CN112737823A - Resource slice allocation method and device and computer equipment - Google Patents
Resource slice allocation method and device and computer equipment Download PDFInfo
- Publication number
- CN112737823A CN112737823A CN202011535215.5A CN202011535215A CN112737823A CN 112737823 A CN112737823 A CN 112737823A CN 202011535215 A CN202011535215 A CN 202011535215A CN 112737823 A CN112737823 A CN 112737823A
- Authority
- CN
- China
- Prior art keywords
- resource allocation
- resource
- target
- allocation scheme
- utility function
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 103
- 238000013468 resource allocation Methods 0.000 claims abstract description 249
- 230000006870 function Effects 0.000 claims abstract description 134
- 239000011159 matrix material Substances 0.000 claims abstract description 50
- 238000005192 partition Methods 0.000 claims abstract description 46
- 238000001228 spectrum Methods 0.000 claims description 22
- 230000015654 memory Effects 0.000 claims description 19
- 230000007704 transition Effects 0.000 claims description 6
- 230000001186 cumulative effect Effects 0.000 claims description 5
- 238000004364 calculation method Methods 0.000 claims description 3
- 238000009826 distribution Methods 0.000 claims description 3
- 238000004590 computer program Methods 0.000 claims description 2
- 230000008859 change Effects 0.000 abstract description 19
- 230000008569 process Effects 0.000 abstract description 10
- 230000004044 response Effects 0.000 abstract description 9
- 238000004422 calculation algorithm Methods 0.000 description 23
- 230000009471 action Effects 0.000 description 19
- 230000002787 reinforcement Effects 0.000 description 13
- 238000010586 diagram Methods 0.000 description 11
- 238000005516 engineering process Methods 0.000 description 7
- 238000005457 optimization Methods 0.000 description 6
- 238000000638 solvent extraction Methods 0.000 description 6
- 238000012545 processing Methods 0.000 description 5
- 239000003795 chemical substances by application Substances 0.000 description 4
- 230000005540 biological transmission Effects 0.000 description 3
- 230000003595 spectral effect Effects 0.000 description 3
- 230000003068 static effect Effects 0.000 description 3
- 238000004891 communication Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000004088 simulation Methods 0.000 description 2
- 238000009827 uniform distribution Methods 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 239000002360 explosive Substances 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/08—Configuration management of networks or network elements
- H04L41/0893—Assignment of logical groups to network elements
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/14—Network analysis or design
- H04L41/147—Network analysis or design for predicting network behaviour
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W16/00—Network planning, e.g. coverage or traffic planning tools; Network deployment, e.g. resource partitioning or cells structures
- H04W16/02—Resource partitioning among network components, e.g. reuse partitioning
- H04W16/10—Dynamic resource partitioning
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W28/00—Network traffic management; Network resource management
- H04W28/02—Traffic management, e.g. flow control or congestion control
- H04W28/10—Flow control between communication endpoints
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W28/00—Network traffic management; Network resource management
- H04W28/16—Central resource management; Negotiation of resources or communication parameters, e.g. negotiating bandwidth or QoS [Quality of Service]
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Quality & Reliability (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention discloses a resource slice allocation method, a device and computer equipment, wherein the method comprises the following steps: acquiring historical data of a target network in a preset time period, and predicting flow values of each link in the future preset time period; determining a link resource partition matrix and a node resource partition matrix according to a preset initial link allocation state and a preset initial node allocation state; then determining a plurality of resource allocation schemes at the current moment; respectively calculating target utility functions of all resource allocation schemes meeting the reliability conditions at the current moment; and taking the resource allocation scheme corresponding to the maximum target utility function as the target resource allocation scheme at the current moment. By implementing the method, historical data and a future utility function are combined, so that the network resource division state at the future moment affects the current division strategy, the current optimal strategy is obtained, and the rapid response to the future network demand change in the resource allocation process is ensured.
Description
Technical Field
The invention relates to the field of communication, in particular to a resource slice allocation method, a resource slice allocation device and computer equipment.
Background
With the rapid development of the 5G network, better bandwidth experience is brought to users, and meanwhile more diversified requirements are brought to the mobile network. Network slicing has received increasing attention as one of the key technologies of 5G. Based on a virtualization technology, a 5G physical network is logically cut into a plurality of virtual end-to-end networks, network slices are isolated from one another, and congestion, overload and configuration adjustment of any one network slice cannot affect other network slices; the 5G network slice enables operators to construct flexible networks to meet various use conditions in different industry vertical fields.
When the 5G network is applied to a power grid scenario, power services can be divided into three categories: mobile application services, control services and information collection services. Each kind of power service has different service quality requirements, and in a network architecture based on slices, the quality of the slices directly affects the performance of the network, so that the dynamic optimization of slice resources is very important.
In the related art, a resource allocation strategy is mainly determined by optimizing according to the current flow, and the flow demand change of a future network cannot be responded, so that the user experience is influenced.
Disclosure of Invention
In view of this, embodiments of the present invention provide a resource slice allocation method, an apparatus, and a computer device, so as to solve the problem that a resource allocation policy is determined only by performing optimization according to a current flow, and a future change in a network flow demand cannot be responded, which affects user experience.
According to a first aspect, an embodiment of the present invention provides a resource slice allocation method, including: acquiring historical data of a target network in a preset time period, wherein the target network comprises a plurality of links and a plurality of nodes, and the links and the nodes comprise a plurality of resource slices; respectively predicting the flow value of each link in a future preset time period according to the historical data; determining a link resource partition matrix and a node resource partition matrix according to a preset initial link allocation state and a preset initial node allocation state; determining a plurality of resource allocation schemes at the current moment according to a topological structure of a target network, a flow value of each link in a future preset time period, a link resource partition matrix and a node resource partition matrix, wherein the resource allocation schemes comprise a link resource allocation scheme and a node allocation scheme; respectively calculating target utility functions of all resource allocation schemes meeting the reliability conditions at the current moment; and taking the resource allocation scheme corresponding to the maximum target utility function as the target resource allocation scheme at the current moment.
With reference to the first aspect, in a first implementation manner of the first aspect, the determining whether each resource allocation scheme satisfies a reliability condition at a current time includes: respectively acquiring a time delay value and a frequency spectrum utilization value of each resource allocation scheme; and judging whether the time delay value of each resource allocation scheme is smaller than the target time delay value or not, and judging whether the frequency spectrum utilization rate of each resource allocation scheme is larger than the target frequency spectrum utilization rate or not.
With reference to the first aspect, in a second embodiment of the first aspect, the method further includes: and distributing a plurality of resource slices on the target network to each network slice at the current moment according to the target resource distribution scheme at the current moment.
With reference to the first aspect, in a third implementation manner of the first aspect, the process of calculating the target utility function of each resource allocation scheme includes: calculating a link utility function and a node utility function of each resource allocation scheme; and determining the target utility function according to the link utility function and the node utility function.
With reference to the third implementation manner of the first aspect, in a fourth implementation manner of the first aspect, the target utility function is determined by the following formula:
Vt＝Vm+Vn,
wherein, VtRepresenting the total utility function, gamma a predetermined state transition factor, VmRepresenting the link utility function, VnRepresenting the node utility function;
calculating the link utility function and the node utility function by the following formulas:
Vπ(s)＝Eπ{Gt|St＝s},
R＝α*SE+β*QoE,
wherein QoE represents a user service experience value, SE represents spectral efficiency, and α and β represent different weightsWeight factor, R represents the reward function of the resource allocation scheme at the current time, GtRepresenting a cumulative reward function.
