A network slice access control method and system based on resource efficiency and auction

Abstract

The present invention provides a network slice access control method and system based on resource efficiency and auction, belonging to the field of communication technology, the method includes counting the total number of slice requests of various types received in the current time slot, and initializing the temporary access decision vector between slices in the current time slot to be a 0 vector; based on each slice, calculating the cumulative access rate under the current cumulative request amount and cumulative access amount, determining whether the cumulative access rate meets the priority requirements of different slices, and determining the order of trying to access different slices according to the dominant resource efficiency under the current available resources, and obtaining the inter-slice access decision; according to the determined inter-slice access decision, using a single item auction mechanism to allocate access quotas within the slice, and completing the control of network slice access. The present invention solves the optimal problem of network slice access control in the scenario where multiple network slice providers and multiple upper-layer service providers coexist.

Description

Technical Field

[0001] The present invention belongs to the field of communication technology, and in particular, to a network slice access control method and system based on resource efficiency and auction. Background Art

[0002] With the development and evolution of 5G networks, network slicing has become the basic carrier for mobile operators to provide specific services. According to the different service quality requirements of different services, operators need to provide different quantities of various physical resources for the corresponding slices to ensure that the actual service quality obtained by users meets the service level agreement (Service Level Agreement, SLA) and obtain benefits from the services they provide.

[0003] Operators use network slicing technology to abstract the physical networks under their jurisdiction into logically independent virtual networks, thereby providing a variety of slice services. They are important participants in the 5G market and are also called slice providers. Thanks to the flexibility of slicing technology, each slice provider can easily build some similar or functionally interchangeable slice entities to provide services to one or several types of businesses. However, the types of network equipment held by different slice providers are often different, and the main deployment locations of some equipment are usually different, so that in addition to supporting overlapping slice types, they can also provide personalized slice services that are more in line with their own network characteristics.

[0004] The upper-layer service provider is another participant in the 5G market. They are responsible for providing various vertical services directly to users. Since the upper-layer service providers do not control the physical network, they need to lease slices that can support their services from slice providers. If multiple slice providers can meet the needs of the upper-layer service providers, rational upper-layer service providers will compare the slice providers and make slice leasing decisions that are more favorable to them.

[0005] After receiving a series of slice lease requests, the slice provider also needs to decide which slice request to accept and how many slices to accept based on its limited physical resource capacity. In contrast, the issues that slice providers need to consider when making decisions are more complicated. First, slices are customized designs for services. They naturally inherit the priority attributes of services, which leads to a hierarchical relationship between slices. The allocation of access quotas between different slices needs to respect this hierarchical difference. Second, for the same slice, the slice provider may receive lease requests from multiple upper-level service providers, and the total lease quota of the slice needs to be redistributed among them. When pursuing maximum benefits, slice providers must consider whether their access behavior violates these restrictions, because once the hierarchical relationship between slices is broken or the allocation within the slice is unfair, it may damage their reputation with some upper-level service providers, causing them to flow to other slice providers. Summary of the invention

[0006] In view of the above-mentioned deficiencies in the prior art, the present invention provides a network slice access control method and system based on resource efficiency and auction, which realizes the optimal problem of network slice access control in a scenario where multiple network slice providers and multiple upper-layer service providers coexist.

[0007] In order to achieve the above purpose, the technical solution adopted by the present invention is:

[0008] This solution provides a network slice access control method based on resource efficiency and auction, including the following steps:

[0009] S1, initialization phase: count the total number of slice requests received in the current time slot, and initialize the temporary access decision vector between slices in the current time slot to 0 vector;

[0010] S2, inter-slice access decision stage: based on each slice, calculate the cumulative access rate under the current cumulative request volume and cumulative access volume, and determine the order of trying to access different slices according to the dominant resource efficiency of different slices under the current available resources, wherein the cumulative access volume includes the historical cumulative access volume and the current temporary access volume;

[0011] S3, Intra-slice access quota allocation stage: Based on the determined inter-slice access decision, the single item auction mechanism is used to allocate access quotas within the slice to complete access control of the network slice.

[0012] The beneficial effect of the present invention is that the present invention intends to design an access control method for maximizing long-term average benefits for slice providers in the scenario of "multiple slice providers and multiple upper-layer service providers". The method quickly determines the access decision between slices based on the dominant resource efficiency of the slices, and maintains the priority difference between slices; then an auction mechanism is designed to allocate access quotas within the slices, maximizing the fairness of the valuation weighted ratio within the slices, thereby maintaining a stable group of slice tenants for the slice provider, and realizing the optimal problem of network slice access control in the scenario of coexistence of multiple network slice providers and multiple upper-layer service providers.

