage degree of each 5G base station within the management range of the 5G base station controller and the preset resource support force degree saturation, specifically including: Based on the current each pre-arranged network slice, evaluate the floating resource shortage degree of each 5G base station after executing the pre-arranged network slice through the 5G base station controller; When the floating resource shortage degree of any one of the 5G base stations is greater than or equal to the resource support degree saturation, the 5G base station controller deletes the pre-arranged network slice with the lowest performance score in each of the pre-arranged network slices required to be executed by the any one of the 5G base stations; the above steps are repeatedly executed until the floating resource shortage degree of each of the 5G base stations is less than the resource support degree saturation; Based on the predicted access base station identifier corresponding to each of the current pre-arranged network slices, the 5G base station controller obtains target network slices of each of the 5G base stations, and distributes each of the target network slices to each of the 5G base stations to instruct the 5G base station to configure the target network slice, and when detecting that the terminal corresponding to the target network slice requests, activate the target network slice for data transmission.

[0087] As a preferred solution, the running state information includes a hardware resource usage rate, a hardware resource maximum usage rate, and a hardware resource minimum usage rate.

[0088] The pre-arrangement device 100 of the 5G network slice provided by the embodiment of the application can obtain at least one pre-arranged network slice in the pre-arrangement time period by predicting the access base station trajectory and the service request type of the user terminal in the pre-arrangement time period based on the historical access base station trajectory information of the user terminal, and arranging the network slice based thereon, so that the preparation work of the 5G network slice can be completed in advance, the configuration efficiency of the 5G network slice and the adaptability under high load or burst traffic are improved, and it is ensured that the 5G network slice can be quickly activated when the user demand changes, and different network resource configurations can be quickly switched.

[0089] Please refer to Figure 3 The third aspect of the embodiment of the application provides an electronic device 200, which comprises a memory 22, a processor 21, and a computer program stored in the memory 22 and capable of running on the processor 21, and the processor 21 implements the pre-arrangement method of the 5G network slice of any one of the first aspect.

[0090] For example, the computer program can be divided into one or more modules / units, which are stored in the memory 22 and executed by the processor 21 to complete the application. The one or more modules / units can be a series of computer program instruction segments capable of completing a specific function, which are used to describe the execution process of the computer program in the electronic device 200.

[0091] The electronic device 200 can include, but is not limited to, a processor 21, a memory 22. Those skilled in the art can understand that the schematic diagram is only an example of the electronic device 200, and does not constitute a limitation on the electronic device 200, and can include more or fewer components than the diagram, or combine certain components, or different components, for example, the electronic device 200 can also include an input / output device, a network access device, a bus, etc.

[0092] The processor 21 can be a central processing unit (CPU), and can also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs) or other programmable logic devices, discrete gates or transistor logic components, discrete hardware components, etc. The general-purpose processor can be a microprocessor or the processor 21 can also be any conventional processor 21, etc., which is the control center of the electronic device 200, and connects each part of the entire electronic device 200 through various interfaces and lines.

[0093] The memory 22 can be used to store computer programs and / or modules, and the processor 21 realizes various functions of the electronic device 200 by running or executing computer programs and / or modules stored in the memory 22, and calling data stored in the memory 22. The memory 22 can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, at least one application required by a function (such as a sound playing function, an image playing function, etc.), etc.; the data storage area can store data created according to the use of the mobile phone (such as audio data, a phone book, etc.), etc. In addition, the memory 22 can include a high-speed random access memory, and can also include a non-volatile memory, for example, a hard disk, a memory, a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, at least one disk storage device, a flash memory device, or other volatile solid-state memory devices.

[0094] The fourth aspect of the embodiment of the present application provides a computer readable storage medium, the computer readable storage medium includes a stored computer program, wherein when the computer program runs, the device where the computer readable storage medium is located executes the pre-arrangement method of the 5G network slice of any one of the first aspect.

[0095] The fifth aspect of the embodiments of the present application provides a computer program product comprising computer programs / instructions which, when executed by a processor, implement the steps of the method for pre-orchestration of 5G network slices according to any of the first aspect.

[0096] If the modules / units of the electronic device are implemented in the form of software function units and sold or used as independent products, they can be stored in a computer readable storage medium. Based on this understanding, all or part of the processes in the above-mentioned embodiment methods can also be completed by a computer program instructing related hardware, and the computer program can be stored in a computer readable storage medium. When the processor executes the computer program, the steps of each method embodiment can be implemented. The computer program includes computer program code, which can be in the form of source code, object code, executable files or some intermediate forms. The computer readable medium can include any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal and software distribution medium, etc.

[0097] The above is the preferred embodiment of the present application. It should be noted that for those skilled in the art, without departing from the principles of the present application, a number of improvements and refinements can be made, which are also considered within the scope of protection of the present application.