According to a second aspect, an embodiment of the present invention provides a resource slice allocation method, including: acquiring a resource allocation scheme of a target network at the last moment; according to the first aspect or the resource slice allocation method according to any embodiment of the first aspect, a target resource allocation scheme of a target network at the current moment is determined; calculating a difference value between the target utility function corresponding to the resource allocation scheme at the previous moment and the target utility function corresponding to the target resource allocation scheme at the current moment; judging whether the difference value is smaller than or equal to a preset threshold value; and if the difference is smaller than or equal to a preset threshold value, determining that the target resource allocation scheme is converged, and taking the target resource allocation scheme as an optimal resource allocation scheme.
With reference to the second aspect, in a first embodiment of the second aspect, the method further comprises: when the difference is greater than the preset threshold, taking the target resource allocation scheme at the current moment as the resource allocation scheme at the current moment, returning to execute the resource slice allocation method according to the first aspect or any embodiment of the first aspect, determining the target resource allocation scheme at the current moment of the target network, to judge whether the difference is less than or equal to the preset threshold, until the difference is less than or equal to the preset threshold, determining that the target resource allocation scheme is converged, and taking the target resource allocation scheme as the optimal resource allocation scheme.
According to a third aspect, an embodiment of the present invention provides a resource slice allocation apparatus, including: the historical data acquisition module is used for acquiring historical data of a target network in a preset time period, the target network comprises a plurality of links and a plurality of nodes, and the links and the nodes comprise a plurality of resource slices; the prediction module is used for respectively predicting the flow value of each link in the future preset time period according to the historical data; the matrix determination module is used for determining a link resource partition matrix and a node resource partition matrix according to a preset initial link allocation state and a preset initial node allocation state; the system comprises a plurality of resource allocation scheme determination modules, a resource allocation module and a resource allocation module, wherein the resource allocation scheme determination modules are used for determining a plurality of resource allocation schemes at the current moment according to a topological structure of a target network, a flow value of each link in a future preset time period, a link resource partition matrix and a node resource partition matrix, and the resource allocation schemes comprise a link resource allocation scheme and a node allocation scheme; the calculation module is used for respectively calculating a target utility function of each resource allocation scheme meeting the reliability condition at the current moment; and the target resource allocation scheme determining module is used for taking the resource allocation scheme corresponding to the maximum target utility function as the target resource allocation scheme at the current moment.
According to a fourth aspect, an embodiment of the present invention provides a resource slice allocation apparatus, including: the resource allocation scheme acquisition module is used for acquiring a resource allocation scheme of a target network at the last moment; a determining module, configured to determine a target resource allocation scheme of a target network at a current time according to the resource slice allocation method according to the first aspect or any embodiment of the first aspect; a difference value calculating module, configured to calculate a difference value between the target utility function corresponding to the resource allocation scheme at the previous time and the target utility function corresponding to the target resource allocation scheme at the current time; the judging module is used for judging whether the difference value is smaller than or equal to a preset threshold value or not; and the optimal resource allocation scheme determining module is used for determining the convergence of the target resource allocation scheme if the difference is smaller than or equal to a preset threshold value, and taking the target resource allocation scheme as the optimal resource allocation scheme.
According to a fifth aspect, an embodiment of the present invention provides a computer device, including: at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores instructions executable by the one processor to cause the at least one processor to perform the steps of the resource slice allocation method according to the first aspect or any one of the embodiments of the first aspect and the steps of the resource slice allocation method according to any one of the embodiments of the second aspect or the second aspect.
According to a sixth aspect, an embodiment of the present invention provides a computer-readable storage medium, on which a computer program is stored, which, when executed by a processor, implements the steps of the resource slice allocation method described in the first aspect or any one of the embodiments of the first aspect, and the steps of the resource slice allocation method described in the second aspect or any one of the embodiments of the second aspect.
The technical scheme of the invention has the following advantages:
1. the invention provides a resource slice allocation method, a device and computer equipment, wherein the resource slice allocation method comprises the following steps: acquiring historical data of a target network within a preset time period; respectively predicting the flow value of each link in a future preset time period according to historical data; determining a link resource partition matrix and a node resource partition matrix according to a preset initial link allocation state and a preset initial node allocation state; determining a plurality of resource allocation schemes according to the topological structure of the target network, the flow value of each link in a future preset time period, a link resource partition matrix and a node resource partition matrix, wherein the resource allocation schemes comprise a link resource allocation scheme and a node allocation scheme; respectively calculating target utility functions of all resource allocation schemes meeting the reliability conditions at the current moment; and taking the resource allocation scheme corresponding to the maximum target utility function as the target resource allocation scheme at the current moment. By implementing the method and the device, the resource allocation method at the current moment is determined by combining the flow value in the future time period and the future utility function, and the problems that the resource allocation strategy is optimized only according to the current flow, future changes cannot be responded, and user experience is influenced in the related technology are solved. By combining historical data, the service flow can be predicted according to the service flow change condition in the future network slice, so that the division condition of the future network resources is deduced; and then, by calculating a future utility function, the network resource division state at a future moment affects the current division strategy, so that the current optimal strategy is obtained, and the rapid response to the future network demand change in the resource allocation process is ensured.
2. The invention provides a resource slice allocation method, a device and computer equipment, wherein the resource slice allocation method comprises the following steps: acquiring a resource allocation scheme of a target network at the last moment; according to the resource slice allocation method in the embodiment, a target resource allocation scheme of a target network at the current moment is determined; calculating a difference value between the target utility function corresponding to the resource allocation scheme at the previous moment and the target utility function corresponding to the target resource allocation scheme at the current moment; judging whether the difference value is smaller than or equal to a preset threshold value; and if the difference is less than or equal to the preset threshold, determining that the target resource allocation scheme is converged, and taking the target resource allocation scheme as the optimal resource allocation scheme. By implementing the method and the device, the resource allocation method at the current moment is determined by combining the flow value in the future time period and the future utility function, and the problems that the resource allocation strategy is optimized only according to the current flow, future changes cannot be responded, and user experience is influenced in the related technology are solved. By combining historical data, the service flow can be predicted according to the service flow change condition in the future network slice, so that the division condition of the future network resources is deduced; and then, by calculating a future utility function, the network resource division state at a future moment affects the current division strategy, so that the current optimal strategy is obtained, and the rapid response to the future network demand change in the resource allocation process is ensured.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and other drawings can be obtained by those skilled in the art without creative efforts.
Fig. 1 is a flowchart of a specific example of a resource slice allocation method in an embodiment of the present invention;
fig. 2 is a flowchart of another specific example of a resource slice allocation method in an embodiment of the present invention;
FIG. 3 is a diagram illustrating a state transition in a resource slice allocation method according to an embodiment of the present invention;
FIG. 4 is a schematic diagram illustrating an average node resource utilization rate of each algorithm in the resource slice allocation method according to the embodiment of the present invention;
fig. 5 is a schematic diagram of average link resource utilization of each algorithm in the resource slice allocation method in the embodiment of the present invention;
fig. 6 is a schematic diagram of an average resource utilization rate of nodes of each algorithm in the resource slice allocation method in the embodiment of the present invention;
fig. 7 is a schematic diagram of link average resource utilization of each algorithm in the resource slice allocation method in the embodiment of the present invention;
fig. 8 is a schematic block diagram of a specific example of a resource slice allocation apparatus in the embodiment of the present invention;
fig. 9 is a schematic block diagram of a specific example of a resource slice allocation apparatus in the embodiment of the present invention;
FIG. 10 is a diagram showing a specific example of a computer device according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
With the rapid development of the power service industry, the explosive growth of various power grid equipment, power terminals and power consumer communication demands forces the world power grid to transform from the traditional power grid to the smart power grid. Due to different requirements of grid use cases, an ultra-reliable, low-delay, flexible and low-cost network is needed, and a 5G network chip has corresponding matching capability, so that perfect combination of 'bits + watts' can be realized. Based on the above background, in order to solve the problem that the change of the traffic demand of the future network cannot be responded to and the user experience is affected because the resource allocation policy is determined only by optimizing according to the current traffic in the related art, embodiments of the present invention provide a resource slice allocation method, an apparatus, and a computer device, which are capable of quickly responding to the change of the network demand in the resource allocation process along with the current demand and the future demand of a service.