[0013] Furthermore, the step S2 comprises the following steps:

[0014] S201. Calculate the cumulative access rate under the current cumulative request volume and cumulative access volume based on each slice;

[0015] S202, determine whether the cumulative access rate meets the priority requirements of different slices, if so, proceed to step S204, otherwise, proceed to step S203;

[0016] S203, select the slice types that do not meet the priority requirements, calculate their dominant resource efficiency under the currently available resources, and access the slice requests in the order of the dominant resource efficiency from high to low;

[0017] If the currently available resources allow access to a certain slice, and the total demand for such slices is greater than the current temporary access quantity, the temporary access quantity of the slice is increased by 1, and the process returns to step S201. If the currently available resources do not allow access to any slice, or the temporary access quantity of the slice is equal to the total demand, the temporary access decision is used as the final access decision for the time slot, the inter-slice access process ends, and the process goes to step S3.

[0018] S204. Calculate the dominant resource efficiency of all slices under the currently available resources, and access the slice requests in the order of the dominant resource efficiency from high to low, and set the feasible condition as follows: the total demand for this type of slice is greater than the current temporary access amount, and after the temporary access amount of this type of slice is increased by 1, its new cumulative access rate meets the priority condition;

[0019] If the currently available resources allow access to a certain slice, and there is a slice that meets the feasible conditions, the temporary access amount of this type of slice is increased by 1, and the process returns to step S201; if there is no slice that meets the feasible conditions, the temporary access decision is used as the final access decision for the time slot, the current inter-slice access process is terminated, and the process proceeds to step S3.

[0020] The beneficial effect of the above further scheme is that the further scheme can make full use of limited resources to improve the efficiency of network slice providers and ensure the priority conditions between slices as much as possible.

[0021] Furthermore, step S3 includes the following steps:

[0022] S301. According to the determined inter-slice access decision, determine whether there is intra-slice contention. If so, proceed to step S302; otherwise, proceed to step S303.

[0023] S302. Use a single item auction mechanism to redistribute the access quotas of the competing slices to complete the control of network slice access.

[0024] S303. Output the inter-slice access decision as the final access plan of the access process to complete the control of network slice access.

[0025] The beneficial effect of the above further scheme is that the further scheme can ensure that each upper-layer service provider reports the valuation of the slice truthfully, and allocates quotas within the slice quickly, efficiently and fairly based on the true valuation.

[0026] Furthermore, the step S302 includes the following steps:

[0027] S3021. Obtain slice valuation, and based on the slice valuation, remove upper-layer service providers whose access quota is less than the slice basic threshold, and calculate the increment generated by the following formula when one slice access quota is added to the remaining upper-layer service providers, and record the access quota a o,n The demand q of the upper service provider is gradually increased from 1 to o,n All incremental data generated when:

[0028]

[0029] Among them, a o,n represents the number of access quotas obtained by the upper layer service provider o from the network slice provider n, q o,n represents the slice access quota demand of the upper-layer service provider, o represents the upper-layer service provider, O s Indicates the set of upper-layer service providers supported by slice s, v o represents the true estimate of the slice by the upper layer service provider o, ε represents the positive bias, and log(·) represents the natural logarithmic function;

[0030] S3022. Arrange the incremental data in descending order using the following constraints, and select the sorted incremental data in sequence until the cumulative number of selected incremental data is equal to the total number of access quotas for the slice, thereby allocating access quotas within the slice and completing the control of network slice access:

[0031]

[0032]

[0033]

[0034]

[0035] Among them, o represents the upper layer service provider, O s represents the set of upper-layer service providers that provide services for slice s, a o,n represents the number of access quotas obtained by the upper layer service provider o from the network slice provider n, a n,s represents the total number of accesses to s-type slices from network slice provider n, Represents a set of integers.

[0036] The beneficial effect of the above further scheme is that the further scheme can ensure that each upper-layer service provider reports the valuation of the slice truthfully, and allocates quotas within the slice quickly, efficiently and fairly based on the true valuation.

[0037] The present invention provides a network slice access control system based on resource efficiency and auction, comprising:

[0038] The first processing module is used to count the total number of slice requests of various types received in the current time slot, and initialize the temporary access decision vector between slices in the current time slot to a 0 vector;

[0039] A second processing module is used to calculate, based on each slice, a cumulative access rate under a current cumulative request amount and a cumulative access amount, and determine an order of trying to access different slices according to dominant resource efficiencies of different slices under currently available resources, wherein the cumulative access amount includes a historical cumulative access amount and a current temporary access amount;

[0040] The third processing module is used to determine whether to redistribute the access quota of the competing slices based on the determined inter-slice access decision by using the single-item auction mechanism to complete the control of network slice access.

