5G network resource dynamic allocation method

Abstract

The invention relates to the field of 5G network resource management, in particular to a 5G network resource dynamic allocation method. The method comprises the following steps: acquiring resource requirements and available resources of nodes in a 5G network, forming resource states of the nodes, and further forming a resource state matrix; a multi-dimensional harmonic mapping distribution algorithm is introduced to update the resource requirements of the nodes, and the resource state matrix is further updated; calculating the global resource pressure of the node based on the updated resource state matrix; performing nonlinear resource adjustment optimization on the updated resource demand to obtain a final resource demand; and performing resource allocation based on the final resource demand. The problem that many scheduling algorithms cannot fully consider resource interaction and mutual influence between nodes is solved; in the prior art, in a high-dynamic environment, the reaction speed is low, and resource allocation is difficult to adjust according to a network state changing in real time; and in a peak load scene, an existing resource allocation algorithm is difficult to realize rapid balance of network resources.

Description

Technical Field

[0001] The present invention relates to the field of 5G network resource management, and in particular to a method for dynamically allocating 5G network resources. Background Art

[0002] With the rapid popularization of 5G networks, more and more devices are connected to the network, including base stations, user equipment, Internet of Things (IoT) devices, smart terminals, etc. The demand for network resources from the above devices is highly dynamic and complex, especially in high-concurrency and high-data-rate application scenarios, such as high-definition video transmission, virtual reality, remote control, etc., making the efficient allocation of network resources a core issue in 5G network management. Traditional network resource management methods often rely on static allocation modes or simple time slice rotation mechanisms. This method can work normally in a low-load environment, but in a high-load and frequently dynamically changing 5G network environment, traditional network resource management methods cannot effectively adapt to the frequently fluctuating resource demands, which can easily lead to low network resource utilization, serious resource waste, or unbalanced resource allocation.

[0003] In addition, existing resource allocation technologies also have many limitations when facing the complexity of 5G networks. First, many scheduling algorithms fail to fully consider the resource interactions and mutual influences between nodes. The differences between node resource requirements and available resources are simply processed linearly, which makes it difficult to achieve efficient resource scheduling when resource requirements are more complex or the differences between nodes are large. Secondly, when facing a highly dynamic environment, the existing technology has a slow response speed and it is difficult to adjust resource allocation according to the real-time changing network status. It is easy for some nodes to be overloaded or insufficient. Especially in peak load scenarios, the existing resource allocation algorithms are difficult to achieve rapid balancing of network resources, which can easily lead to resource bottlenecks, thereby affecting the performance and user experience of the entire network. The above-mentioned deficiencies have brought huge challenges to the stability of 5G network performance and resource utilization. Summary of the invention

[0004] The present invention provides a 5G network resource dynamic allocation method to solve the problem that many scheduling algorithms fail to fully consider the resource interaction and mutual influence between nodes, and the difference between node resource demand and available resources is simply processed linearly, resulting in difficulty in achieving efficient resource scheduling when resource demand is more complex or the difference between nodes is large; the existing technology has a slow response speed when facing a highly dynamic environment, and it is difficult to adjust resource allocation according to the real-time changing network status, which is prone to resource overload or resource shortage of some nodes; and in peak load scenarios, the existing resource allocation algorithm is difficult to achieve rapid balancing of network resources, which is prone to resource bottlenecks, thereby affecting the performance of the entire network and user experience.

[0005] A 5G network resource dynamic allocation method comprises the following steps:

[0006] S1: Obtain the resource requirements and available resources of nodes in the 5G network to form the resource status of the nodes; construct a resource status matrix based on the resource status of all nodes; introduce a multi-dimensional harmonic mapping allocation algorithm to update the resource requirements of the nodes; update the resource status matrix according to the updated resource requirements of the nodes;

[0007] S2: Based on the updated resource state matrix, calculate the global resource pressure of the node; construct a global energy function, and combine the global resource pressure to perform nonlinear resource adjustment optimization on the updated resource demand to obtain the final resource demand; allocate resources based on the final resource demand.

