Network slice configuration method, apparatus, and storage medium
By parsing network slice request information using a trained NLP model, configuring parameters based on business scenarios and priority information, and employing fuzzy matching when necessary, the problem of inaccurate network slice parameter configuration in existing technologies is solved, achieving more efficient and accurate slice parameter configuration.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- CHINA TELECOM CORP LTD
- Filing Date
- 2021-12-24
- Publication Date
- 2026-08-04
AI Technical Summary
Existing automated configuration methods for network slicing parameters cannot accurately perceive the actual network resource status of operators, resulting in low configuration accuracy and poor customer experience.
By using a trained NLP model to parse network slice request information and extract slice parameter labels, and combining business scenario information, priority information and slice function requirement parameter information, network slice function parameters are configured, and fuzzy matching algorithm is used to adjust when accurate configuration is not possible.
It improves the accuracy and efficiency of network slicing parameter configuration, reduces the risk of human error, and enhances the user experience.
Smart Images
Figure CN116367174B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of communication technology, and in particular to a network slicing configuration method, apparatus, and storage medium. Background Technology
[0002] Network slicing, also known simply as slicing, is an on-demand network deployment method. The application of slicing provides flexible configuration capabilities for differentiated services in communication networks. In 5G network slicing applications, a rich variety of differentiated network parameter configurations will be implemented to meet the diverse network configuration needs of users. The number of differentiated slice parameters will continue to increase, requiring coverage of all network configuration requirements. Automated auxiliary configuration of network slice parameters is essential for achieving efficient and high-value configuration. Current automated auxiliary configuration of network slice parameters typically uses typical Natural Language Processing (NLP) models. However, existing NLP models cannot accurately perceive the actual network resource status of operators, resulting in low accuracy in slice parameter configuration and a poor customer experience. Summary of the Invention
[0003] In view of this, one technical problem to be solved by the present invention is to provide a network slicing configuration method, apparatus and storage medium.
[0004] According to a first aspect of this disclosure, a network slice configuration method is provided, comprising: parsing network slice request information and extracting slice parameter tags corresponding to the network slice request; determining business scenario information, priority information, and slice function requirement parameter information based on the slice parameter tags; configuring network slice function parameters according to a parameter configuration strategy and based on the business scenario information, the priority information, and the slice function requirement parameter information; and configuring network slice function parameters by fuzzy matching when it is determined that network slice function parameter configuration cannot be performed according to the parameter configuration strategy.
[0005] Optionally, parsing the network slice request information and extracting the slice parameter labels corresponding to the network slice request includes: using a trained NLP model to parse the network slice request information and extracting slice parameters and corresponding slice parameter labels; wherein, the slice parameter labels include: business scenario labels, network function slice parameter labels, priority auxiliary labels, and parameter type labels.
[0006] Optionally, determining the service scenario information, priority information, and slice function requirement parameter information based on the slice parameter label includes: obtaining the service scenario information corresponding to the service scenario label; wherein, the service scenario includes: eMBB, mMTC, and uRLLC scenarios; obtaining the priority information corresponding to the priority auxiliary label; obtaining the slice function requirement parameter information corresponding to the network slice function parameter label; wherein, the slice function requirement parameters include: bandwidth, reliability, and latency parameters; and setting corresponding type attributes for the service scenario information, the priority information, and the slice function requirement parameter information according to the parameter type label.
[0007] Optionally, configuring network slice function parameters according to the parameter configuration strategy and based on the business scenario information, the priority information, and the slice function requirement parameter information includes: if the user is determined to be a dedicated slice user based on the priority information, then configuring network slice function parameters based on the slice function requirement parameter information; wherein, for default parameters in the network slice function parameters, a preset parameter with the highest priority corresponding to the business scenario information is selected and configured based on the parameter configuration strategy; after the network slice function parameters are configured, a network slice function parameter threshold corresponding to the business scenario information is obtained based on the parameter configuration strategy; the currently configured network slice function parameter value is verified based on the network slice function parameter threshold, and corresponding processing is performed based on the verification result.
[0008] Optionally, configuring network slice function parameters according to the parameter configuration strategy and based on the business scenario information, the priority information, and the slice function requirement parameter information includes: if the user is determined to be a non-dedicated slice user based on the priority information, then determining the non-dedicated priority level based on the priority information; obtaining the corresponding parameter increase or decrease threshold based on the parameter configuration strategy and the non-dedicated priority level; and configuring network slice function parameters based on the slice function requirement parameter information and the parameter increase or decrease threshold.
[0009] Optionally, the step of configuring fuzzy matching for network slicing function parameters includes: for non-dedicated slice users, if it is determined that network slicing function parameters cannot be configured according to the parameter configuration strategy, then fuzzy matching is performed on the network slicing function parameters to determine the slice function parameter configuration group.
[0010] Optionally, the step of performing fuzzy matching configuration on network slice function parameters to determine slice function parameter configuration combinations includes: obtaining allocable function attribute parameter values; generating a set of function attribute parameters based on function attribute parameter values with the same function attribute; obtaining function requirement parameter values based on the slice function requirement parameter information; generating function requirement parameter combinations based on the function requirement parameter values; and determining a set of function attribute parameters corresponding to each function requirement parameter value; calculating the difference between each function attribute parameter value in the set of function attribute parameters and its corresponding function requirement parameter; selecting candidate function parameter values from the set of function attribute parameters based on the median value selection rule and the difference; generating candidate function parameter combinations based on all candidate function parameter values; generating function parameter combinations based on the allocable function attribute parameter values; matching the parameter values in the candidate function parameter combinations with the corresponding parameter values in the function parameter combinations; selecting the function parameter combination with the most successfully matched parameter values as the network function preset parameter combination; and if the number of network function preset parameter combinations is greater than 1, calculating the error between each network function preset parameter combination and the function requirement parameter combination, and determining the slice function parameter configuration combination based on the error.