An embodiment of the present invention provides a resource slice allocation method, as shown in fig. 1, including:
step S11: acquiring historical data of a target network in a preset time period, wherein the target network comprises a plurality of links and a plurality of nodes, and the links and the nodes comprise a plurality of resource slices; in this embodiment, the preset time period may be a historical time period determined according to a user requirement, or the like; the target network may be a core network, the core network includes a plurality of links and a plurality of nodes, the network resources of the core network are discretized, that is, the network resources on the links are divided into a plurality of resource slices, and the network resources on each node are divided into a plurality of resource slices, the resource slices can independently provide the network resources for different 5G network slices, that is, each resource slice can be used as a minimum allocation unit for resource allocation; the historical data of the target network may include historical data packets of each link and each node in the previous period, and the historical data packets include: the service type of the data and its corresponding network slice number, packet length, timestamp, etc. Specifically, historical data packets of each link and each node on the core network within a period of time are obtained.
Step S12: respectively predicting the flow value of each link in a future preset time period according to historical data; in this embodiment, the future preset time period may be corresponding to the preset time period described in the above embodiment, for example, the historical data packet in the first twelve hours of the current time is obtained, the flow value in the future twelve hours of the current time may be correspondingly predicted, and the preset time period may be specifically determined according to an actual application scenario, which is not limited in this embodiment of the present invention. Specifically, according to the data demand of the core network in the historical time period, the traffic demand of various service packets in each link in the future discrete time can be predicted.
Step S13: determining a link resource partition matrix and a node resource partition matrix according to a preset initial link allocation state and a preset initial node allocation state; in this embodiment, the preset initial link allocation state may be a preset allocation condition of resource slices on each link in the core network, and similarly, the preset initial node allocation state may be a preset allocation condition of multiple resource slices on each node in the core network; the link resource partition matrix may be a matrix describing an allocation state of each link on the core network, and similarly, the node resource partition matrix may be a matrix describing an allocation state of each node on the core network.
In particular, there may be k network slices in the core network, the initial allocation state S of link iaMay be ai＝(b1，b2，...，bn) Wherein b iskIs the number of resource slices to which the kth network slice is allocated,
is the total number of resource slices on the link i, and then the link resource partition matrix B of the whole core network can be obtainedm×nWherein each row vector describes the resource allocation of each link, i.e. each row vector describes the resource slice partition of each link, defining Sa＝Bm×nWherein m links exist in the core network. Initial assignment State S for node jnMay be nj＝(c1，c2，...，cn) Wherein c iskIs the number of resource slices to which the kth network slice is allocated,
is the total number of resource slices on the node j, and then the node resource partition matrix C of the whole core network can be obtainedk×nWherein each row vector describes the resource allocation of each node, that is to sayEach row vector describes a resource slice partition of each node, defining Sn＝Cp×nWherein, there are p nodes in the core network.
Step S14: determining a plurality of resource allocation schemes at the current moment according to the topological structure of the target network, the flow value of each link in a future preset time period, a link resource partition matrix and a node resource partition matrix, wherein the resource allocation schemes comprise a link resource allocation scheme and a node allocation scheme; in this embodiment, the topology of the target network may be a network topology of a core network, and is used to represent connection relationships between links and nodes in the core network; the flow value of each link in the future preset time period may be a future flow value of each link in the core network predicted and determined by the method described in the above embodiment; specifically, a plurality of different resource allocation schemes, including a resource slice allocation scheme on each link and an allocation scheme of a plurality of resource slices on each node, may be determined through the topology of the target network, the traffic prediction value of each service packet on each link, the link resource partition matrix, and the node resource partition matrix.
Step S15: respectively calculating target utility functions of all resource allocation schemes meeting the reliability conditions at the current moment; in this embodiment, each resource allocation scheme that satisfies the reliability condition at the current time may be that, when the resource allocation scheme is implemented, the corresponding spectrum utilization index and the corresponding time delay index satisfy a preset threshold; the target utility function for each resource allocation scheme may be the corresponding quality of service and frequency efficiency when implementing the resource allocation scheme. Specifically, the target utility function values of the resource allocation schemes meeting the reliability requirements of the nodes are calculated respectively.
Step S16: and taking the resource allocation scheme corresponding to the maximum target utility function as the target resource allocation scheme at the current moment. In this embodiment, the target utility function values of the resource allocation schemes that meet the reliability requirement at the current time are compared, and the resource allocation scheme corresponding to the largest target utility function value is determined as the target resource allocation scheme at the current time, for example, the network state at the current time may be S00The resource allocation schemes that satisfy the reliability requirement at the current time determined by the method described in the above embodiment are a00、a01、a02That is, the actions that can be executed at the current time are a00、a01、a02(ii) a Resource allocation scheme a00Corresponding to network state S10Resource allocation scheme a01Corresponding to network state S11Resource allocation scheme a02Corresponding to network state S12State S10The corresponding value of the target utility function is r10State S11The corresponding value of the target utility function is r11State S12The corresponding value of the target utility function is r12。
When r is10＞r11And r is10＞r12When it is, r is10Corresponding resource allocation scheme a10As a target resource allocation scheme for the current time.
The invention provides a resource slice allocation method, which comprises the following steps: acquiring historical data of a target network within a preset time period; respectively predicting the flow value of each link in a future preset time period according to historical data; determining a link resource partition matrix and a node resource partition matrix according to a preset initial link allocation state and a preset initial node allocation state; determining a plurality of resource allocation schemes at the current moment according to the topological structure of the target network, the flow value of each link in a future preset time period, a link resource partition matrix and a node resource partition matrix, wherein the resource allocation schemes comprise a link resource allocation scheme and a node allocation scheme; respectively calculating target utility functions of all resource allocation schemes meeting the reliability conditions at the current moment; and taking the resource allocation scheme corresponding to the maximum target utility function as the target resource allocation scheme at the current moment.
By implementing the method and the device, the resource allocation method at the current moment is determined by combining the flow value in the future time period and the target utility function, and the problems that the resource allocation strategy is optimized only according to the current flow, the resource allocation strategy cannot be adaptively adjusted in response to future changes, and the user experience is influenced in the related art are solved. By combining historical data, the service flow can be predicted according to the service flow change condition in the future network slice, so that the division condition of the future network resources is deduced; and then, by calculating a future utility function, the network resource division state at a future moment affects the current division strategy, so that the current optimal strategy is obtained, and the rapid response to the future network demand change in the resource allocation process is ensured.