[0008] Preferably, the S1 specifically includes:

[0009] The specific process of the multi-dimensional harmonic mapping allocation algorithm is as follows:

[0010] The interaction coefficients between nodes are calculated, and the harmonic mapping value of each node is calculated using the interaction coefficients and the resource state matrix; the resource requirements of the nodes are updated according to the harmonic mapping value.

[0011] Preferably, the S1 specifically includes:

[0012] The formula for harmonic mapping value is as follows:

[0013]

[0014] in, is the harmonic mapping value of node i; S i is the resource status of node i; i and j are index variables of nodes in the 5G network; N is the total number of nodes in the 5G network; α ij represents the interaction coefficient between nodes; R i is the resource requirement of node i; R j is the resource requirement of node j; C i is the available resources of node i; C j is the available resources of node j; ε is a constant to prevent the denominator from being zero, which is 10 -9 ; exp is the exponential operation; is an exponential decay term.

[0015] Preferably, the S1 specifically includes:

[0016] The resource requirement update formula is as follows:

[0017]

[0018] Among them, R′ iis the updated resource requirement of node i; λ is the adjustment coefficient.

[0019] Preferably, the S2 specifically includes:

[0020] The global resource pressure calculation formula is:

[0021]

[0022] Among them, P i is the global resource pressure of node i; R′ j is the updated resource requirement of node j.

[0023] Preferably, the S2 specifically includes:

[0024] Based on the updated resource status matrix, a global energy function is constructed; the global energy function combines the interaction coefficient between nodes, the nonlinear difference between resource demand and available resources, and measures the balance of 5G network resource allocation.

[0025] Preferably, the S2 specifically includes:

[0026] Nonlinear resource adjustment optimization is based on the gradient descent method, introducing fractional power and hyperbolic sine functions, and combining global resource pressure to obtain the final resource demand of the node by minimizing the global energy function.

[0027] Preferably, the S2 specifically includes:

[0028] The update formula for nonlinear resource adjustment optimization is:

[0029]

[0030] in, is the resource requirement of node i after nonlinear resource adjustment optimization; η is the learning rate; is the resource requirement R′ of the global energy function E(S) for node i i gradient.

[0031] The beneficial effects of the technical solution of the present invention are:

[0032] 1. The multi-dimensional harmonic mapping allocation algorithm proposed in the present invention realizes accurate dynamic allocation based on node resource requirements and available resources by calculating the interaction coefficients between nodes and the harmonic mapping values ​​of nodes. By dynamically adjusting resource requirements according to network load, network resource utilization is maximized to ensure the flexibility and accuracy of resource allocation. Compared with traditional static allocation methods, the present invention can better adapt to real-time changes in 5G networks and avoid resource waste and bottlenecks.

[0033] 2. The global energy function is introduced to combine the interaction coefficient between nodes, the nonlinear difference between resource demand and available resources. It not only takes into account the resource demand of the node itself, but also comprehensively optimizes the global resource allocation by calculating the relationship between nodes in the entire network to ensure that resources are evenly distributed in the network; it reduces network congestion and delay problems, improves the overall performance of the network, and significantly reduces the probability of resource bottlenecks, especially in high-load scenarios.

[0034] 3. Through nonlinear resource adjustment and optimization, the resource requirements of nodes can be adaptively adjusted so that the resource allocation in the network always remains optimal; through the gradient descent method, the adjustment of node resource requirements is dynamically responded to, and finally the optimal resource allocation is provided through full network optimization and iteration; it not only improves the utilization efficiency of network resources, but also effectively balances the network load, ensuring that all nodes in the network can obtain fair resource allocation, thereby improving the stability and reliability of the 5G network. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 This is a flow chart of a 5G network resource dynamic allocation method described in the present invention. DETAILED DESCRIPTION

[0036] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined purpose of the invention, the technical scheme in the embodiment of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiment of the present invention. Obviously, the described embodiment is only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0037] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0038] The following is a detailed description of a specific scheme of a 5G network resource dynamic allocation method provided by the present invention in conjunction with the accompanying drawings.