[0011] Optionally, calculating the error between the preset parameter combinations of each network function and the functional requirement parameter combinations, and determining the slice function parameter configuration combination based on the error, includes:
[0012]
[0013] Calculate the weighted average of the deviation rates corresponding to all functional requirement parameter values as the error; obtain the deviation threshold corresponding to the non-exclusive priority level; select the smallest error less than the deviation threshold from all errors as the target error, and use the network function preset parameter combination corresponding to the target error as the slice function parameter configuration combination.
[0014] Optionally, if the number of preset network function parameter combinations is 1, then this preset network function parameter combination is used as the slice function parameter configuration combination.
[0015] According to a second aspect of this disclosure, a network slice configuration apparatus is provided, comprising: an information parsing module, configured to parse network slice request information and extract slice parameter tags corresponding to the network slice request; a parameter determination module, configured to determine business scenario information, priority information, and slice function requirement parameter information based on the slice parameter tags; a parameter configuration module, configured to configure network slice function parameters according to a parameter configuration strategy and based on the business scenario information, the priority information, and the slice function requirement parameter information; and a fuzzy matching module, configured to perform fuzzy matching configuration of the network slice function parameters when it is determined that network slice function parameter configuration cannot be performed according to the parameter configuration strategy.
[0016] Optionally, the information parsing module is used to parse the network slice request information using a trained NLP model, and extract slice parameters and corresponding slice parameter labels; wherein, the slice parameter labels include: business scenario labels, network function slice parameter labels, priority auxiliary labels, and parameter type labels.
[0017] Optionally, the parameter determination module is used to obtain the service scenario information corresponding to the service scenario label; wherein the service scenario includes: eMBB, mMTC, uRLLC scenario; obtain the priority information corresponding to the priority auxiliary label; obtain the slice function requirement parameter information corresponding to the network slice function parameter label; wherein the slice function requirement parameters include: bandwidth, reliability, and latency parameters; and set corresponding type attributes for the service scenario information, the priority information, and the slice function requirement parameter information according to the parameter type label.
[0018] Optionally, the parameter configuration module is specifically used to configure network slice function parameters based on the slice function requirement parameter information if the user is determined to be a dedicated slice user based on the priority information; wherein, for the default parameters in the network slice function parameters, the preset parameter with the highest priority corresponding to the business scenario information is selected and configured based on the parameter configuration strategy; after the network slice function parameters are configured, the network slice function parameter threshold corresponding to the business scenario information is obtained based on the parameter configuration strategy; the currently configured network slice function parameter value is verified according to the network slice function parameter threshold, and corresponding processing is performed based on the verification result.
[0019] Optionally, the parameter configuration module is specifically used to: if a user is determined to be a non-dedicated slice user based on the priority information, determine the non-dedicated priority level based on the priority information; obtain the corresponding parameter increase or decrease threshold based on the parameter configuration strategy and the non-dedicated priority level; and configure the network slice function parameters based on the slice function requirement parameter information and the parameter increase or decrease threshold.
[0020] Optionally, the fuzzy matching module is used to perform fuzzy matching configuration on the network slicing function parameters for non-dedicated slice users if it is determined that network slicing function parameter configuration cannot be performed according to the parameter configuration strategy, in order to determine the slice function parameter configuration group.
[0021] Optionally, the fuzzy matching module includes: a set determination unit, configured to obtain allocable functional attribute parameter values, generate a functional attribute parameter set based on functional attribute parameter values with the same functional attributes; obtain functional requirement parameter values based on the slice functional requirement parameter information, generate functional requirement parameter combinations based on the functional requirement parameter values, and determine the functional attribute parameter set corresponding to each functional requirement parameter value; a parameter selection unit, configured to calculate the difference between each functional attribute parameter value in the functional attribute parameter set and the corresponding functional requirement parameter; select candidate functional parameter values from the functional attribute parameter set based on the intermediate value selection rule and the difference; a combination selection unit, configured to generate candidate functional parameter combinations based on all candidate functional parameter values, and generate functional parameter combinations based on the allocable functional attribute parameter values; match the parameter values in the candidate functional parameter combinations with the corresponding parameter values in the functional parameter combinations, and select the functional parameter combination with the most successfully matched parameter values as the network function preset parameter combination; an error calculation unit, configured to calculate the error between each network function preset parameter combination and the functional requirement parameter combination if the number of network function preset parameter combinations is greater than 1; and a combination determination unit, configured to determine the slice functional parameter configuration combination based on the error.
[0022] Optionally, the error calculation unit is used for
[0023] The error calculation unit calculates the weighted average of the deviation rates corresponding to all functional requirement parameter values as the error; the combination determination unit is used to obtain the deviation threshold corresponding to the non-exclusive priority level; selects the smallest error less than the deviation threshold from all errors as the target error, and uses the network function preset parameter combination corresponding to the target error as the slice function parameter configuration combination.
[0024] Optionally, the combination determination unit is configured to use the network function preset parameter combination as the slice function parameter configuration combination if the number of network function preset parameter combinations is 1.