[0041] The beneficial effect of the present invention is that the present invention intends to design an access control method for maximizing long-term average benefits for slice providers in the scenario of "multiple slice providers and multiple upper-layer service providers". The method quickly determines the access decision between slices based on the dominant resource efficiency of the slices, and maintains the priority difference between slices; then an auction mechanism is designed to allocate access quotas within the slices, maximizing the fairness of the valuation weighted ratio within the slices, thereby maintaining a stable group of slice tenants for the slice provider, and realizing the optimal problem of network slice access control in the scenario of coexistence of multiple network slice providers and multiple upper-layer service providers. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1The figure is a flow chart of the method of the present invention.

[0043] Figure 2 This is a performance comparison chart of the present invention and three existing methods in terms of long-term average basic benefits in this embodiment.

[0044] Figure 3 The figure is a comparison chart of the performance of the present invention and three existing methods in terms of long-term average real benefits in this embodiment.

[0045] Figure 4 Schematic diagram of fairness indicators among slices of each method in this embodiment.

[0046] Figure 5 Schematic diagram of the valuation weighted ratio fairness index of the VWPFA mechanism in this embodiment.

[0047] Figure 6 It is a schematic diagram of the system structure of the present invention. DETAILED DESCRIPTION

[0048] The specific implementation modes of the present invention are described below so that those skilled in the art can understand the present invention. However, it should be clear that the present invention is not limited to the scope of the specific implementation modes. For those of ordinary skill in the art, as long as various changes are within the spirit and scope of the present invention as defined and determined by the attached claims, these changes are obvious, and all inventions and creations utilizing the concept of the present invention are protected.

[0049] Example

[0050] Before describing the present invention, the following concepts are described.

[0051] 1. User service request: User service request is the demand initiated by the end user to communicate with the service entity to the upper layer service provider. Different types of user service requests focus on different service indicators, such as latency, bandwidth, packet loss rate, jitter, etc. The upper layer service provider needs to rent the corresponding network slice from the network slice provider to provide end users with services that meet their performance requirements.

[0052] 2. Heterogeneous service request queues: When a service request initiated by a user reaches the corresponding upper-layer service provider, it may not be immediately approved for service. Instead, it first enters the service queue maintained by the upper-layer service provider and waits for the service notification of the upper-layer service provider. Since different upper-layer service providers support different types of services, the queues they maintain are also heterogeneous. The upper-layer service provider provides service opportunities for user requests in its queue in order according to the principle of first come first served.

[0053] The service request initiated by the end user has a maximum tolerable waiting time. If the waiting time of the request in the queue exceeds the maximum tolerable waiting time, the end user will cancel the service request and leave the queue. In addition, if the user has a service demand, but finds that too many requests have accumulated in the queue of the upper-level service provider that can provide services to him, the user is likely to give up initiating the request directly, because the estimated waiting time of the request initiated in the heterogeneous queue is likely to exceed the user's tolerance limit.

[0054] 3. Slice leasing request: The upper layer service provider needs to instantiate the service network element before providing services to the end user, but they lack real physical devices and need to rent from the slice provider in units of slices, that is, initiate a slice leasing request. The present invention uses a resource demand vector to characterize the resource cost of instantiating a network slice with specific QoS requirements, and assumes that for the same slice provider, the resource overhead of instantiating the same slice remains unchanged.

[0055] After the upper-layer service provider successfully obtains the required network slice, it will immediately instantiate the service entity and provide services to the user. After the user service request is connected, it will last for several time slots, during which the network slice carrying the service will always remain active and consume a corresponding amount of physical resources until the user terminates the service. The slice provider can then destroy the slice and reclaim the physical resources occupied by the slice.

[0056] 4. Slice priority: There are natural priority differences between different services. For example, emergency communication services take precedence over ordinary data services. The bundling relationship between slices and services makes the slices inherit the priority of the services. The willingness to pay of upper-layer service providers who provide different services will also lead to preferences of slice providers when making access decisions. The traditional definition of priority requires that when two requests are received, the request with the higher priority is always accepted first. In a slice-based service model, this approach may lead to resource monopoly, which violates the concept of sharing. The present invention believes that in the new scenario, the cumulative access rate of high-priority slice requests is not lower than the cumulative access rate of low-priority slices, which means that the priority requirements are complied with.