[0039] See attached Figure 1 , which shows a flow chart of a 5G network resource dynamic allocation method provided by an embodiment of the present invention, the method comprising the following steps:

[0040] S1: Obtain the resource requirements and available resources of nodes in the 5G network to form the resource status of the nodes; construct a resource status matrix based on the resource status of all nodes; introduce a multi-dimensional harmonic mapping allocation algorithm to update the resource requirements of the nodes; update the resource status matrix according to the updated resource requirements of the nodes;

[0041] It is defined that there are N nodes in the 5G network, the resource demand of the node is represented by Ri, the available resources are represented by Ci, and the resource demand and available resources constitute the resource status of the node, which is represented by (Ri, Ci), where i is the index variable of the node.

[0042] Specifically, the node is a network unit in the 5G network, such as a base station, a user device (such as a mobile phone, an IoT device) or an entity that requires communication resources; the resource demand is the amount of network resources currently required by the node, which is related to communication bandwidth, computing resources, storage requirements, etc.; the available resources are the amount of network resources currently available to the node, indicating the total amount of resources that the node can obtain from the 5G network, such as the wireless spectrum bandwidth currently allocable to the base station, and the transmission rate and connection resources obtained by the user device from the 5G network; the resource status is a combination of the node resource demand and available resources, that is, the resource allocation situation of each node.

[0043] The resource status of all nodes in the 5G network constitutes a resource status matrix, which contains the resource requirements and available resources of all nodes in the current 5G network. Each row of the resource status matrix corresponds to the resource status of a node. Since the resource requirements and available resources of each node are dynamically changing, it is necessary to dynamically allocate network resources according to the real-time status of the 5G network.

[0044] The embodiment of the present invention proposes a multi-dimensional harmonic mapping allocation algorithm, which updates the resource requirements of the nodes in combination with the interaction coefficients between nodes and the resource status matrix, reflects the pressure of the nodes in resource allocation, further performs nonlinear resource adjustment optimization on the resource requirements of the updated nodes, and allocates resources based on the final resource requirements obtained by optimization. The algorithm is suitable for dynamic network resource allocation scenarios, can effectively improve resource utilization, balance network load, and reduce bottlenecks and delays caused by uneven resource allocation.

[0045] The specific process of the multi-dimensional harmonic mapping allocation algorithm is as follows:

[0046] First, the interaction coefficients between nodes are calculated, and the interaction coefficients and resource status matrix are used to further calculate the harmonic mapping value of each node to adjust the resource requirements.

[0047] The interaction coefficient between the nodes is α ij It is the interaction coefficient between nodes i and j. The interaction coefficient is expressed as a scalar, ranging from 0 to α. ij ≤1. The closer the value is to 1, the stronger the association between nodes. The interaction coefficient is based on the physical distance between nodes or the network topology, and is obtained by calculating the Euclidean distance between nodes or the similarity of business requirements. The specific calculation method of the interaction coefficient is selected according to the specific implementation scenario and is not limited here.

[0048] The harmonic mapping value is used to measure the pressure of the node in resource allocation. Specifically, the harmonic mapping value takes into account the resource status relationship between the node and other nodes, as well as the difference between the node's own resource demand and available resources, and comprehensively measures the pressure of the node in resource allocation. The larger the harmonic mapping value, the greater the imbalance between the node's resource demand and the available resources; the smaller the harmonic mapping value or close to zero, the closer the node's resource demand is to the available resources.

[0049] The formula for harmonic mapping value is as follows:

[0050]

[0051] in, is the harmonic mapping value of node i; S i is the resource status of node i; i and j are index variables of nodes in the 5G network; N is the total number of nodes in the 5G network, that is, the total number of nodes participating in resource allocation in the 5G network; R i is the resource requirement of node i; R j is the resource requirement of node j; C i is the available resources of node i; C j is the available resources of node j; ε is a constant to prevent the denominator from being zero, which is 10 -9 ; exp is the exponential operation; is an exponential decay term, which is used to decay the difference between the resource demand and available resources of node i.