[0025] According to a third aspect of this disclosure, a network slicing configuration apparatus is provided, comprising: a memory; and a processor coupled to the memory, the processor being configured to perform the method described above based on instructions stored in the memory.
[0026] According to a fourth aspect of this disclosure, a computer-readable storage medium is provided that stores computer instructions which are executed by a processor using the method described above.
[0027] The network slicing configuration method, apparatus, and storage medium disclosed herein can realize multi-parameter configuration in complex network slicing application scenarios. It can improve the efficiency, accuracy, and availability of slice parameter configuration for operation and maintenance personnel and end users, reduce the risk of human error, and improve the user experience. Attached Figure Description
[0028] To more clearly illustrate the technical solutions in the embodiments of this disclosure or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this disclosure. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0029] Figure 1 This is a flowchart illustrating an embodiment of the network slicing configuration method according to the present disclosure;
[0030] Figure 2 This is a schematic diagram illustrating the process of configuring a dedicated slice user according to one embodiment of the network slice configuration method of this disclosure;
[0031] Figure 3 This is a schematic diagram illustrating the process of configuring a non-exclusive shared slice user according to one embodiment of the network slice configuration method of this disclosure;
[0032] Figure 4 This is a schematic diagram illustrating the process of performing fuzzy matching on non-exclusive shared slice users in one embodiment of the network slice configuration method according to the present disclosure.
[0033] Figure 5 This is a schematic diagram illustrating an application scenario of the network slicing configuration method disclosed herein.
[0034] Figure 6This is a schematic diagram of a module of an embodiment of a network slicing configuration apparatus according to the present disclosure;
[0035] Figure 7 This is a schematic diagram of a fuzzy matching module in one embodiment of a network slicing configuration apparatus according to the present disclosure;
[0036] Figure 8 This is a schematic diagram of a module of another embodiment of a network slicing configuration apparatus according to the present disclosure. Detailed Implementation
[0037] The present disclosure will now be described more fully with reference to the accompanying drawings, which illustrate exemplary embodiments of the present disclosure. The technical solutions of the embodiments of the present disclosure will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present disclosure, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present disclosure without creative effort are within the scope of protection of the present disclosure.
[0038] Figure 1 This is a flowchart illustrating an embodiment of the network slicing configuration method according to the present disclosure, as follows: Figure 1 As shown:
[0039] Step 101: Parse the network slice request information and extract the slice parameter tags corresponding to the network slice request.
[0040] In one embodiment, a user can input network slice request information through a UI interface, and the trained NLP model processes the network slice request information to extract slice parameter labels.
[0041] Step 102: Determine business scenario information, priority information, and slice function requirement parameter information based on slice parameter labels.
[0042] Step 103: Configure network slicing function parameters according to the parameter configuration strategy and based on business scenario information, priority information, and slicing function requirement parameter information. The parameter configuration strategy can be preset and includes multiple configuration strategies.
[0043] Step 104: If it is determined that network slicing function parameters cannot be configured according to the parameter configuration strategy, then fuzzy matching is performed on the network slicing function parameters. Various fuzzy matching algorithms can be used for fuzzy matching configuration of the network slicing function parameters.
[0044] In one embodiment, a trained NLP (Natural Language Processing) model is used to parse network slice request information, identify network slice request information, label network slice request information, and extract slice parameters and corresponding slice parameter labels.
[0045] Slice parameter information includes business scenario information, priority information, and slice functional requirement parameters. Slice parameter tags include business scenario tags, network function slice parameter tags, priority auxiliary tags, and parameter type tags.
[0046] The network slice configuration method disclosed herein provides a network slice configuration method based on the use of NLP models for auxiliary parameter extraction, and can provide a strategy matching method for multi-parameter configuration of slices.
[0047] Retrieve business scenario information corresponding to the business scenario tag. Business scenarios include eMBB, mMTC, uRLLC, etc. Retrieve priority information corresponding to the priority auxiliary tag. Retrieve slice function requirement parameter information corresponding to the network slice function parameter tag. Slice function requirement parameters include bandwidth, reliability, latency, etc. Set the corresponding type attributes for the business scenario information, priority information, and slice function requirement parameter information according to the parameter type tag.
[0048] For example, to create an NLP model, you train it using a large number of training samples, calibrate key parameters, and train a usable NLP model. A trained NLP model can then extract slice parameter labels along multiple dimensions.
[0049] Business scenario tags: Tags corresponding to the three major 5G business scenarios, namely eMBB, mMTC, and uRLLC;
[0050] Network Function Slice Parameter Tags: Network function slice parameter tags include tags such as bandwidth, reliability, and latency parameters;
[0051] Priority auxiliary labels: used for subsequent "strategy matching"; priority auxiliary labels can also be sent by the NLP model to the backend system with information such as username, and the priority auxiliary labels returned by the backend system.
[0052] Parameter type label: Used for subsequent parameter threshold selection, indicating the parameter type attribute, including numeric, option, and text types. For example, business scenario information can be labeled as option type, and slice function requirement parameter information can be labeled as numeric type.
[0053] Figure 2This is a schematic diagram illustrating the process of configuring a dedicated slice user according to one embodiment of the network slice configuration method of this disclosure, such as... Figure 2 As shown:
[0054] Step 201: If the user is determined to be a dedicated slice user based on the priority information, then the network slice function parameters are configured based on the slice function requirement parameter information.
[0055] In one embodiment, for the default parameters in the network slicing function parameters, the preset parameters with the highest priority corresponding to the business scenario information are selected and configured based on the parameter configuration strategy.