As an optional embodiment of the present invention, the step S15 of determining whether each resource allocation plan satisfies the reliability condition at the current time includes:
firstly, respectively obtaining a time delay value and a frequency spectrum utilization value of each resource allocation scheme; in this embodiment, the delay value corresponding to each resource allocation scheme may be a sum of a transmission delay, a propagation delay, a processing delay, and a queuing delay, where the transmission delay is time required for transmitting data, the transmission delay is time required for transmitting data in a link or a network, the processing delay may be time for receiving data by a target node, and the queuing delay may be time for waiting for processing data; the spectrum utilization may be a utilization degree of a plurality of resource slices on each node and each link in the current core network. Specifically, a time delay value and a frequency spectrum utilization rate corresponding to each resource allocation scheme are respectively obtained.
Secondly, judging whether the time delay value of each resource allocation scheme is smaller than the target time delay value, and judging whether the frequency spectrum utilization rate of each resource allocation scheme is larger than the target frequency spectrum utilization rate. In this embodiment, the target delay value may be an index value determined according to the specific service and the actual application scenario, and the target spectrum utilization rate may be an index value determined according to the specific service and the actual application scenario. Specifically, whether the time delay value and the frequency spectrum utilization rate corresponding to each resource allocation scheme meet respective index values is judged, and when the time delay value is smaller than a target time delay value, the resource allocation scheme is determined to meet the time delay requirement; when the frequency spectrum utilization rate is greater than the target frequency spectrum utilization rate, determining that the resource allocation scheme meets the requirement of the frequency spectrum utilization rate; when the resource allocation scheme meets the time delay requirement and the frequency spectrum utilization requirement, the resource allocation scheme is determined to meet the reliability requirement of the current node at the current moment.
Specifically, the current link state may be SaThe current node state may be SnThe target utility function at the current time, i.e., the corresponding link utility function and node utility function, can be calculated. When transferred to the adjacent state S within time t1a' and SnFirstly, checking whether the current resource allocation scheme meets the reliability requirement, and when the current reliability requirement is met, obtaining a corresponding target utility function in a new state, and then performing next state conversion, namely determining the resource allocation scheme at the current moment. In particular, when adjacent state Sa' and Sn' when the current reliability requirement is not satisfied, the adjacent state Sa' and Sn' Mark as invalid State, at which point it can continue to fall back to original State Sa、SnThen proceed to the next time, for Sa"and Sn"is verified.
As an optional embodiment of the present invention, the resource slice allocation method further includes:
and distributing a plurality of resource slices on the target network to each network slice at the current moment according to the target resource distribution scheme at the current moment.
As an optional embodiment of the present invention, in the step S15, the process of calculating the target utility function of each resource allocation plan includes:
firstly, calculating a link utility function and a node utility function of each resource allocation scheme; in this embodiment, the utility function V of the linkmMay be the utility function V of each network slice on the linkmjSuperposition of (2); similarly, the node utility function VnMay be the utility function V of each network slice on the nodenjAnd (3) superposition.
Calculating a link utility function and a node utility function by the following formulas:
Vπ(s)＝Eπ{Gt|St＝s}，
Vm，j，π(s)＝Em，jπ{Gt|St＝s}，
Vn，j，π(s)＝En，jπ{Gt|St＝s}，
R＝α*SE+β*QoE，
wherein QoE represents a user service experience value, SE represents spectrum efficiency, alpha and beta represent different weight factors, R represents a reward function of a resource allocation scheme at the current moment, and G represents a reward function of a resource allocation scheme at the current momenttRepresenting a cumulative reward function.
In particular, the reward function may be a weighted sum of the normalized spectral efficiency and the user service experience value.
And secondly, determining a target utility function according to the link utility function and the node utility function. In this embodiment, the target utility function is determined by the following formula:
Vt＝Vm+Vn，
wherein the content of the first and second substances,
denotes, gamma denotes a preset state transition factor, VmRepresenting the link utility function, VnRepresenting a node utility function.
Specifically, the method according to the embodiment of the present invention is a value-based reinforcement learning algorithm, thereby allocating network resources in a core network. Q (s, a) in reinforcement learning represents the expectation that the agent can obtain a reward by taking an action a in the state s. The reinforcement learning algorithm has several key factors such as states, actions, rewards, and cost functions, and when the method described in this embodiment is applied to a specific application scenario in which resource slices are allocated, the mapping relationship between each key factor in reinforcement learning and each index in an actual application scenario is as follows:
the state in reinforcement learning refers to the current situation returned by the environment, and the current network situation can be taken as the state of reinforcement learning for the problem of network slice resource allocation. Assuming that the total number of users in the network is not changed, and the service type of each user changes with time, the present invention selects the number of downlink data packets of each service type as the state in the reinforcement learning.
The actions in reinforcement learning represent all possible actions that an agent may take, and for network slice resource allocation problems, a resource allocation scheme may be taken as an action. In the method according to the embodiment of the present invention, all possible resource slice allocation schemes in the next state are all actions that the agent can take in the next state.
Rewards in reinforcement learning represent immediate return values of the environment that can be used to assess the quality of certain actions of the agent. In this embodiment, a reward function of the weighted sum of the spectrum efficiency SE and the quality of service QoE may be used as a reward for reinforcement learning. The spectral efficiency is defined as the total data volume transmitted by the system in unit time divided by the total bandwidth of the system, and is normalized for problem analysis. The QoE considering delay is defined as a ratio of a downlink data amount satisfying a service level agreement delay to a total downlink data amount, and is a dimensionless number less than 1. The reward function calculates the reward which can be generated by a specific resource allocation method at a certain moment, and the size of the reward can reflect the influence of executing the resource allocation to a certain extent.
In the reinforcement learning algorithm, the cost function represents the long-term return. The size of the reward can reflect the quality of the effect of executing the resource allocation to a certain extent, however, the next state to be transferred affects the subsequent state, so that an action affects the subsequent state in addition to the current state, and the action with the maximum instant reward is not necessarily large, but the method of the embodiment aims to maximize the accumulated reward in a period of time instead of the instant reward. The expectation of accumulated earnings for a period of time may be represented by a cost function, and the optimal resource slice allocation scheme may be determined by determining a maximized cost function in response to the policy.
The cumulative reward may be calculated by the following equation:
wherein gamma is a discount factor, gamma is more than or equal to 0 and less than or equal to 1, and the reward obtained by adopting a certain action at the moment t is RtThen the subsequent reward sequence is Rt+1，Rt+2，Rt+3，...。
The cost function is defined as the cumulative reward of expected values at state s when policy π is employed. The cost function V can be calculated by the following formulaπ(s)：
An embodiment of the present invention provides a resource slice allocation method, as shown in fig. 2, including:
step S21: acquiring a resource allocation scheme of a target network at the last moment; in this embodiment, the resource allocation scheme at the previous time may be a preset initial resource allocation scheme, or may be a resource allocation scheme at any time.
Step S22: according to the resource slice allocation method described in the above embodiment, a target resource allocation scheme of a target network at the current time is determined; in this embodiment, according to the resource slice allocation method described in the foregoing embodiment, the allocation schemes of multiple resource slices on each link and the allocation schemes of multiple resource slices on each node in the current network state of the core network can be determined, that is, the actions that can be performed at the current time can be determined by the method described in the foregoing embodiment.
Step S23: calculating a difference value between the target utility function corresponding to the resource allocation scheme at the previous moment and the target utility function corresponding to the target resource allocation scheme at the current moment; in this embodiment, it is first determined whether the target utility function corresponding to the resource allocation scheme at this time has converged, specifically, by calculating a difference between a target utility function value corresponding to the resource allocation scheme at the current time and a target utility function value corresponding to the resource allocation scheme at the previous time.