[0057] 5. Slice fairness: The definition of slice fairness is divided into inter-slice fairness and intra-slice fairness. The present invention believes that inter-slice fairness needs to respect the priority differences of slices, that is, it is meaningful to talk about inter-slice fairness only when the cumulative reception rate of high-priority slices is not lower than the cumulative reception rate of low-priority slices. On this basis, inter-slice fairness is defined as the uniformity of the difference in cumulative reception rates of adjacent priority slices. In addition, the inter-slice fairness index can intuitively reflect whether the inter-slice priority conditions are met: if they are met, the inter-slice fairness index is non-zero, otherwise it is zero.

[0058] Intra-slice fairness considers the fairness of allocating access quotas for the same slice among multiple upper-layer service providers that are in competition or interchangeable for users. The present invention uses valuation-weighted proportional fairness to measure the fairness of intra-slice access quota allocation, where valuation refers to the upper-layer service provider's estimate of the slice's value.

[0059] 6. Dominant resource efficiency: The dominant resource of a slice refers to the resource type that is exhausted first when the currently available resources are only used to access this type of slice; the dominant resource efficiency refers to the ratio of the basic service value of the slice to its dominant resource demand in the current state. It reflects the efficiency rate of different slices when consuming the same amount of bottleneck resources.

[0060] In this embodiment, there are multiple network slice providers and multiple upper layer service providers in the 5G market envisioned by the present invention, represented by symbols N and O respectively. Each network slice provider can only support a limited number of network slices, represented by symbols S n Represents the set of slice types supported by network slice provider n. A network slice provider needs to pay a certain amount of physical resources to instantiate a network slice, which is represented by the symbol represents the resource vector that network slice provider n needs to consume to instantiate s slices, r n Represents its physical resource capacity. After a network slice provider receives a slice request, it will receive a certain amount of basic benefits per time slot as a reward for maintaining the normal operation of the slice, represented by the symbol Represents the basic benefit that slice provider n obtains from slice type s.

[0061] At the beginning of each time slot, the user group will generate a batch of service requests, and they will enter the service queues of each upper-layer service provider according to the user's wishes. The upper-layer service provider will decide the number of slice requests to be initiated to each optional slice provider according to certain strategies based on the cumulative access rate of slices of the optional slice provider, the fairness of access between slices, and other information.

[0062] After receiving the slice access request for this time slot, the slice provider summarizes the total number of new requests for each type of slice, checks the current amount of available physical resources, and gives the number of accesses for each type of slice. Then, if there are multiple upper-layer service providers applying for the same type of slice, it is necessary to further determine the allocation plan for the total access quota for this type of slice among these competitors.

[0063] The access control method for maximizing the long-term average benefits of network slice providers proposed by the present invention in the scenario of multiple network slice providers and multiple upper-layer service providers divides the access control process into time slots. The complete logical process is divided into two parts: Inter-slice Access Decision Based on Dominant Resource Efficiency (DREPSAC) and Intra-slice Access Quota Allocation Scheme Based on Valuation Weighted Proportional Fairness (VWPFA). The detailed process is as follows:

[0064] like Figure 1 As shown, the present invention provides a network slice access control method based on resource efficiency and auction, and its implementation method is as follows:

[0065] S1, initialization phase: count the total number of slice requests received in the current time slot, and initialize the temporary access decision vector between slices in the current time slot to 0 vector;

[0066] S2, inter-slice access decision stage: based on each slice, the cumulative access rate under the current cumulative request amount and cumulative access amount is calculated, and the order of trying to access different slices is determined according to the dominant resource efficiency of different slices under the current available resources, wherein the cumulative access amount includes the historical cumulative access amount and the current temporary access amount, and the implementation method is as follows:

[0067] S201. Calculate the cumulative access rate under the current cumulative request volume and cumulative access volume based on each slice;

[0068] S202, determine whether the cumulative access rate meets the priority requirements of different slices, if so, proceed to step S204, otherwise, proceed to step S203;

[0069] S203, select the slice types that do not meet the priority requirements, calculate their dominant resource efficiency under the currently available resources, and access the slice requests in the order of the dominant resource efficiency from high to low;

[0070] If the currently available resources allow access to a certain slice, and the total demand for such slices is greater than the current temporary access quantity, the temporary access quantity of the slice is increased by 1, and the process returns to step S201. If the currently available resources do not allow access to any slice, or the temporary access quantity of the slice is equal to the total demand, the temporary access decision is used as the final access decision for the time slot, the inter-slice access process ends, and the process goes to step S3.