[0056] Step 202: After the network slicing function parameters are configured, obtain the network slicing function parameter thresholds corresponding to the business scenario information based on the parameter configuration strategy.
[0057] Step 203: Verify the currently configured network slicing function parameter values according to the network slicing function parameter threshold, and perform corresponding processing based on the verification results.
[0058] For example, the priority information corresponding to the priority auxiliary label is obtained as priority A, where priority A is the priority possessed by a dedicated slice user. Based on priority A, user 1 is determined to be a dedicated slice user, and the network slice function parameters are configured directly based on the slice function requirement parameter information.
[0059] The parameter configuration strategy presets commonly used slicing parameter strategies. For default parameters, the default high-priority preset parameters are filled in. That is, for the default parameters in the network slicing function parameters, the preset parameters with the highest priority corresponding to the business scenario information (e.g., eMBB scenario) are selected and configured based on the parameter configuration strategy.
[0060] After the network slicing function parameters are configured, the threshold values for the network slicing function parameters corresponding to the business scenario information (e.g., eMBB scenario) are obtained based on the parameter configuration strategy. If at least one of the parameters such as bandwidth, reliability, and latency in the network slicing function parameters is greater than the corresponding network slicing function parameter threshold, an alarm is triggered and the case is transferred to manual processing; if the values of all parameters such as bandwidth, reliability, and latency in the network slicing function parameters are less than or equal to the corresponding network slicing function parameter threshold, the slice configuration is considered successful.
[0061] Figure 3 This is a schematic diagram illustrating the process of configuring a non-exclusive shared slice user according to one embodiment of the network slice configuration method of this disclosure, as shown below. Figure 3 As shown:
[0062] Step 301: If the user is determined to be a non-dedicated slice user based on the priority information, then the non-dedicated priority level is determined based on the priority information.
[0063] Step 302: Based on the parameter configuration strategy and according to the non-exclusive priority level, obtain the corresponding parameter increase or decrease threshold.
[0064] Step 303: Configure network slicing function parameters based on slicing function requirement parameter information and parameter increase or decrease thresholds.
[0065] For example, the priority information corresponding to multiple priority auxiliary tags is obtained as follows: Priority B, Priority C, Priority D, and Priority E. Priority B, Priority C, Priority D, and Priority E are the priorities possessed by users who do not have exclusive slices.
[0066] Based on priority B, User 2 is determined to be a non-dedicated slice user, and its non-dedicated priority level is determined to be B. Based on priority C, User 3 is determined to be a non-dedicated slice user, and its non-dedicated priority level is determined to be C. Based on priority D, User 4 is determined to be a non-dedicated slice user, and its non-dedicated priority level is determined to be D. Based on priority E, User 5 is determined to be a non-dedicated slice user, and its non-dedicated priority level is determined to be E.
[0067] Based on the parameter configuration strategy and according to the non-exclusive priority level, the corresponding parameter increase or decrease thresholds are obtained: the parameter increase or decrease threshold for non-exclusive priority level B is +5%, the parameter increase or decrease threshold for non-exclusive priority level C is +10%, the parameter increase or decrease threshold for non-exclusive priority level D is -5%, and the parameter increase or decrease threshold for non-exclusive priority level E is -10%.
[0068] For example, when configuring slice parameters, the bandwidth, reliability, and latency parameters in the slice function requirement information for user 2 are increased by 5% each; the bandwidth, reliability, and latency parameters in the slice function requirement information for user 3 are increased by 10% each; the bandwidth, reliability, and latency parameters in the slice function requirement information for user 4 are decreased by 5% each; and the bandwidth, reliability, and latency parameters in the slice function requirement information for user 5 are decreased by 5% each.
[0069] In one embodiment, for non-dedicated slice users, if it is determined that network slice function parameters cannot be configured according to the parameter configuration policy, then fuzzy matching is performed on the network slice function parameters to determine the slice function parameter configuration group.
[0070] For example, obtain the network slicing function parameter threshold corresponding to the non-dedicated priority level B. If, after configuring the network slicing function parameters, at least one of the parameter values such as bandwidth, reliability, and latency configured for user 2 is greater than the corresponding network slicing function parameter threshold, it is determined that network slicing function parameter configuration cannot be performed according to the parameter configuration strategy, and then fuzzy matching configuration is performed on the network slicing function parameters.
[0071] Figure 4 This is a schematic diagram illustrating the process of fuzzy matching for non-exclusive shared slice users in one embodiment of the network slice configuration method according to this disclosure, as shown below. Figure 4 As shown:
[0072] Step 401: Obtain the assignable functional attribute parameter values, and generate a functional attribute parameter set based on the functional attribute parameter values with the same functional attributes.
[0073] Step 402: Obtain functional requirement parameter values based on slice functional requirement parameter information, generate functional requirement parameter combinations based on functional requirement parameter values, and determine the set of functional attribute parameters corresponding to each functional requirement parameter value.
[0074] Step 403: Calculate the difference between each functional attribute parameter value in the functional attribute parameter set and the corresponding functional requirement parameter.
[0075] Step 404: Select candidate parameter values for functions from the set of function attribute parameters based on the intermediate value selection rule and the difference.
[0076] Step 405: Generate a combination of candidate function parameters based on all candidate function parameter values, and generate a combination of function parameters based on the allocable function attribute parameter values.
[0077] Step 406: Match the parameter values in the candidate parameter combinations with the corresponding parameter values in the function parameter combinations, and select the function parameter combination with the most successfully matched parameter values as the preset parameter combination for network functions.