Step S24: judging whether the difference value is smaller than or equal to a preset threshold value; in this embodiment, the preset threshold may be one or zero, or other values may be determined according to user requirements and actual application scenarios, and whether the target utility function corresponding to the resource allocation scheme at the current time has converged may be determined according to a relationship between the difference and the preset threshold.
Step S25: and if the difference is less than or equal to the preset threshold, determining that the target resource allocation scheme is converged, and taking the target resource allocation scheme as the optimal resource allocation scheme. In this embodiment, when the calculated difference is zero, at this time, it may be determined that the target utility function corresponding to the resource allocation scheme at the current time determined by the foregoing embodiment has converged, and the resource allocation scheme at the current time determined by the foregoing embodiment may be used as the optimal resource allocation scheme.
The invention provides a resource slice allocation method, which comprises the following steps: acquiring a resource allocation scheme of a target network at the last moment; according to the resource slice allocation method in the embodiment, a target resource allocation scheme of a target network at the current moment is determined; calculating a difference value between a target utility function corresponding to the resource allocation scheme at the previous moment and a target utility function corresponding to the target resource allocation scheme at the current moment; judging whether the difference value is smaller than or equal to a preset threshold value; and if the difference is less than or equal to the preset threshold, determining that the target resource allocation scheme is converged, and taking the target resource allocation scheme as the optimal resource allocation scheme.
By implementing the method and the device, the resource allocation method at the current moment is determined by combining the flow value in the future time period and the future utility function, and the problems that the resource allocation strategy is optimized only according to the current flow, future changes cannot be responded, and user experience is influenced in the related technology are solved. By combining historical data, the service flow can be predicted according to the service flow change condition in the future network slice, so that the division condition of the future network resources is deduced; and then, by calculating a utility function, the network resource division state at the future moment affects the current division strategy, so that the current optimal strategy is obtained, and the rapid response to the future network demand change in the resource allocation process is ensured.
As an optional embodiment of the present invention, the resource slice allocation method, as shown in fig. 2, further includes:
step S26: and when the difference value is greater than the preset threshold value, returning to execute the steps S11-S15 described in the above embodiment until the difference value is less than or equal to the preset threshold value, determining that the target resource allocation scheme converges, and taking the target resource allocation scheme as the optimal resource allocation scheme. In this embodiment, when the calculated difference is not zero, at this time, it may be determined that the target utility function corresponding to the resource allocation scheme at the current time determined by the foregoing embodiment has not converged, and by continuing to execute the method described in the foregoing embodiment, a next resource allocation scheme is determined until the value of the target utility function corresponding to the resource allocation scheme converges.
In the following, with reference to a specific schematic diagram 3, the transition between states in a practical application scenario of the method according to the embodiment of the present invention is described, for example, the initial state may be s00The initial utility function may be r00(ii) a In taking action a00In the case of (1), i.e. action a00After completion, state s00Transition to state s with a certain probability10When the corresponding utility function is r10(ii) a In taking action a01In the case of (2), state s00Transition to state s with a certain probability11When the corresponding utility function is r11(ii) a In taking action a02In the case of (2), state s00Transition to state s with a certain probability12When the corresponding utility function is r12(ii) a When the utility function corresponding to each action is calculated, that is, the utility function corresponding to each state at the current time is calculated, in this case, the calculation result may be r10＞r11And r is10＞r12Now, the explanation requires action a00State from s00Jump to s10And then, the corresponding utility function is maximum, namely, the spectrum utilization rate is higher and the user service quality is better.
When the slave state is from s00Jump to s10Then, the current state is s10At this point the computation transitions to state s20Or s21And determining the jump action of the next state according to the utility function.
In the following, performance differences between the algorithm corresponding to the embodiment of the present invention and other existing algorithms are described with reference to simulation result diagrams 4 to 7, specifically, the algorithm 1 is a network slice dynamic optimization algorithm based on a proportion, and slice resources are proportionally divided according to current traffic; the algorithm 2 is based on a fair static network slice allocation scheme, which fairly allocates network resources without adjusting according to changes in network traffic. The algorithm 3 is a resource slice allocation method according to the embodiment of the present invention, and specifically is a 5G network slice resource dynamic optimization scheme based on reinforcement learning. The horizontal axis of the diagram indicates the maximum number of content requests per slice.
Through experiments, simulation of the 5G core network is carried out, wherein the core node network nodes are assumed to be in distributed cooperation, so that the network resources in the core network are kept unchanged, and the quantity of request data in the network is increased to compare the utilization rate of the three algorithms on the network resources. Assuming that there are four core network nodes, the number of resources provided by each node conforms to a uniform distribution centered at 40, the core network node resources include computational resources, storage resources, etc., the generation probability of a link between two core network nodes is p, and the number of resources per link conforms to a uniform distribution centered at 55. It is assumed that two core network slices are constructed simultaneously and the number of request packets per slice is randomly generated within a certain range. For example, when the maximum number of content requests per box of slices is N, the number of requests per slice in each link is a random number between 0 and N. The number of packet requests per link can be obtained in turn and then the number of packet requests passing through each node will be obtained. Its value is the sum of the data volumes of each link connected to the node, and the total data volume of the T groups can be generated by the demand of each slice for a period of time in the future.
As shown in fig. 4 and fig. 5, the average resource utilization of the nodes and the average resource utilization of the links of each algorithm in the 5G core network are described; when the maximum content request of each core network slice is increased, namely the data volume of the core network is increased, the resource utilization rate of the three algorithms is improved, and the resource utilization rate of the two dynamic adjustment algorithms is higher than that of the static partitioning algorithm. This is mainly due to the dynamic resource allocation results better matching the network requirements and thus better utilization of the core network resources.
As shown in fig. 6 and fig. 7, the average resource utilization of each algorithm node and the average resource utilization of the link at a future time in the 5G core network are described. The algorithm provided by the invention, namely the 5G network slice resource dynamic optimization scheme based on reinforcement learning, has the highest average resource utilization rate, the static partitioning algorithm based on fairness is the second, and the partitioning algorithm based on proportion is the last, wherein the network slice dynamic optimization scheme based on proportion is the most unstable. Intuitively, the increase in the maximum number of content requests per core network slice mainly brings about two changes: firstly, the total amount of data in a core network is increased, so that resources are more fully utilized; secondly, the traffic demand change range of each core network slice becomes large. This means that when the network traffic changes, the scale-based partitioning algorithm needs to repeatedly adjust the network partitioning result, and the continuous change of the core network resources will consume a lot of time and resources. Therefore, it is slow and unstable to future changes in network demand. Therefore, when the network partitioning result does not match the traffic at the future time, the utilization rate of the resource is low.
An embodiment of the present invention further provides a resource slice allocation apparatus, as shown in fig. 8, including:
a historical data obtaining module 31, configured to obtain historical data of a target network in a preset time period, where the target network includes multiple links and multiple nodes, and the links and the nodes include multiple resource slices; the detailed implementation can be referred to the related description of step S11 in the above method embodiment.
The prediction module 32 is configured to respectively predict flow values of each link in a future preset time period according to the historical data; the detailed implementation can be referred to the related description of step S12 in the above method embodiment.
A matrix determining module 33, configured to determine a link resource partition matrix and a node resource partition matrix according to a preset initial link allocation state and a preset initial node allocation state; the detailed implementation can be referred to the related description of step S13 in the above method embodiment.