[0071] S204. Calculate the dominant resource efficiency of all slices under the currently available resources, and access the slice requests in the order of the dominant resource efficiency from high to low, and set the feasible condition as follows: the total demand for this type of slice is greater than the current temporary access amount, and after the temporary access amount of this type of slice is increased by 1, its new cumulative access rate meets the priority condition;

[0072] If the currently available resources allow access to a certain slice, and there is a slice that meets the feasible condition, the temporary access amount of this type of slice is increased by 1, and the process returns to step S201; if there is no slice that meets the feasible condition, the temporary access decision is used as the final access decision for the time slot, the inter-slice access process is terminated, and the process goes to step S3;

[0073] In this embodiment, the network slice provider calculates the cumulative access rate under the current cumulative request volume and cumulative access volume for each slice it provides, where the cumulative access volume includes the historical cumulative access volume and the current temporary access volume. Check whether the cumulative access rate meets the priority requirements of different slices:

[0074] If the priority requirement is not met, select the slice types that violate the priority requirement and calculate their dominant resource efficiency under the currently available resources. Available resources refer to the total resource capacity minus the amount of resources occupied by slices that are still active, minus the amount of resources pre-allocated for temporary access decisions. Then try to access the slice request in order of dominant resource efficiency from high to low. If the available resources allow access to a slice, and the total demand for this type of slice is greater than the current temporary access volume, then add 1 to the temporary access volume of the slice and return to step S2; if no slice can be accessed, then the temporary access decision at this time is directly used as the final access decision for the time slot, ending the current inter-slice access process and entering step S3;

[0075] If the priority requirements are met, the dominant resource efficiency of all slices under the current available resources is calculated, and then the access slice request is attempted in the order of the dominant resource efficiency from high to low. If the available resources allow access to a certain slice, and the total demand for this type of slice is greater than the current temporary access volume, and after the temporary access volume of this type of slice increases by 1 unit, its new cumulative access rate still meets the priority conditions, then the temporary access volume of the slice is increased by 1, and the process returns to step S2; if there is no slice that meets these three conditions, the temporary access decision at this time is used as the final access decision for the time slot, the inter-slice access process ends, and the process goes to step S3.

[0076] S3, intra-slice access quota allocation phase: Based on the determined inter-slice access decision, the single item auction mechanism is used to allocate access quotas within the slice to complete access control of the network slice. The implementation method is as follows:

[0077] S301. According to the determined inter-slice access decision, determine whether there is intra-slice contention. If so, proceed to step S302; otherwise, proceed to step S303.

[0078] S302: Use the single item auction mechanism to redistribute the access quota of the competing slices to complete the control of network slice access. The implementation method is as follows:

[0079] S3021. Obtain slice valuation, and based on the slice valuation, remove upper-layer service providers whose access quota is less than the slice basic threshold, and calculate the increment generated by the following formula when one slice access quota is added to the remaining upper-layer service providers, and record the access quota a o,n The demand q of the upper service provider is gradually increased from 1 to o,n All incremental data generated when:

[0080]

[0081] Among them, a o,n represents the number of access quotas obtained by the upper layer service provider o from the network slice provider n, q o,n represents the slice access quota demand of the upper-layer service provider, o represents the upper-layer service provider, O s Indicates the set of upper-layer service providers supported by slice s, v o represents the true estimate of the slice by the upper layer service provider o, ε represents the positive bias, and log(·) represents the natural logarithmic function;

[0082] S3022. Arrange the incremental data in descending order using the following constraints, and select the sorted incremental data in sequence until the cumulative number of selected incremental data is equal to the total number of access quotas for the slice, thereby allocating access quotas within the slice and completing the control of network slice access:

[0083]

[0084]

[0085]

[0086]

[0087] Among them, o represents the upper layer service provider, O s represents the set of upper-layer service providers that provide services for slice s, a o,n represents the number of access quotas obtained by the upper layer service provider o from the network slice provider n, a n,s represents the total number of accesses to s-type slices from network slice provider n, represents a set of integers;

[0088] S303. Output the inter-slice access decision as the final access plan of the access process to complete the control of network slice access.