[0078] In one embodiment, if the number of network function preset parameter combinations is 1, then this network function preset parameter combination is used as the slice function parameter configuration combination. If the number of network function preset parameter combinations is greater than 1, then the error between each network function preset parameter combination and the function requirement parameter combination is calculated, and the slice function parameter configuration combination is determined based on the error.
[0079]
[0080] Calculate the weighted average of the deviation rates corresponding to all functional requirement parameter values as the error, and obtain the deviation threshold corresponding to the non-exclusive priority level. Select the error that is smaller than the deviation threshold and the smallest among all errors as the target error, and use the network function preset parameter combination corresponding to the target error as the slice function parameter configuration combination.
[0081] For example, for special slice parameters that cannot be matched according to the preset common slice parameter strategy group, a fuzzy matching algorithm should be used to select the preset common slice parameter strategy group that can be matched the nearest one within the allowable deviation range.
[0082] Obtain allocable functional attribute parameter values, categorize and aggregate them according to functional attributes, and generate three functional attribute parameter sets based on functional attribute parameter values with the same functional attributes: bandwidth set A = {10M, 100M, 1000M}, reliability set B = {99%, 99.99%}, and latency set C = {10ms, 100ms}, etc., where 10M, 100M, 1000M, 99%, 99.99%, 10ms, 100ms, etc. are all allocable functional attribute parameter values.
[0083] Multiple functional parameter combinations are generated based on the assignable functional attribute parameter values. For example, multiple functional parameter groups are: (A1, B1, C1) = {10M, 99%, 10ms}, (A2, B2, C2) = {1000M, 99.99%, 10ms}, (A3, B3, C3) = {1000M, 99%, 100ms}, etc.
[0084] The functional requirement parameter values are obtained based on the slice functional requirement parameter information, and the functional requirement parameter combination (A0, B0, C0) is generated based on the functional requirement parameter values. For example, the user's functional requirement parameter combination is (A0, B0, C0) = {500M, 99.9%, 50ms}, where the functional requirement parameter values are 500M bandwidth, 99.9% reliability, and 50ms latency, respectively.
[0085] Functional attribute parameter (bandwidth) set A c If there are two different parameters in {A1, A2, A3} = {5M, 1000M, 1000M}, then calculate the set of functional attribute parameters A. c The difference between each functional attribute parameter value and the corresponding functional requirement parameter A0 is used to select the appropriate value from the functional attribute parameter set A0 according to the median value selection rule. c Select candidate parameter values for the function.
[0086] For example, the function candidate parameter value A' = middle(A1-A0, A2-A0, A3-A0) = {-490M, 500M, 500M}, where middle() is the function to take the middle value, and A' is the parameter 1000M corresponding to A2 and A3. There are two optional parameters A2 and A3 for parameter A.
[0087] Based on the same method, the functional candidate parameter values B' = middle(B1-B0, B2-B0, B3-B0) and C' = middle(C1-C0, C2-C0, C3-C0), that is, A' takes 1000M to correspond to A2 and A3, B' takes 0.09% to correspond to B2, and C' takes 10ms to correspond to C1 and C2.
[0088] Generate candidate parameter combinations based on all candidate parameter values, for example, (A', B', C'). Match the parameter values in the candidate parameter combination (A', B', C') with the corresponding parameter values in the function parameter combinations (A1, B1, C1) = {10M, 99%, 10ms}, (A2, B2, C2) = {1000M, 99.99%, 10ms}, (A3, B3, C3) = {1000M, 99%, 100ms}, etc. Select the function parameter combination with the most successful matches as the network function preset parameter combination, that is, the function parameter combination (A2, B2, C2) = {1000M, 99.99%, 10ms} is the network function preset parameter combination.
[0089] If the number of preset parameter combinations for network functions is greater than one, the group with the smallest deviation rate p is selected using an error verification method. For example, there are two preset parameter combinations for network functions: (A2, B2, C2) and (A3, B3, C3).
[0090] Error verification is performed by checking the error deviation rate as a percentage. Taking the network function preset parameter combination (A2, B2, C2) as an example, for the functional requirement parameter A0, the corresponding candidate functional parameter value is selected as A2. The deviation rate is p1 = (A2 - A0) / the absolute value of A0, where p' is a deviation threshold set based on operational experience and corresponding to the non-dedicated priority level.
[0091] The deviation rates corresponding to all functional requirement parameter values are weighted and averaged to form the error p = (p1 + p2 + p3) / 3. If the error p is greater than the deviation threshold p', an alarm is triggered and manual processing is initiated.
[0092] Among all errors corresponding to the network function preset parameter combinations (A2, B2, C2) and (A3, B3, C3), the smallest error that is less than the deviation threshold is selected as the target error. The network function preset parameter combination corresponding to the target error is then used as the slice function parameter configuration combination. For example, if the error of the network function preset parameter combination (A2, B2, C2) is small and less than the deviation threshold, then the slice function parameter configuration combination is the network function preset parameter combination (A2, B2, C2).
[0093] In one embodiment, such as Figure 5 As shown, the UI module includes a form for configuring manual network slice parameters and a form for text or voice input. Users can input network slice request information through the UI module. The UI module forwards the user-input text or voice (network slice request information) to the NLP Model for processing via the UI-Server. The NLP Model then parses the network slice request information, extracting the slice parameters and corresponding slice parameter tags. The network slice configuration method described in any of the above embodiments is executed in the UI-Server. The network slice function parameters are configured using this method, and the slices are issued after manual confirmation of the network slice function parameters.