A plurality of resource allocation schemes determining modules 34, configured to determine a plurality of resource allocation schemes at the current time according to a topology structure of the target network, a traffic value of each link in a future preset time period, a link resource partition matrix, and a node resource partition matrix, where the resource allocation schemes include a link resource allocation scheme and a node allocation scheme; the detailed implementation can be referred to the related description of step S14 in the above method embodiment.
The calculating module 35 is configured to calculate a target utility function of each resource allocation scheme that meets the reliability condition at the current time; the detailed implementation can be referred to the related description of step S15 in the above method embodiment.
And a target resource allocation scheme determining module 36, configured to use the resource allocation scheme corresponding to the maximum target utility function as the target resource allocation scheme at the current time. The detailed implementation can be referred to the related description of step S16 in the above method embodiment.
The invention provides a resource slice allocation device, comprising: a historical data obtaining module 31, configured to obtain historical data of a target network within a preset time period; the prediction module 32 is configured to respectively predict flow values of each link in a future preset time period according to the historical data; a matrix determining module 33, configured to determine a link resource partition matrix and a node resource partition matrix according to a preset initial link allocation state and a preset initial node allocation state; a plurality of resource allocation schemes determining modules 34, configured to determine a plurality of resource allocation schemes at the current time according to a topology structure of the target network, a traffic value of each link in a future preset time period, a link resource partition matrix, and a node resource partition matrix, where the resource allocation schemes include a link resource allocation scheme and a node allocation scheme; the calculating module 35 is configured to calculate a target utility function of each resource allocation scheme that meets the reliability condition at the current time; and a target resource allocation scheme determining module 36, configured to use the resource allocation scheme corresponding to the maximum target utility function as the target resource allocation scheme at the current time.
By implementing the method and the device, the resource allocation method at the current moment is determined by combining the flow value in the future time period and the future utility function, and the problems that the resource allocation strategy is optimized only according to the current flow, future changes cannot be responded, and user experience is influenced in the related technology are solved. By combining historical data, the service flow can be predicted according to the service flow change condition in the future network slice, so that the division condition of the future network resources is deduced; and then, by calculating a future utility function, the network resource division state at a future moment affects the current division strategy, so that the current optimal strategy is obtained, and the rapid response to the future network demand change in the resource allocation process is ensured.
An embodiment of the present invention further provides a resource slice allocation apparatus, as shown in fig. 9, including:
a resource allocation scheme obtaining module 41, configured to obtain a resource allocation scheme of a target network at a previous time; the detailed implementation can be referred to the related description of step S21 in the above method embodiment.
A determining module 42, configured to determine a target resource allocation scheme of a target network at a current time according to the resource slice allocation method described in the foregoing embodiment; the detailed implementation can be referred to the related description of step S22 in the above method embodiment.
A difference value calculating module 43, configured to calculate a difference value between the target utility function corresponding to the resource allocation scheme at the previous time and the target utility function corresponding to the target resource allocation scheme at the current time; the detailed implementation can be referred to the related description of step S23 in the above method embodiment.
A judging module 44, configured to judge whether the difference is smaller than or equal to a preset threshold; the detailed implementation can be referred to the related description of step S24 in the above method embodiment.
And an optimal resource allocation scheme determining module 45, configured to determine that the target resource allocation scheme is converged if the difference is smaller than or equal to the preset threshold, and use the target resource allocation scheme as the optimal resource allocation scheme. The detailed implementation can be referred to the related description of step S25 in the above method embodiment.
The invention provides a resource slice allocation device, comprising: a resource allocation scheme obtaining module 41, configured to obtain a resource allocation scheme of a target network at a previous time; a determining module 42, configured to determine a target resource allocation scheme of a target network at a current time according to the resource slice allocation method in the foregoing embodiment; a difference value calculating module 43, configured to calculate a difference value between the target utility function corresponding to the resource allocation scheme at the previous time and the target utility function corresponding to the target resource allocation scheme at the current time; a judging module 44, configured to judge whether the difference is smaller than or equal to a preset threshold; and an optimal resource allocation scheme determining module 45, configured to determine that the target resource allocation scheme is converged if the difference is smaller than or equal to the preset threshold, and use the target resource allocation scheme as the optimal resource allocation scheme.
By implementing the method and the device, the resource allocation method at the current moment is determined by combining the flow value in the future time period and the future utility function, and the problems that the resource allocation strategy is optimized only according to the current flow, future changes cannot be responded, and user experience is influenced in the related technology are solved. By combining historical data, the service flow can be predicted according to the service flow change condition in the future network slice, so that the division condition of the future network resources is deduced; and then, by calculating a future utility function, the network resource division state at a future moment affects the current division strategy, so that the current optimal strategy is obtained, and the rapid response to the future network demand change in the resource allocation process is ensured.
An embodiment of the present invention further provides a computer device, as shown in fig. 10, the computer device may include a processor 51 and a memory 52, where the processor 51 and the memory 52 may be connected by a bus 50 or in another manner, and fig. 10 takes the example of connection by the bus 50.
The processor 51 may be a Central Processing Unit (CPU). The Processor 51 may also be other general purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field Programmable Gate Arrays (FPGAs) or other Programmable logic devices, discrete Gate or transistor logic devices, discrete hardware components, or combinations thereof.
The memory 52, which is a non-transitory computer readable storage medium, may be used for storing non-transitory software programs, non-transitory computer executable programs, and modules, such as program instructions/modules corresponding to the resource slice allocation method in the embodiment of the present invention. The processor 51 executes various functional applications and data processing of the processor by running non-transitory software programs, instructions and modules stored in the memory 52, that is, implements the resource slice allocation method in the above-described method embodiments.
The memory 52 may include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function; the storage data area may store data created by the processor 51, and the like. Further, the memory 52 may include high speed random access memory, and may also include non-transitory memory, such as at least one magnetic disk storage device, flash memory device, or other non-transitory solid state storage device. In some embodiments, the memory 52 may optionally include memory located remotely from the processor 51, and these remote memories may be connected to the processor 51 via a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
The one or more modules are stored in the memory 52 and when executed by the processor 51 perform a resource slice allocation method as in the embodiment of fig. 1 or fig. 2.
The details of the computer device can be understood by referring to the corresponding related descriptions and effects in the embodiments shown in fig. 1 or fig. 2, and are not described herein again.
The embodiment of the present invention further provides a non-transitory computer readable medium, where the non-transitory computer readable storage medium stores a computer instruction, and the computer instruction is used to enable a computer to execute the resource slice allocation method described in any one of the above embodiments, where the storage medium may be a magnetic Disk, an optical Disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a Flash Memory (Flash Memory), a Hard Disk (Hard Disk Drive, abbreviated as HDD), or a Solid-State Drive (SSD); the storage medium may also comprise a combination of memories of the kind described above.
It should be understood that the above examples are only for clarity of illustration and are not intended to limit the embodiments. Other variations and modifications will be apparent to persons skilled in the art in light of the above description. And are neither required nor exhaustive of all embodiments. And obvious variations or modifications therefrom are within the scope of the invention.
Claims (11)
1. A method for resource slice allocation, comprising:
acquiring historical data of a target network in a preset time period, wherein the target network comprises a plurality of links and a plurality of nodes, and the links and the nodes comprise a plurality of resource slices;
respectively predicting the flow value of each link in a future preset time period according to the historical data;
determining a link resource partition matrix and a node resource partition matrix according to a preset initial link allocation state and a preset initial node allocation state;
determining a plurality of resource allocation schemes at the current moment according to a topological structure of a target network, a flow value of each link in a future preset time period, a link resource partition matrix and a node resource partition matrix, wherein the resource allocation schemes comprise a link resource allocation scheme and a node allocation scheme;
respectively calculating target utility functions of all resource allocation schemes meeting the reliability conditions at the current moment;
and taking the resource allocation scheme corresponding to the maximum target utility function as the target resource allocation scheme at the current moment.