[0089] In this embodiment, after determining the inter-slice access decision, the present invention needs to further check whether multiple upper-layer service providers request the same slice at the same time, in other words, whether there is intra-slice competition. If there is no intra-slice competition, there is no need to further redistribute the inter-slice access decision, and it is directly output as the final access plan for the entire access process, ending the access control process of the current time slot; if there is intra-slice competition, it is still necessary to redistribute the access quota of the competing slices. To this end, the present invention designs a single item auction mechanism (VWPFA) that uses the slice basic threshold as the threshold and satisfies the dominant strategy incentive compatibility (DSIC attribute), which encourages competing upper-layer service providers to honestly report their valuations of the slices to the network slice provider, thereby allocating more slice access quotas to upper-layer service providers who need it more, so as to achieve a higher valuation weighted proportional fairness within the slice, while further improving the actual benefits of the network slice provider.

[0090] Among them, the VWPFA auction mechanism aims to optimize the fairness of the valuation weighted ratio between competing upper-layer service providers by allocating slice access quotas, that is, to find an intra-slice allocation plan. So that for other feasible allocation solutions {a o,n |o∈O s}, formula (1) is established:

[0091]

[0092] Among them, O s Indicates the set of upper-layer service providers that slice s can support. In the optimal allocation scheme, network slice provider n grants slice access quota to upper layer service provider o, v o represents the real valuation of the slice by the upper layer service provider, and ε represents a positive bias to prevent division by 0. It can be proved that the target allocation scheme that makes formula (1) valid is the optimal solution of formula (2).

[0093]

[0094]

[0095]

[0096]

[0097]

[0098] Formulas (3) to (6) are the constraints that need to be considered. Formula (3) indicates that the sum of the slice access quotas actually obtained by all competing upper-layer service providers cannot exceed the total access quota of this type of slice; Formula (4) indicates that the access quota obtained by the upper-layer service provider o must be non-negative; Formula (5) indicates that the slice quota that exceeds the demand of the upper-layer service provider o cannot be allocated to the upper-layer service provider o, q o,n That is, the number of slice requests initiated by the upper-layer service provider o; Formula (6) indicates that the number of slices obtained by each upper-layer service provider must be an integer.

[0099] In this embodiment, the slice provider asks the competing upper-layer service providers for their slice valuations. After obtaining the valuation information, the slice provider first eliminates the upper-layer service providers whose valuations are less than the slice basic threshold. For each remaining upper-layer service provider, the slice provider calculates the increment generated by formula (2) when adding a slice access quota to it, and records the increase in access quotas from 1 to the demand q of the upper-layer service provider. o,n All incremental data generated when

[0100] In this embodiment, the incremental data of all competing upper-layer service providers are sorted in descending order, and the sorted incremental data are selected in turn until the cumulative number of selected incremental data is equal to the total number of access quotas for the slice. The cumulative number of selected incremental data for each upper-layer service provider is the number of slice access quotas it finally obtains. Among them, the allocation scheme given above can maximize the value of formula (2) under constraints (3) to (6). In addition, under this allocation scheme, the allocation function of the upper-layer service provider remains monotonically non-decreasing. That is, when the valuations reported by other upper-layer service providers remain unchanged, a competitor increases the valuation information reported by himself, then the access quota allocated to him under the high valuation is not less than the access quota allocated under the low valuation. Among them, in order to prevent competing upper-layer service providers from defrauding more access quotas by inflating valuations, the network slice provider also needs to set a threshold for each access quota, and force each upper-layer service provider who hopes to obtain an access quota to honestly report its valuation through a reasonable threshold setting scheme. Myerson's lemma states that when the allocation function is monotonically non-decreasing, such a threshold setting scheme exists and is unique, that is, the threshold is set according to the valuation at the point where the allocation function occurs a step.

[0101] In this embodiment, in order to enable the competing upper layer service provider to report its own valuation of the slice, the slice provider needs to determine the threshold of each slice access quota allocated to the competing upper layer service provider per unit time:

[0102] A1. For example, after filtering out the incremental data of the upper-layer service provider o, the highest incremental data among the other incremental data that are not granted access quotas is taken as the unselected incremental data;

[0103] A2. If such unselected increments exist, the valuation of the upper-layer service provider o is gradually reduced until its final selected incremental data is reduced to the same value as the unselected incremental value just screened out. The valuation at this time is called the critical threshold, and then proceed to step A3; if there are no unselected increments, a slice basic threshold is set for all access quotas of the upper-layer service provider o whose values ​​have not yet been determined, and the entire access control process ends.