[0094] The network slice configuration method in the above embodiments can realize multi-parameter configuration in complex network slice application scenarios, which can improve the efficiency and accuracy of slice parameter configuration for operation and maintenance personnel and end users, and reduce the risk of human error.
[0095] In one embodiment, such as Figure 6 As shown, this disclosure provides a network slicing configuration device 60, including an information parsing module 61, a parameter determination module 62, a parameter configuration module 63, and a fuzzy matching module 64. The information parsing module 61 parses the network slicing request information and extracts the slice parameter tags corresponding to the network slicing request. The parameter determination module 62 determines business scenario information, priority information, and slice functional requirement parameter information based on the slice parameter tags.
[0096] The parameter configuration module 63 configures network slicing function parameters according to the parameter configuration strategy and based on business scenario information, priority information, and slicing function requirement parameter information. The fuzzy matching module 64 performs fuzzy matching configuration on the network slicing function parameters when it determines that network slicing function parameter configuration cannot be performed according to the parameter configuration strategy.
[0097] In one embodiment, the information parsing module 61 uses a trained NLP model to parse the network slice request information and extract slice parameters and corresponding slice parameter labels; wherein, the slice parameter labels include business scenario labels, network function slice parameter labels, priority auxiliary labels, and parameter type labels.
[0098] The parameter determination module 62 acquires the business scenario information corresponding to the business scenario tag; the business scenarios include eMBB, mMTC, uRLLC, etc. The parameter determination module 62 acquires the priority information corresponding to the priority auxiliary tag, and acquires the slice function requirement parameter information corresponding to the network slice function parameter tag; the slice function requirement parameters include bandwidth, reliability, latency, etc. The parameter determination module 62 sets the corresponding type attributes for the business scenario information, priority information, and slice function requirement parameter information according to the parameter type tag.
[0099] If the parameter configuration module 63 determines that a user is a dedicated slice user based on priority information, it configures the network slice function parameters based on the slice function requirement parameter information. For default parameters in the network slice function parameters, the module selects the preset parameter with the highest priority corresponding to the business scenario information for configuration based on the parameter configuration strategy. After configuring the network slice function parameters, the parameter configuration module 63 obtains the network slice function parameter threshold corresponding to the business scenario information based on the parameter configuration strategy. The parameter configuration module 63 verifies the currently configured network slice function parameter values according to the network slice function parameter thresholds and performs corresponding processing based on the verification results.
[0100] If the parameter configuration module 63 determines that a user is a non-dedicated slice user based on priority information, it determines the non-dedicated priority level based on the priority information. Based on the parameter configuration strategy and the non-dedicated priority level, the parameter configuration module 63 obtains the corresponding parameter increase or decrease thresholds. The parameter configuration module 63 configures the network slicing function parameters based on the slice function requirement parameter information and the parameter increase or decrease thresholds.
[0101] In one embodiment, if the fuzzy matching module 64 determines that network slicing function parameters cannot be configured according to the parameter configuration policy for non-dedicated slice users, it performs fuzzy matching configuration on the network slicing function parameters to determine the slice function parameter configuration group.
[0102] like Figure 7As shown, the fuzzy matching module 64 includes: a set determination unit 641, a parameter selection unit 642, a combination selection unit 643, an error calculation unit 644, and a combination determination unit 645. The set determination unit 641 acquires allocable functional attribute parameter values and generates a set of functional attribute parameters based on the functional attribute parameter values with the same functional attributes. The set determination unit 641 obtains functional requirement parameter values based on sliced functional requirement parameter information, generates functional requirement parameter combinations based on the functional requirement parameter values, and determines the set of functional attribute parameters corresponding to each functional requirement parameter value.
[0103] The parameter selection unit 642 calculates the difference between each functional attribute parameter value in the functional attribute parameter set and the corresponding functional requirement parameter. Based on the intermediate value selection rule and the difference, the parameter selection unit 642 selects candidate functional parameter values from the functional attribute parameter set.
[0104] The combination selection unit 643 generates a combination of candidate function parameters based on all candidate function parameter values, and generates a combination of function parameters based on the allocable function attribute parameter values. The combination selection unit 643 matches the parameter values in the candidate function parameter combinations with the corresponding parameter values in the function parameter combinations, and selects the function parameter combination with the most successfully matched parameter values as the preset parameter combination for network functions.
[0105] If the number of network function preset parameter combinations is 1, the error calculation unit 644 uses this network function preset parameter combination as the slice function parameter configuration combination. If the number of network function preset parameter combinations is greater than 1, the error calculation unit 644 calculates the error between each network function preset parameter combination and the function requirement parameter combination, and determines the slice function parameter configuration combination based on the error.
[0106]
[0107] Error calculation unit 644 calculates the weighted average of the deviation rates corresponding to all functional requirement parameter values, as the target error. Combination determination unit 645 obtains the deviation threshold corresponding to the non-exclusive priority level, selects the smallest error less than the deviation threshold from all errors as the target error, and combines the network function preset parameters corresponding to the target error as the slice function parameter configuration combination.
[0108] In one embodiment, such as Figure 8 As shown, this disclosure provides a network slicing configuration apparatus, including a memory 81, a processor 82, a communication interface 83, and a bus 84. The memory 81 is used to store instructions, and the processor 82 is coupled to the memory 81. The processor 82 is configured to execute the network slicing configuration method described above based on the instructions stored in the memory 81.