2. The method of claim 1, wherein determining whether each resource allocation scheme satisfies the reliability condition at the current time comprises:
respectively acquiring a time delay value and a frequency spectrum utilization value of each resource allocation scheme;
and judging whether the time delay value of each resource allocation scheme is smaller than the target time delay value or not, and judging whether the frequency spectrum utilization rate of each resource allocation scheme is larger than the target frequency spectrum utilization rate or not.
3. The method of claim 1, further comprising:
and distributing a plurality of resource slices on the target network to each network slice at the current moment according to the target resource distribution scheme at the current moment.
4. The method of claim 1, wherein the step of calculating the target utility function for each resource allocation scheme comprises:
calculating a link utility function and a node utility function of each resource allocation scheme;
and determining the target utility function according to the link utility function and the node utility function.
5. The method of claim 4, wherein the target utility function is determined by the following formula:
Vt＝Vm+Vn,
wherein, VtotalRepresenting the target utility function, VtRepresenting the total utility function, gamma a predetermined state transition factor, VmRepresenting the link utility function, VnRepresenting the node utility function;
calculating the link utility function and the node utility function by the following formulas:
Vπ(s)＝Eπ{Gt|St＝s},
R＝α*SE+β*QoE,
wherein QoE represents a user service experience value, SE represents spectrum efficiency, alpha and beta represent different weight factors, and R represents an incentive function of the resource allocation scheme at the current momentNumber, GtRepresenting a cumulative reward function.
6. A method for resource slice allocation, comprising:
acquiring a resource allocation scheme of a target network at the last moment;
the resource slice allocation method according to any of claims 1-5, determining a target resource allocation scheme for a current time of a target network;
calculating a difference value between the target utility function corresponding to the resource allocation scheme at the previous moment and the target utility function corresponding to the target resource allocation scheme at the current moment;
judging whether the difference value is smaller than or equal to a preset threshold value;
and if the difference is smaller than or equal to a preset threshold value, determining that the target resource allocation scheme is converged, and taking the target resource allocation scheme as an optimal resource allocation scheme.
7. The method of claim 6, further comprising:
when the difference is larger than a preset threshold, returning to the step of executing the resource slice allocation method according to any one of claims 1 to 5, determining the target resource allocation scheme of the target network at the current moment to the step of judging whether the difference is smaller than or equal to the preset threshold until the difference is smaller than or equal to the preset threshold, determining that the target resource allocation scheme is converged, and taking the target resource allocation scheme as an optimal resource allocation scheme.
8. A resource slice allocation apparatus, comprising:
the historical data acquisition module is used for acquiring historical data of a target network in a preset time period, the target network comprises a plurality of links and a plurality of nodes, and the links and the nodes comprise a plurality of resource slices;
the prediction module is used for respectively predicting the flow value of each link in the future preset time period according to the historical data;
the matrix determination module is used for determining a link resource partition matrix and a node resource partition matrix according to a preset initial link allocation state and a preset initial node allocation state;
the system comprises a plurality of resource allocation scheme determination modules, a resource allocation module and a resource allocation module, wherein the resource allocation scheme determination modules are used for determining a plurality of resource allocation schemes at the current moment according to a topological structure of a target network, a flow value of each link in a future preset time period, a link resource partition matrix and a node resource partition matrix, and the resource allocation schemes comprise a link resource allocation scheme and a node allocation scheme;
the calculation module is used for respectively calculating a target utility function of each resource allocation scheme meeting the reliability condition at the current moment;
and the target resource allocation scheme determining module is used for taking the resource allocation scheme corresponding to the maximum target utility function as the target resource allocation scheme at the current moment.
9. A resource slice allocation apparatus, comprising:
the resource allocation scheme acquisition module is used for acquiring a resource allocation scheme of a target network at the last moment;
a determining module, configured to determine a target resource allocation scheme of a target network at a current time according to the resource slice allocation method according to any one of claims 1 to 5;
a difference value calculating module, configured to calculate a difference value between the target utility function corresponding to the resource allocation scheme at the previous time and the target utility function corresponding to the target resource allocation scheme at the current time;
the judging module is used for judging whether the difference value is smaller than or equal to a preset threshold value or not;
and the optimal resource allocation scheme determining module is used for determining the convergence of the target resource allocation scheme if the difference is smaller than or equal to a preset threshold value, and taking the target resource allocation scheme as the optimal resource allocation scheme.
10. A computer device, comprising: at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores instructions executable by the one processor to cause the at least one processor to perform the steps of the resource slice allocation method of any one of claims 1-7.
11. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the resource slice allocation method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011535215.5A CN112737823A (en) | 2020-12-22 | 2020-12-22 | Resource slice allocation method and device and computer equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011535215.5A CN112737823A (en) | 2020-12-22 | 2020-12-22 | Resource slice allocation method and device and computer equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112737823A true CN112737823A (en) | 2021-04-30 |
Family
ID=75604278
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011535215.5A Pending CN112737823A (en) | 2020-12-22 | 2020-12-22 | Resource slice allocation method and device and computer equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112737823A (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113329414A (en) * | 2021-06-07 | 2021-08-31 | 深圳聚创致远科技有限公司 | Smart power grid slice distribution method based on reinforcement learning |
| CN113612635A (en) * | 2021-07-29 | 2021-11-05 | 西安电子科技大学 | Network slice instance resource allocation method based on horizontal/vertical expansion and contraction combination |
| CN113705959A (en) * | 2021-05-11 | 2021-11-26 | 北京邮电大学 | Network resource allocation method and electronic equipment |
| CN113726571A (en) * | 2021-08-30 | 2021-11-30 | 电子科技大学 | Network resource dynamic allocation method based on network slice |
| CN113840333A (en) * | 2021-08-16 | 2021-12-24 | 国网河南省电力公司信息通信公司 | Power grid resource allocation method and device, electronic equipment and storage medium |
| CN114158073A (en) * | 2021-11-29 | 2022-03-08 | 中国联合网络通信集团有限公司 | Network slice deployment method, device, equipment and storage medium |