[0104] A3. Compare the critical threshold with the slice basic threshold specified by the slice provider. If the critical threshold is less than or equal to the slice basic threshold, the slice basic threshold is set for all access quotas whose values ​​have not yet been determined by the upper-layer service provider o, and the entire access control process ends; on the contrary, if the critical threshold is higher than the slice basic threshold, the upper-layer service provider o reports to the slice provider with a valuation slightly lower than the critical threshold, and he will lose an access quota obtained from the back to the front (that is, he will lose the last access quota first). Therefore, the present invention sets a critical threshold for the access quota, and then changes the valuation of the upper-layer service provider o to a value that is infinitely lower than this critical threshold, and recalculates its incremental data information under the new valuation, determines the unselected increment at this time, and returns to step A2.

[0105] In this embodiment, the present invention (DREPSAC+VWPFA) can help network slice providers seek higher long-term average benefits in an environment with multiple network slice providers and multiple upper-layer service providers. Figure 2 The performance of the present invention and three existing methods in terms of long-term average basic benefits is given. It can be seen that DREPSAC is significantly higher than PAGE and MQSAC, and slightly higher than the DSARA method which requires training cost. Figure 3 The performance of the present invention and three existing methods in terms of long-term average real benefits is given. It can be seen that the VWPFA mechanism further improves the actual benefits of DREPSAC through auctions between slices.

[0106] In this embodiment, the present invention effectively maintains the priority differences between slices. Figure 4 The inter-slice fairness index of each method is given. If the index is not zero, it means that the cumulative access rate of each slice is strictly monotonically non-decreasing in the order of slice priority, that is, it meets the requirements of the present invention for the inter-slice priority condition. It can be seen that other existing methods cannot achieve this.

[0107] In this embodiment, the present invention can ensure that the competing upper-layer service providers honestly report their valuations of the slices to the slice provider, thereby better improving the fairness of the allocation of access quotas within the slice. The present invention designs a single-item auction mechanism that takes the slice base threshold as the reserved value and satisfies the dominant strategy incentive compatibility (DSIC attribute), which encourages the competing upper-layer service providers to honestly report their valuations of the slices to the network slice provider, thereby allocating more slice access quotas to the upper-layer service providers who need it more, so as to achieve a higher valuation weighted proportional fairness within the slice. Figure 5 The valuation weighted proportional fairness index of the VWPFA auction mechanism and the scheme (OP) that allocates based on the proportion of request quantity is given. It can be seen that when making access decisions between fixed slices, the VWPFA mechanism has better intra-slice fairness.

[0108] Example 2

[0109] like Figure 6 As shown, the present invention provides a network slice access control system based on resource efficiency and auction, including:

[0110] The first processing module is used to count the total number of slice requests of various types received in the current time slot, and initialize the temporary access decision vector between slices in the current time slot to a 0 vector;

[0111] A second processing module is used to calculate, based on each slice, a cumulative access rate under a current cumulative request amount and a cumulative access amount, and determine an order of trying to access different slices according to dominant resource efficiencies of different slices under currently available resources, wherein the cumulative access amount includes a historical cumulative access amount and a current temporary access amount;

[0112] The third processing module is used to determine whether to redistribute the access quota of the competing slices based on the determined inter-slice access decision by using the single-item auction mechanism to complete the control of network slice access.

[0113] like Figure 6 The network slice access control system based on resource efficiency and auction provided by the illustrated embodiment can execute the technical solution shown in the network slice access control method of the above method embodiment. Its implementation principle and beneficial effects are similar and will not be repeated here.

[0114] In this embodiment, the present application can divide the functional units according to the network slice access control method. For example, each function can be divided into each functional unit, or two or more functions can be integrated into one processing unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of a software functional unit. It should be noted that the division of units in the present invention is schematic and is only a logical division. There may be other division methods in actual implementation.

[0115] In this embodiment, in order to realize the principles and beneficial effects of the network slice access control method, the network slice access control system includes hardware structures and / or software modules corresponding to the execution of each function. Those skilled in the art should easily realize that, in combination with the schematic units and algorithm steps described in the embodiments disclosed in the present invention, the present invention can be implemented in the form of hardware and / or a combination of hardware and computer software. Whether a function is executed in a hardware or computer software driven manner depends on the specific application and design constraints of the technical solution. Different methods can be used for each specific application to implement the described function, but such implementation should not be considered to exceed the scope of this application.

[0116] In this embodiment, the present invention intends to design an access control method for maximizing long-term average benefits for slice providers in the scenario of "multiple slice providers and multiple upper-layer service providers". The method quickly determines the access decision between slices based on the dominant resource efficiency of the slices, maintains the priority difference between slices; and then designs an auction mechanism to allocate access quotas within the slices, maximizing the fairness of the valuation weighted ratio within the slices, thereby maintaining a stable group of slice tenants for the slice provider and improving its long-term average benefits.