[0109] The memory 81 can be a high-speed RAM, non-volatile memory, or a memory array. The memory 81 may also be divided into blocks, and these blocks can be combined into virtual volumes according to certain rules. The processor 82 can be a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the network slicing configuration method of this disclosure.
[0110] In one embodiment, this disclosure provides a computer-readable storage medium storing computer instructions that, when executed by a processor, implement the method as described in any of the above embodiments.
[0111] The network slicing configuration method, apparatus, and storage medium in the above embodiments can realize multi-parameter configuration in complex network slicing application scenarios. They can improve the efficiency, accuracy, and availability of slice parameter configuration for operation and maintenance personnel and end users, reduce the risk of human error, and improve the user experience.
[0112] The methods and systems of this disclosure may be implemented in many ways. For example, they may be implemented by software, hardware, firmware, or any combination of software, hardware, and firmware. The above-described order of steps for the methods is for illustrative purposes only, and the steps of the methods of this disclosure are not limited to the order specifically described above unless otherwise specifically stated. Furthermore, in some embodiments, this disclosure may also be implemented as a program recorded on a recording medium, the program including machine-readable instructions for implementing the methods according to this disclosure. Thus, this disclosure also covers recording media storing programs for performing the methods according to this disclosure.
[0113] The description in this disclosure is provided for illustrative and descriptive purposes only and is not intended to be exhaustive or to limit the disclosure to its forms. Many modifications and variations will be apparent to those skilled in the art. The embodiments were chosen and described in order to better illustrate the principles and practical application of this disclosure and to enable those skilled in the art to understand this disclosure and to design various embodiments with various modifications suitable for a particular purpose.
Claims
1. A network slicing configuration method, comprising: Parse the network slice request information and extract the slice parameter tags corresponding to the network slice request; Based on the slice parameter labels, business scenario information, priority information, and slice function requirement parameter information are determined; Configure network slicing function parameters according to the parameter configuration strategy and based on the business scenario information, the priority information, and the slicing function requirement parameter information; For non-dedicated slice users, when it is determined that network slicing function parameters cannot be configured according to the aforementioned parameter configuration strategy, fuzzy matching configuration is performed on the network slicing function parameters, including: Get the assignable functional attribute parameter values, and generate a set of functional attribute parameters based on the functional attribute parameter values with the same functional attributes; Based on the sliced functional requirement parameter information, functional requirement parameter values are obtained, functional requirement parameter combinations are generated based on the functional requirement parameter values, and a set of functional attribute parameters corresponding to each functional requirement parameter value is determined. Calculate the difference between each functional attribute parameter value in the set of functional attribute parameters and the corresponding functional requirement parameter; Based on the median value selection rule and the difference, select candidate function parameter values from the set of function attribute parameters; Generate a combination of candidate function parameters based on all candidate function parameter values, and generate a combination of function parameters based on the allocable function attribute parameter values. The parameter values in the candidate parameter combinations are matched with the corresponding parameter values in the function parameter combinations, and the function parameter combination with the most successful matching parameter values is selected as the preset parameter combination for network functions.
2. The method as described in claim 1, wherein parsing the network slice request information and extracting the slice parameter tags corresponding to the network slice request includes: The trained NLP model is used to parse the network slice request information and extract the slice parameters and corresponding slice parameter labels. The slice parameter labels include: business scenario labels, network function slice parameter labels, priority auxiliary labels, and parameter type labels.
3. The method as described in claim 2, wherein determining the business scenario information, priority information, and slice functional requirement parameter information based on the slice parameter labels includes: Obtain the business scenario information corresponding to the business scenario tag; wherein, the business scenarios include: eMBB, mMTC, and uRLLC scenarios; Obtain the priority information corresponding to the priority auxiliary label; Obtain the slice function requirement parameter information corresponding to the network slice function parameter label; wherein, the slice function requirement parameters include: bandwidth, reliability, and latency parameters; The corresponding type attributes are set for the business scenario information, the priority information, and the slice function requirement parameter information according to the parameter type label.
4. The method as described in claim 3, wherein configuring network slicing function parameters according to the parameter configuration strategy and based on the business scenario information, the priority information, and the slicing function requirement parameter information includes: If the user is determined to be a dedicated slice user based on the priority information, then the network slice function parameters are configured based on the slice function requirement parameter information; wherein, for the default parameters in the network slice function parameters, the preset parameters with the highest priority corresponding to the business scenario information are selected and configured based on the parameter configuration strategy. After the network slicing function parameters are configured, the network slicing function parameter thresholds corresponding to the business scenario information are obtained based on the parameter configuration strategy. The network slicing function parameter values are validated based on the network slicing function parameter threshold, and corresponding processing is performed based on the validation results.
5. The method as described in claim 3, wherein configuring network slicing function parameters according to the parameter configuration strategy and based on the business scenario information, the priority information, and the slicing function requirement parameter information includes: If a user is determined to be a non-dedicated slice user based on the priority information, then the non-dedicated priority level is determined based on the priority information. Based on the parameter configuration strategy and according to the non-exclusive priority level, obtain the corresponding parameter increase or decrease threshold; The network slicing function parameters are configured based on the slicing function requirement parameters and the threshold for increasing or decreasing the parameters.
6. The method of claim 5, further comprising: If the number of preset parameter combinations for network functions is greater than 1, then the error between each preset parameter combination for network functions and the combination of function requirements parameters is calculated, and the configuration combination of slice function parameters is determined based on the error.