Claims

1. A pre-arrangement method for 5G network slices, characterized in that, include: Based on the historical access base station trajectory information of each user terminal, the predicted access base station trajectory and target service request type of each user terminal within the pre-arranged time period are determined. Based on the predicted access base station trajectory and the target service request type, network slice orchestration is performed to obtain at least one pre-arranged network slice within the pre-arranged time period.

2. The 5G network slice pre-arrangement method as described in claim 1, characterized in that, The step of determining the predicted access base station trajectory and target service request type for each user terminal within a pre-arranged time period based on the historical access base station trajectory information of each user terminal specifically includes: According to the preset prediction time length, the pre-arranged time period is divided into time segments to obtain at least one prediction time period; Based on the start and end times of each of the predicted time periods, an initialized predicted access trajectory template corresponding to each of the predicted time periods is generated. The historical access base station trajectory information and the initial predicted access trajectory template are input into the access trajectory prediction model based on the transformer structure to perform access trajectory prediction, so as to obtain the predicted access trajectory information of each user terminal in each predicted time period. Based on the predicted access trajectory information, the predicted access base station trajectory and the target service request type of each user terminal within the pre-arranged time period are determined.

3. The 5G network slicing pre-arrangement method as described in claim 2, characterized in that, The step of determining the predicted access base station trajectory and the target service request type for each user terminal within the pre-arranged time period based on the predicted access trajectory information specifically includes: Based on the predicted access trajectory information, the predicted access base station identifier, predicted access application IP, and target transmission channel slice type of the user terminal in each predicted time period are determined. When it is detected that the user terminal has at least two target predicted access trajectory information that are continuous in the predicted time period and have the same predicted access base station identifier, predicted access application IP and target transmission channel slice type, the target predicted access trajectory information is merged to obtain predicted access trajectory merging information. Based on the predicted access base station identifier and predicted access application IP in the predicted access trajectory merging information and / or the predicted access base station identifier and predicted access application IP in the predicted access trajectory information that did not participate in the merging, the predicted access base station trajectory and the target service request type are determined respectively.

4. The 5G network slice pre-arrangement method as described in claim 2 or 3, characterized in that, The historical access base station trajectory information includes at least the historical access start time, historical access end time, historical access base station identifier, historical access application IP, and historical transmission channel slice type.

5. The 5G network slice pre-arrangement method as described in claim 1, characterized in that, The step of performing network slice orchestration based on the predicted access base station trajectory and the target service request type to obtain at least one pre-arranged network slice within the pre-arrangement time period specifically includes: Based on the predicted access base station trajectory and the target service request type, at least one target service request link is determined for each user terminal; Based on the link resources of the target service request link and the corresponding target service request type, the performance of the target service request link is evaluated to obtain a performance score for each target service request link. Network slices are constructed for each target service request link whose performance score is greater than the average performance score, to obtain at least one pre-arranged network slice within the pre-arranged time period; wherein the average performance score is the average of the performance scores of all target service request links within the pre-arranged time period.

6. The 5G network slice pre-arrangement method as described in claim 5, characterized in that, The method further includes: Obtain the slice link information and user behavior clustering degree corresponding to each of the pre-arranged network slices; wherein, the user behavior clustering degree is determined based on the number of terminals that are consistent with the terminal access behavior information corresponding to the pre-arranged network slices selected from each historical network slice; the terminal access behavior information includes access time, access base station identifier, access application IP and transmission channel slice type; Based on the slice link information, the user behavior aggregation degree, and the current pre-arrangement accuracy, each pre-arranged network slice is adjusted; wherein, the pre-arrangement accuracy is determined based on the pre-arranged network slice update status within a preset detection time period prior to the current time.

7. The 5G network slice pre-arrangement method as described in claim 6, characterized in that, The step of adjusting each pre-arranged network slice based on the slice link information, the user behavior clustering degree, and the current pre-arrangement accuracy specifically includes: The accuracy of the pre-arranged network slice is calculated based on the slice link information and the user behavior clustering degree. Delete the pre-arranged network slices whose accuracy is less than the pre-arranged accuracy.

8. The 5G network slice pre-arrangement method as described in claim 6, characterized in that, The method specifically determines the current pre-arrangement accuracy through the following steps: Obtain the update status of the pre-arranged network slices within the preset detection time period from the 5G base station controller; Based on the pre-arranged network slice update status, determine the number of pre-arranged network slice updates corresponding to each pre-arranged time period within the preset detection time period; The current pre-arrangement accuracy is determined based on the preset weight coefficients corresponding to each pre-arranged time period within the preset detection time period, the number of pre-arranged network slices, and the number of pre-arranged network slice updates.