[0052] The harmonic mapping formula introduces fractional power and exponential decay terms. When there is a large numerical difference between resource demand and available resources, the power operation is used to amplify the impact of the difference on resource demand. At the same time, the exponential decay term can weaken the impact of a small difference when the difference between resource demand and available resources is small.

[0053] The resource requirements of the node are updated according to the harmonized mapping value so that the updated resource requirements can reflect the pressure of the node in resource allocation. The resource requirement update formula is as follows:

[0054]

[0055] Among them, R′ i is the updated resource requirement of node i, that is, the current resource requirement of node i; R i is the resource requirement of node i before updating; λ is the adjustment coefficient used to control the reconciliation mapping value The impact on resource demand is set according to the network load in the specific implementation scenario; It is the average value of the ratio of resource demand to available resources of all nodes in the network, which measures the relationship between resource demand and available resources of network nodes;

[0056] The resource demand update formula introduces the exponential adjustment term of the harmonic mapping value and the average value of the ratio of the whole network resource demand to the available resources, so that the updated resource demand can better reflect the pressure of the node in resource allocation and provide a basis for the optimization of the whole network resources.

[0057] Finally, the resource status matrix is ​​updated according to the updated resource demand, and the corresponding resource demand value in the resource status matrix is ​​replaced, so that the resource status matrix always reflects the current resource allocation situation.

[0058] S2: Based on the updated resource state matrix, calculate the global resource pressure of the node; construct a global energy function, and combine the global resource pressure to perform nonlinear resource adjustment optimization on the updated resource demand to obtain the final resource demand; allocate resources based on the final resource demand.

[0059] First, based on the updated resource status matrix, the global resource pressure of the node is calculated to reflect the pressure of each node's resource demand relative to the resource status of the entire network and provide a basis for the subsequent optimization process. Specifically, the larger the value of global resource pressure, the heavier the resource pressure of the node in the entire network, and the greater the pressure of resource demand relative to available resources. The global resource pressure calculation formula is:

[0060]

[0061] Among them, P i is the global resource pressure of node i, which indicates the resource demand of node i relative to the global resource pressure of other nodes; R′ j is the updated resource requirement of node j.

[0062] At the same time, based on the updated resource state matrix, a global energy function is constructed to reflect the balance of the current network resource allocation and provide a basis for the subsequent optimization process. The global energy function combines the interaction coefficient between nodes, the nonlinear difference between resource demand and available resources, and measures the balance of the entire network resource allocation. The smaller the value of the global energy function, the more balanced the resource allocation.

[0063] The formula of the global energy function is:

[0064]

[0065] Where E(S) is the global energy function, which is used to measure the balance of resource allocation in the entire network; s is the resource state matrix; cosh is the hyperbolic cosine function, which is used to smooth the nonlinear effect of the difference between resource demand and available resources; R′ iis the updated resource demand of node i. The global energy function introduces the hyperbolic cosine function to smooth the difference between resource demand and available resources, and introduces fractional power to enhance the sensitivity to resource differences, so that the global energy function can reflect the resource imbalance of the entire network.

[0066] Next, the updated resource requirements are optimized by nonlinear resource adjustment to obtain the final resource requirements. The nonlinear resource adjustment optimization is based on the gradient descent method, introduces fractional power and hyperbolic sine function, and combines global resource pressure to obtain the final resource requirements of the node by minimizing the global energy function. Based on the final resource requirements, network resources are dynamically allocated to achieve network resource balance, improve resource utilization efficiency, reduce bottlenecks and delays caused by uneven resource allocation, and optimize the performance and stability of the 5G network.

[0067] The update formula for nonlinear resource adjustment optimization is:

[0068]

[0069] in, is the resource demand of node i after nonlinear resource adjustment optimization; R′ i is the current resource demand of node i; η is the learning rate, which is used to control the resource update step size and is obtained through experimental tuning; is the resource requirement R′ of the global energy function E(S) for node i i gradient.

[0070] gradient The calculation formula of introduces the hyperbolic sine function to enhance the dynamic response to the node resource demand update, and can adjust the update amplitude when the resource demand changes greatly. The specific calculation formula is as follows:

[0071]

[0072] Among them, sinh is a hyperbolic sine function, which is used to enhance the dynamic regulation effect of changes in resource demand.