Claims

1. A network slice access control method based on resource efficiency and auction, characterized in that: The following steps are involved: S1, initialization phase: count the total number of slice requests received in the current time slot, and initialize the temporary access decision vector between slices in the current time slot to 0 vector; S2, inter-slice access decision stage: based on each slice, the cumulative access rate under the current cumulative request amount and cumulative access amount is calculated, and the order of trying to access different slices is determined according to the dominant resource efficiency of different slices under the current available resources, wherein the cumulative access amount includes the historical cumulative access amount and the current temporary access amount, which is specifically: S201. Calculate the cumulative access rate under the current cumulative request volume and cumulative access volume based on each slice; S202, determine whether the cumulative access rate meets the priority requirements of different slices, if so, proceed to step S204, otherwise, proceed to step S203; S203, select the slice types that do not meet the priority requirements, calculate their dominant resource efficiency under the currently available resources, and access the slice requests in the order of the dominant resource efficiency from high to low; If the currently available resources allow access to a certain slice, and the total demand for such slices is greater than the current temporary access quantity, the temporary access quantity of the slice is increased by 1, and the process returns to step S201. If the currently available resources do not allow access to any slice, or the temporary access quantity of the slice is equal to the total demand, the temporary access decision is used as the final access decision for the time slot, the inter-slice access process ends, and the process goes to step S3. S204. Calculate the dominant resource efficiency of all slices under the currently available resources, and access the slice requests in the order of the dominant resource efficiency from high to low, and set the feasible condition as follows: the total demand for this type of slice is greater than the current temporary access amount, and after the temporary access amount of this type of slice is increased by 1, its new cumulative access rate meets the priority condition; If the currently available resources allow access to a certain slice, and there is a slice that meets the feasible condition, the temporary access amount of this type of slice is increased by 1, and the process returns to step S201; if there is no slice that meets the feasible condition, the temporary access decision is used as the final access decision for the time slot, the inter-slice access process is terminated, and the process goes to step S3; S3, intra-slice access quota allocation stage: Based on the determined inter-slice access decision, the single item auction mechanism is used to allocate access quotas within the slice to complete access control of the network slice.

2. The network slice access control method based on resource efficiency and auction according to claim 1, characterized in that: The step S3 comprises the following steps: S301. According to the determined inter-slice access decision, determine whether there is intra-slice contention. If so, proceed to step S302; otherwise, proceed to step S303. S302. Use a single item auction mechanism to redistribute the access quotas of the competing slices to complete the control of network slice access. S303. Output the inter-slice access decision as the final access plan of the access process to complete the control of network slice access.

3. The network slice access control method based on resource efficiency and auction according to claim 2, characterized in that: The step S302 includes the following steps: S3021. Obtain slice valuation, and based on the slice valuation, remove upper-layer service providers whose access quota is less than the slice basic threshold, and calculate the increment generated by the following formula when one slice access quota is added to the remaining upper-layer service providers, and record the access quota a o,n The demand q of the upper service provider is gradually increased from 1 to o,n All incremental data generated when: Among them, a o,n represents the number of access quotas obtained by the upper layer service provider o from the network slice provider n, q o,n represents the slice access quota demand of the upper-layer service provider, o represents the upper-layer service provider, O s Indicates the set of upper-layer service providers supported by slice s, v o represents the true estimate of the slice by the upper layer service provider o, ε represents the positive bias, and log(·) represents the natural logarithmic function; S3022. Arrange the incremental data in descending order using the following constraints, and select the sorted incremental data in sequence until the cumulative number of selected incremental data is equal to the total number of access quotas for the slice, thereby allocating access quotas within the slice and completing the control of network slice access: Among them, a n,s represents the total number of accesses to s-type slices from network slice provider n, Represents a set of integers.

4. A network slice access control system according to any one of claims 1 to 3, wherein: include: The first processing module is used to count the total number of slice requests of various types received in the current time slot, and initialize the temporary access decision vector between slices in the current time slot to a 0 vector; A second processing module is used to calculate, based on each slice, a cumulative access rate under a current cumulative request amount and a cumulative access amount, and determine an order of trying to access different slices according to dominant resource efficiencies of different slices under currently available resources, wherein the cumulative access amount includes a historical cumulative access amount and a current temporary access amount; The third processing module is used to determine whether to redistribute the access quota of the competing slices based on the determined inter-slice access decision by using the single-item auction mechanism to complete the control of network slice access.

Images