9. The 5G network slice pre-arrangement method as described in claim 6 or 7, characterized in that, The slice link information includes channel quality, user experience score, and slice link complexity; wherein, the user experience score is determined based on the ratio between the user complaint rate of historical network slices that are consistent with the terminal access behavior information corresponding to the pre-arranged network slice and the average user complaint rate of each historical network slice; the slice link complexity is obtained based on the product between the slice link length and the non-link transmission computing power consumption.

10. The 5G network slice pre-arrangement method as described in claim 5, characterized in that, The method further includes: Each of the pre-arranged network slices is sent to the 5G base station controller; The 5G base station controller distributes each pre-arranged network slice to each 5G base station based on the floating resource scarcity and preset resource support saturation of each 5G base station within its management scope; wherein the floating resource scarcity is determined based on the operating status information of the 5G base station after executing the pre-arranged network slice.

11. The 5G network slice pre-arrangement method as described in claim 10, characterized in that, The step of distributing pre-arranged network slices to each of the 5G base stations through the 5G base station controller, based on the floating resource scarcity and preset resource support saturation of each 5G base station within the management scope of the 5G base station controller, specifically includes: Based on the current pre-arranged network slices, the floating resource scarcity of each 5G base station after executing the pre-arranged network slices is evaluated by the 5G base station controller; When the floating resource scarcity of any 5G base station is greater than or equal to the resource support saturation, the 5G base station controller deletes the pre-arranged network slice with the lowest performance score among the pre-arranged network slices to be executed by any 5G base station; repeat the above steps until the floating resource scarcity of each 5G base station is less than the resource support saturation. Based on the predicted access base station identifiers corresponding to each of the current pre-arranged network slices, the target network slices of each of the 5G base stations are obtained through the 5G base station controller, and each of the target network slices is distributed to each of the 5G base stations to instruct the 5G base stations to configure the target network slices, and when a terminal request corresponding to the target network slice is detected, the target network slice is activated to perform data transmission.

12. The 5G network slice pre-arrangement method as described in claim 10, characterized in that, The operational status information includes hardware resource utilization rate, maximum hardware resource utilization rate, and minimum hardware resource utilization rate.

13. A pre-arrangement device for 5G network slicing, characterized in that, include: The terminal access prediction module is used to determine the predicted access base station trajectory and target service request type of each user terminal within a pre-arranged time period based on the historical access base station trajectory information of each user terminal. The network slice pre-arrangement module is used to perform network slice arrangement based on the predicted access base station trajectory and the target service request type, and obtain at least one pre-arranged network slice within the pre-arrangement time period.

14. The 5G network slicing pre-arrangement device as described in claim 13, characterized in that, The terminal access prediction module is used to determine the predicted access base station trajectory and target service request type of each user terminal within a pre-arranged time period based on the historical access base station trajectory information of each user terminal. Specifically, it includes: According to the preset prediction time length, the pre-arranged time period is divided into time segments to obtain at least one prediction time period; Based on the start and end times of each of the predicted time periods, an initialized predicted access trajectory template corresponding to each of the predicted time periods is generated. The historical access base station trajectory information and the initial predicted access trajectory template are input into the access trajectory prediction model based on the transformer structure to perform access trajectory prediction, so as to obtain the predicted access trajectory information of each user terminal in each predicted time period. Based on the predicted access trajectory information, the predicted access base station trajectory and the target service request type of each user terminal within the pre-arranged time period are determined.

15. The 5G network slicing pre-arrangement device as described in claim 14, characterized in that, The terminal access prediction module is used to determine the predicted access base station trajectory and the target service request type of each user terminal within the pre-arranged time period based on the predicted access trajectory information, specifically including: Based on the predicted access trajectory information, the predicted access base station identifier, predicted access application IP, and target transmission channel slice type of the user terminal in each predicted time period are determined. When it is detected that the user terminal has at least two target predicted access trajectory information that are continuous in the predicted time period and have the same predicted access base station identifier, predicted access application IP and target transmission channel slice type, the target predicted access trajectory information is merged to obtain predicted access trajectory merging information. Based on the predicted access base station identifier and predicted access application IP in the predicted access trajectory merging information and / or the predicted access base station identifier and predicted access application IP in the predicted access trajectory information that did not participate in the merging, the predicted access base station trajectory and the target service request type are determined respectively.

16. The 5G network slicing pre-arrangement device as described in claim 14 or 15, characterized in that, The historical access base station trajectory information includes at least the historical access start time, historical access end time, historical access base station identifier, historical access application IP, and historical transmission channel slice type.