| CN115174410A (en) * | 2022-07-27 | 2022-10-11 | 阿里巴巴（中国）有限公司 | Resource allocation method, device and equipment |
| WO2022239321A1 (en) * | 2021-05-11 | 2022-11-17 | 株式会社Ｋｄｄｉ総合研究所 | Control device, resource allocation control method, and computer program |
| CN115696466A (en) * | 2022-09-23 | 2023-02-03 | 广州爱浦路网络技术有限公司 | S-NSSAI priority determination method and device |
| CN116112448A (en) * | 2023-02-28 | 2023-05-12 | 保信科技(杭州)有限公司 | Broadband resource allocation scheduling method |
| WO2024007208A1 (en) * | 2022-07-06 | 2024-01-11 | Huawei Technologies Co., Ltd. | Systems and methods to minimize resource utilization across multiple domains to assure quality of service |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20180317133A1 (en) * | 2017-04-28 | 2018-11-01 | NEC Laboratories Europe GmbH | Method and system for network slice allocation |
| CN108965024A (en) * | 2018-08-01 | 2018-12-07 | 重庆邮电大学 | A kind of virtual network function dispatching method of the 5G network slice based on prediction |
| CN109167787A (en) * | 2018-09-06 | 2019-01-08 | 电子科技大学 | A kind of mobile edge calculations safety in network calculates the method for optimizing resources of unloading |
| CN110519783A (en) * | 2019-09-26 | 2019-11-29 | 东华大学 | 5G network based on enhancing study is sliced resource allocation methods |
| CN110881199A (en) * | 2019-12-06 | 2020-03-13 | 国网河南省电力公司许昌供电公司 | Dynamic allocation method, device and system for network slice resources |
| CN111866953A (en) * | 2019-04-26 | 2020-10-30 | 中国移动通信有限公司研究院 | Network resource allocation method, device and storage medium |
-
2020
- 2020-12-22 CN CN202011535215.5A patent/CN112737823A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20180317133A1 (en) * | 2017-04-28 | 2018-11-01 | NEC Laboratories Europe GmbH | Method and system for network slice allocation |
| CN108965024A (en) * | 2018-08-01 | 2018-12-07 | 重庆邮电大学 | A kind of virtual network function dispatching method of the 5G network slice based on prediction |
| CN109167787A (en) * | 2018-09-06 | 2019-01-08 | 电子科技大学 | A kind of mobile edge calculations safety in network calculates the method for optimizing resources of unloading |
| CN111866953A (en) * | 2019-04-26 | 2020-10-30 | 中国移动通信有限公司研究院 | Network resource allocation method, device and storage medium |
| CN110519783A (en) * | 2019-09-26 | 2019-11-29 | 东华大学 | 5G network based on enhancing study is sliced resource allocation methods |
| CN110881199A (en) * | 2019-12-06 | 2020-03-13 | 国网河南省电力公司许昌供电公司 | Dynamic allocation method, device and system for network slice resources |
Cited By (20)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2022239321A1 (en) * | 2021-05-11 | 2022-11-17 | 株式会社Ｋｄｄｉ総合研究所 | Control device, resource allocation control method, and computer program |
| CN113705959A (en) * | 2021-05-11 | 2021-11-26 | 北京邮电大学 | Network resource allocation method and electronic equipment |
| CN113705959B (en) * | 2021-05-11 | 2023-08-15 | 北京邮电大学 | Network resource allocation method and electronic equipment |
| CN113329414A (en) * | 2021-06-07 | 2021-08-31 | 深圳聚创致远科技有限公司 | Smart power grid slice distribution method based on reinforcement learning |
| CN113329414B (en) * | 2021-06-07 | 2023-01-10 | 深圳聚创致远科技有限公司 | Smart power grid slice distribution method based on reinforcement learning |
| CN113612635A (en) * | 2021-07-29 | 2021-11-05 | 西安电子科技大学 | Network slice instance resource allocation method based on horizontal/vertical expansion and contraction combination |
| CN113612635B (en) * | 2021-07-29 | 2022-08-12 | 西安电子科技大学 | Network slice instance resource allocation method based on combination of horizontal and vertical expansion and contraction |
| CN113840333B (en) * | 2021-08-16 | 2023-11-10 | 国网河南省电力公司信息通信公司 | Power grid resource allocation method and device, electronic equipment and storage medium |
| CN113840333A (en) * | 2021-08-16 | 2021-12-24 | 国网河南省电力公司信息通信公司 | Power grid resource allocation method and device, electronic equipment and storage medium |
| US11792137B2 (en) | 2021-08-30 | 2023-10-17 | University Of Electronic Science And Technology Of China | Dynamic network resource allocation method based on network slicing |
| CN113726571B (en) * | 2021-08-30 | 2022-07-12 | 电子科技大学 | Network resource dynamic allocation method based on network slice |
| CN113726571A (en) * | 2021-08-30 | 2021-11-30 | 电子科技大学 | Network resource dynamic allocation method based on network slice |
| CN114158073A (en) * | 2021-11-29 | 2022-03-08 | 中国联合网络通信集团有限公司 | Network slice deployment method, device, equipment and storage medium |
| CN114158073B (en) * | 2021-11-29 | 2024-01-09 | 中国联合网络通信集团有限公司 | Network slice deployment method, device, equipment and storage medium |
| WO2024007208A1 (en) * | 2022-07-06 | 2024-01-11 | Huawei Technologies Co., Ltd. | Systems and methods to minimize resource utilization across multiple domains to assure quality of service |
| CN115174410A (en) * | 2022-07-27 | 2022-10-11 | 阿里巴巴（中国）有限公司 | Resource allocation method, device and equipment |
| CN115696466A (en) * | 2022-09-23 | 2023-02-03 | 广州爱浦路网络技术有限公司 | S-NSSAI priority determination method and device |
| CN115696466B (en) * | 2022-09-23 | 2023-06-06 | 广州爱浦路网络技术有限公司 | Method and device for determining priority of S-NSSAI |
| CN116112448A (en) * | 2023-02-28 | 2023-05-12 | 保信科技(杭州)有限公司 | Broadband resource allocation scheduling method |
| CN116112448B (en) * | 2023-02-28 | 2024-05-31 | 保信科技(杭州)有限公司 | Broadband resource allocation scheduling method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112737823A (en) | Resource slice allocation method and device and computer equipment | |
| CN111953759B (en) | Collaborative computing task unloading and transferring method and device based on reinforcement learning | |
| CN111953758B (en) | Edge network computing unloading and task migration method and device | |
| US11431645B2 (en) | Method and apparatus for handling network congestion, and system | |
| US8144590B2 (en) | Distributed resource allocation in stream processing systems | |
| CN112822050B (en) | Method and apparatus for deploying network slices | |
| CN110764912A (en) | Self-adaptive task scheduler and method | |
| CN109614215A (en) | Stream scheduling method, device, equipment and medium based on deeply study | |
| CN112003660B (en) | Dimension measurement method of resources in network, calculation force scheduling method and storage medium | |
| CN115794407A (en) | Computing resource allocation method and device, electronic equipment and nonvolatile storage medium | |
| WO2020134133A1 (en) | Resource allocation method, substation, and computer-readable storage medium | |
| Mason et al. | Using distributed reinforcement learning for resource orchestration in a network slicing scenario | |
| Chen et al. | Latency minimization for mobile edge computing networks | |
| Zheng et al. | Adaptive resource scheduling mechanism in P2P file sharing system | |
| CN117938755B (en) | Data flow control method, network switching subsystem and intelligent computing platform | |
| Xu et al. | Schedule or wait: Age-minimization for IoT big data processing in MEC via online learning | |
| CN115580882A (en) | Dynamic network slice resource allocation method and device, storage medium and electronic equipment | |
| CN113271221B (en) | Network capacity opening method and system and electronic equipment | |
| KR20130060350A (en) | Method and apparatus for scheduling communication traffic in atca-based equipment | |
| Li et al. | Optimal service selection and placement based on popularity and server load in multi-access edge computing | |
| CN116302404B (en) | Resource decoupling data center-oriented server non-perception calculation scheduling method | |
| CN115562841B (en) | Cloud video service self-adaptive resource scheduling system and method | |
| Zeydan et al. | A multi-criteria decision making approach for scaling and placement of virtual network functions | |
| CN110138863B (en) | Self-adaptive consistency protocol optimization method based on Multi-Paxos grouping | |
| Wang et al. | Offloading Strategies for Mobile Edge Computing Based on Multi-Attribute Preferences in Smart Grids |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210430 |
|
| RJ01 | Rejection of invention patent application after publication |