7. The method as described in claim 6, wherein calculating the error between each network function preset parameter combination and the function requirement parameter combination, and determining the slice function parameter configuration combination based on the error, comprises: Calculate the deviation rate between the candidate function parameter values in the preset parameter combination of the network function and the combination of function requirement parameters. = Calculate the weighted average of the deviation rates corresponding to all functional requirement parameter values, as the error; Obtain the deviation threshold corresponding to the non-exclusive priority level; The smallest error among all errors, which is less than the deviation threshold, is selected as the target error, and the network function preset parameter combination corresponding to the target error is used as the slice function parameter configuration combination.
8. The method of claim 6, further comprising: If the number of preset parameter combinations for network functions is 1, then this preset parameter combination for network functions is used as the configuration combination for slice function parameters.
9. A network slicing configuration device, comprising: The information parsing module is used to parse network slice request information and extract slice parameter tags corresponding to the network slice request; The parameter determination module is used to determine business scenario information, priority information, and slice function requirement parameter information based on the slice parameter labels. The parameter configuration module is used to configure network slicing function parameters according to the parameter configuration strategy and based on the business scenario information, the priority information and the slicing function requirement parameter information; The fuzzy matching module is used to perform fuzzy matching configuration of network slicing function parameters for non-dedicated slice users when it is determined that the network slicing function parameters cannot be configured according to the parameter configuration strategy. The fuzzy matching module includes: The set determination unit is used to obtain allocable functional attribute parameter values, generate a functional attribute parameter set based on functional attribute parameter values with the same functional attributes, obtain functional requirement parameter values based on the sliced functional requirement parameter information, generate functional requirement parameter combinations based on the functional requirement parameter values, and determine the functional attribute parameter set corresponding to each functional requirement parameter value. The parameter selection unit is used to calculate the difference between each functional attribute parameter value in the functional attribute parameter set and the corresponding functional requirement parameter; and to select candidate functional parameter values from the functional attribute parameter set based on the intermediate value selection rule and the difference. The combination selection unit is used to generate a combination of functional candidate parameters based on all functional candidate parameter values, and to generate a combination of functional parameters based on the allocable functional attribute parameter values; the parameter values in the combination of functional candidate parameters are matched with the corresponding parameter values in the combination of functional parameters, and the combination of functional parameters with the most successfully matched parameter values is selected as the preset parameter combination of network functions.
10. The apparatus of claim 9, wherein, The information parsing module is used to parse the network slice request information using a trained NLP model, and extract slice parameters and corresponding slice parameter labels; wherein, the slice parameter labels include: business scenario labels, network function slice parameter labels, priority auxiliary labels, and parameter type labels.
11. The apparatus of claim 10, wherein, The parameter determination module is used to obtain the service scenario information corresponding to the service scenario label; wherein the service scenario includes: eMBB, mMTC, uRLLC scenario; obtain the priority information corresponding to the priority auxiliary label; obtain the slice function requirement parameter information corresponding to the network slice function parameter label; wherein the slice function requirement parameters include: bandwidth, reliability, and latency parameters; and set corresponding type attributes for the service scenario information, the priority information, and the slice function requirement parameter information according to the parameter type label.
12. The apparatus of claim 11, wherein, The parameter configuration module is specifically used to configure network slice function parameters based on the slice function requirement parameter information if the user is determined to be a dedicated slice user based on the priority information; wherein, for the default parameters in the network slice function parameters, the preset parameter with the highest priority corresponding to the business scenario information is selected and configured based on the parameter configuration strategy; after the network slice function parameters are configured, the network slice function parameter threshold corresponding to the business scenario information is obtained based on the parameter configuration strategy; the currently configured network slice function parameter value is verified according to the network slice function parameter threshold, and corresponding processing is performed based on the verification result.
13. The apparatus of claim 11, wherein, The parameter configuration module is specifically used to determine the non-dedicated slice priority level based on the priority information if the user is determined to be a non-dedicated slice user based on the priority information; to obtain the corresponding parameter increase or decrease threshold based on the parameter configuration strategy and the non-dedicated slice priority level; and to configure the network slice function parameters based on the slice function requirement parameter information and the parameter increase or decrease threshold.
14. The apparatus of claim 13, wherein, The fuzzy matching module includes: An error calculation unit is used to calculate the error between each network function preset parameter combination and the function requirement parameter combination if the number of network function preset parameter combinations is greater than 1. A combination determination unit is used to determine the combination of slice function parameter configurations based on the error.
15. The apparatus of claim 14, wherein, The error calculation unit is used to calculate the deviation rate between the candidate function parameter values in the preset parameter combination of the network function and the combination of function requirement parameters. = The error calculation unit calculates the weighted average of the deviation rates corresponding to all functional requirement parameter values, as the error. The combination determination unit is used to obtain the deviation threshold corresponding to the non-exclusive priority level; select the smallest error less than the deviation threshold from all errors as the target error, and use the network function preset parameter combination corresponding to the target error as the slice function parameter configuration combination.
16. The apparatus of claim 14, wherein, The combination determination unit is used to use the network function preset parameter combination as the slice function parameter configuration combination if the number of network function preset parameter combinations is 1.
17. A network slicing configuration apparatus, comprising: Memory; And a processor coupled to the memory, the processor being configured to perform the method as described in any one of claims 1 to 8 based on instructions stored in the memory.
18. A computer-readable storage medium storing computer instructions that are executed by a processor according to any one of claims 1 to 8.