17. The 5G network slicing pre-arrangement device as described in claim 13, characterized in that, The network slice pre-arrangement module is used to perform network slice arrangement based on the predicted access base station trajectory and the target service request type, to obtain at least one pre-arranged network slice within the pre-arrangement time period, specifically including: Based on the predicted access base station trajectory and the target service request type, at least one target service request link is determined for each user terminal; Based on the link resources of the target service request link and the corresponding target service request type, the performance of the target service request link is evaluated to obtain a performance score for each target service request link. Network slices are constructed for each target service request link whose performance score is greater than the average performance score, to obtain at least one pre-arranged network slice within the pre-arranged time period; wherein the average performance score is the average of the performance scores of all target service request links within the pre-arranged time period.

18. The 5G network slicing pre-arrangement device as described in claim 17, characterized in that, The device is also used for: Obtain the slice link information and user behavior clustering degree corresponding to each of the pre-arranged network slices; wherein, the user behavior clustering degree is determined based on the number of terminals that are consistent with the terminal access behavior information corresponding to the pre-arranged network slices selected from each historical network slice; the terminal access behavior information includes access time, access base station identifier, access application IP and transmission channel slice type; Based on the slice link information, the user behavior aggregation degree, and the current pre-arrangement accuracy, each pre-arranged network slice is adjusted; wherein, the pre-arrangement accuracy is determined based on the pre-arranged network slice update status within a preset detection time period prior to the current time.

19. The 5G network slicing pre-arrangement device as described in claim 18, characterized in that, The device is used to adjust each of the pre-arranged network slices based on the slice link information, the user behavior aggregation degree, and the current pre-arrangement accuracy, specifically including: The accuracy of the pre-arranged network slice is calculated based on the slice link information and the user behavior clustering degree. Delete the pre-arranged network slices whose accuracy is less than the pre-arranged accuracy.

20. The 5G network slicing pre-arrangement device as described in claim 18, characterized in that, The device determines the current pre-arrangement accuracy through the following steps: Obtain the update status of the pre-arranged network slices within the preset detection time period from the 5G base station controller; Based on the pre-arranged network slice update status, determine the number of pre-arranged network slice updates corresponding to each pre-arranged time period within the preset detection time period; The current pre-arrangement accuracy is determined based on the preset weight coefficients corresponding to each pre-arranged time period within the preset detection time period, the number of pre-arranged network slices, and the number of pre-arranged network slice updates.

21. The 5G network slicing pre-arrangement device as described in claim 18 or 19, characterized in that, The slice link information includes channel quality, user experience score, and slice link complexity; wherein, the user experience score is determined based on the ratio between the user complaint rate of historical network slices that are consistent with the terminal access behavior information corresponding to the pre-arranged network slice and the average user complaint rate of each historical network slice; the slice link complexity is obtained based on the product between the slice link length and the non-link transmission computing power consumption.

22. The 5G network slicing pre-arrangement device as described in claim 17, characterized in that, The device is also used for: Each of the pre-arranged network slices is sent to the 5G base station controller; The 5G base station controller distributes each pre-arranged network slice to each 5G base station based on the floating resource scarcity and preset resource support saturation of each 5G base station within its management scope; wherein the floating resource scarcity is determined based on the operating status information of the 5G base station after executing the pre-arranged network slice.

23. The 5G network slicing pre-arrangement device as described in claim 22, characterized in that, The device is used to distribute each pre-arranged network slice to each of the 5G base stations through the 5G base station controller, based on the floating resource scarcity and preset resource support saturation of each 5G base station within the management scope of the 5G base station controller, specifically including: Based on the current pre-arranged network slices, the floating resource scarcity of each 5G base station after executing the pre-arranged network slices is evaluated by the 5G base station controller; When the floating resource scarcity of any 5G base station is greater than or equal to the resource support saturation, the 5G base station controller deletes the pre-arranged network slice with the lowest performance score among the pre-arranged network slices to be executed by any 5G base station; repeat the above steps until the floating resource scarcity of each 5G base station is less than the resource support saturation. Based on the predicted access base station identifiers corresponding to each of the current pre-arranged network slices, the target network slices of each of the 5G base stations are obtained through the 5G base station controller, and each of the target network slices is distributed to each of the 5G base stations to instruct the 5G base stations to configure the target network slices, and when a terminal request corresponding to the target network slice is detected, the target network slice is activated to perform data transmission.

24. The 5G network slicing pre-arrangement device as described in claim 22, characterized in that, The operational status information includes hardware resource utilization rate, maximum hardware resource utilization rate, and minimum hardware resource utilization rate.

25. An electronic device, characterized in that, It includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the 5G network slice pre-arrangement method according to any one of claims 1 to 12.

26. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored computer program, wherein, when the computer program is executed, it controls the device on which the computer-readable storage medium is located to perform the 5G network slicing pre-arrangement method according to any one of claims 1 to 12.

27. A computer program product, characterized in that, It includes a computer program / instruction that, when executed by a processor, implements the steps of the 5G network slicing pre-arrangement method according to any one of claims 1 to 12.

Images