[0073] The change threshold and maximum number of iterations are set according to the specific implementation scenario. When the change of the global energy function is less than the change threshold or the number of iterations reaches the maximum number of iterations, the nonlinear resource adjustment optimization is stopped, and the resource demand of the node updated by the nonlinear resource adjustment optimization is output as the final resource demand for resource allocation.

[0074] Match the final resource demand of the node with the available resources. If the final resource demand of the node is less than or equal to the available resources, allocate the corresponding resources to the node; if the final resource demand of the node is greater than the available resources, select a fair allocation mechanism based on the specific implementation scenario, such as proportionally reducing the resource demand of each node to ensure that all nodes in the network can obtain reasonable allocation under limited resources. After the resource allocation is completed, the resource status matrix is ​​updated according to the resource allocation results to reflect the current resource allocation status of the network.

[0075] In summary, a 5G network resource dynamic allocation method is completed.

[0076] The order of the embodiments of the invention is for description only and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0077] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

[0078] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that the technical solutions described in the above embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention.

Claims

1. A 5G network resource dynamic allocation method, characterized in that: The following steps are involved: S1: Obtain the resource requirements and available resources of nodes in the 5G network to form the resource status of the nodes; construct a resource status matrix based on the resource status of all nodes; introduce a multi-dimensional harmonic mapping allocation algorithm to update the resource requirements of the nodes; update the resource status matrix according to the updated resource requirements of the nodes; S2: Calculate the global resource pressure of the node based on the updated resource status matrix; Construct a global energy function, and combine it with the global resource pressure to perform nonlinear resource adjustment optimization on the updated resource demand to obtain the final resource demand; Resource allocation is performed based on final resource requirements.

2. The 5G network resource dynamic allocation method according to claim 1, characterized in that: The S1 specifically includes: The specific process of the multi-dimensional harmonic mapping allocation algorithm is as follows: calculate the interaction coefficient between nodes, use the interaction coefficient and the resource state matrix to calculate the harmonic mapping value of each node; update the resource requirements of the node according to the harmonic mapping value.

3. The 5G network resource dynamic allocation method according to claim 2, characterized in that: The S1 specifically includes: The formula for harmonic mapping value is as follows: in, is the harmonic mapping value of node i; S i is the resource status of node i; i and j are index variables of nodes in the 5G network; N is the total number of nodes in the 5G network; α ij represents the interaction coefficient between nodes; R i is the resource requirement of node i; R j is the resource requirement of node j; C i is the available resources of node i; C j is the available resources of node j; ε is a constant to prevent the denominator from being zero, which is 10 -9 ; exp is the exponential operation; is an exponential decay term.

4. The 5G network resource dynamic allocation method according to claim 3, characterized in that: The S1 specifically includes: The resource requirement update formula is as follows: Among them, R′ i is the updated resource requirement of node i; λ is the adjustment coefficient.

5. The 5G network resource dynamic allocation method according to claim 1, characterized in that: The S2 specifically includes: The global resource pressure calculation formula is: Among them, P i is the global resource pressure of node i; R′ j is the updated resource requirement of node j.

6. The 5G network resource dynamic allocation method according to claim 1, characterized in that: The S2 specifically includes: Based on the updated resource status matrix, a global energy function is constructed; the global energy function combines the interaction coefficient between nodes, the nonlinear difference between resource demand and available resources, and measures the balance of 5G network resource allocation.

7. The 5G network resource dynamic allocation method according to claim 6, characterized in that: The S2 specifically includes: Nonlinear resource adjustment optimization is based on the gradient descent method, introducing fractional power and hyperbolic sine functions, and combining global resource pressure to obtain the final resource demand of the node by minimizing the global energy function.

8. The 5G network resource dynamic allocation method according to claim 7, characterized in that: The S2 specifically includes: The update formula for nonlinear resource adjustment optimization is: in, is the resource requirement of node i after nonlinear resource adjustment optimization; η is the learning rate; It is the gradient of the global energy function E(S) with respect to the resource demand R′i of node i.

Images