CN113507413B - Route optimization method and device and computing equipment - Google Patents
Route optimization method and device and computing equipment Download PDFInfo
- Publication number
- CN113507413B CN113507413B CN202110831817.3A CN202110831817A CN113507413B CN 113507413 B CN113507413 B CN 113507413B CN 202110831817 A CN202110831817 A CN 202110831817A CN 113507413 B CN113507413 B CN 113507413B
- Authority
- CN
- China
- Prior art keywords
- link
- candidate
- links
- transmission duration
- transmission
- 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.)
- Active
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L45/00—Routing or path finding of packets in data switching networks
- H04L45/14—Routing performance; Theoretical aspects
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L45/00—Routing or path finding of packets in data switching networks
- H04L45/02—Topology update or discovery
- H04L45/08—Learning-based routing, e.g. using neural networks or artificial intelligence
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The application discloses a route optimization method, a route optimization device and computing equipment, and relates to the technical field of communication. The method specifically comprises the steps that after the computing equipment obtains the transmission duration corresponding to N kinds of candidate link expenses of M links, the shortest transmission duration in the transmission durations corresponding to the N kinds of candidate link expenses is obtained based on a heuristic algorithm. And then, the computing device takes the candidate link cost corresponding to the shortest transmission time as the link cost of the M links. The link cost of at least one link included in any two candidate link costs in the N candidate link costs is different, and both N and M are integers greater than or equal to 2.
Description
Technical Field
The present application relates to the field of communications technologies, and in particular, to a method and an apparatus for route optimization, and a computing device.
Background
Currently, in a network architecture of an operator, an Internet Protocol (IP) bearer network carries a large amount of services (e.g., dedicated line services, content distribution network services, etc.) of the operator. With the continuous upgrading of network architecture and service requirements, communication lines between networks are more and more, the flow of communication transmission is larger and more, the network architecture is more and more complex, more and more services need to be carried by the IP bearer network, and higher requirements and challenges are provided for the IP bearer network. In order to ensure the efficiency of the network, the route optimization aiming at the IP layer is important. At the IP layer, the routing parameter for forwarding a datagram is mainly link cost (metric), which is used to indicate the best path for a router to forward the datagram to the destination. The router selects the path with the minimum path cost according to the design of the link cost, and the path cost comprises a plurality of link costs. Therefore, the design of the link overhead needs to be matched with the existing network requirements, and the existing network resources can be fully utilized on the premise of ensuring the service requirements.
At present, link cost is usually designed by operation and maintenance personnel according to experience, and under the condition that a router transmits datagrams according to the link cost designed manually, the utilization rate of network resources is low, and the transmission time of the datagrams is long. Therefore, how to design the link overhead, improve the utilization rate of network resources, and reduce the transmission duration of datagrams is an urgent problem to be solved.
Disclosure of Invention
According to the route optimization method, the route optimization device and the computing equipment, the utilization rate of network resources can be improved and the transmission time of the datagram can be reduced due to the designed link overhead.
In order to achieve the purpose, the following technical scheme is adopted in the application:
in a first aspect, the present application provides a method of route optimization, the method being performed by a computing device. The method comprises the following steps: the computing device obtains transmission durations corresponding to N kinds of candidate link overheads of the M links, further obtains the shortest transmission duration in the transmission durations corresponding to the N kinds of candidate link overheads based on a heuristic algorithm, and takes the candidate link overheads corresponding to the shortest transmission duration as the link overheads of the M links. The candidate link cost of at least one link included in any two candidate link costs in the N candidate link costs is different, and both N and M are integers greater than or equal to 2.
In this way, the computing device selects the candidate link cost corresponding to the shortest transmission duration from the N candidate link costs pre-configured for the M links based on the heuristic algorithm, and takes the candidate link cost as the link cost of the M links. The link cost is not required to be manually designed by operation and maintenance personnel according to experience, the intelligence of the link cost design process is realized, the designed link cost is ensured to be the link cost with the shortest average transmission time of the whole network, the utilization rate of network resources is effectively improved, and the transmission time of the datagram is reduced.
In a second aspect, the present application provides a route optimization device, including: a communication unit and a processing unit.
The communication unit is configured to obtain transmission durations corresponding to N types of candidate link overheads of M links, where the candidate link overheads of at least one link included in any two types of candidate link overheads of the N types of candidate link overheads are different, and N and M are integers greater than or equal to 2. The processing unit is configured to obtain the shortest transmission duration among the transmission durations corresponding to the N candidate link overheads based on a heuristic algorithm. The processing unit is further configured to use the candidate link cost corresponding to the shortest transmission duration as the link cost of the M links.
In one possible design, the N types of candidate link overheads include an initial link overhead and an updated link overhead, and the transmission duration corresponding to the N types of candidate link overheads includes an initial transmission duration and an updated transmission duration.
In one possible design, the heuristic includes a particle swarm algorithm.
In one possible design, the processing unit is specifically configured to compare the first transmission duration with the second transmission duration; if the first transmission time length is less than the second transmission time length, determining the first transmission time length as the transmission time length of the first link group and the transmission time length of the link group set; comparing the third transmission duration with the fourth transmission duration; if the third transmission duration is less than the fourth transmission duration, determining the third transmission duration as the transmission duration of the second link group; comparing the third transmission duration with the first transmission duration; and if the third transmission time length is less than the first transmission time length, determining the third transmission time length as the transmission time length of the link group set. The N candidate link overheads comprise a first candidate link overhead, a second candidate link overhead, a third candidate link overhead and a fourth candidate link overhead, wherein the first candidate link overhead is associated with a first transmission duration of the first link group, the second candidate link overhead is associated with a second transmission duration of the first link group, the third candidate link overhead is associated with a third transmission duration of the second link group, and the fourth candidate link overhead is associated with a fourth transmission duration of the second link group; the link group set includes a first link group including M links and a second link group including M links.
In one possible design, the communication unit is further configured to, before obtaining transmission durations corresponding to multiple candidate link overheads of the M links, obtain link information of the M links, where the link information of each link includes an initial link overhead, and at least one of a longitude and a latitude of the transmitting end, a route type of the transmitting end, a longitude and a latitude of the receiving end, a route type of the receiving end, a link transmission duration, and a link bandwidth; and the processing unit is also used for determining a link cost value range according to the link information of the M links, wherein the various candidate link costs of the M links belong to the link cost value range.
In one possible design, the processing unit is specifically configured to classify the M links based on a clustering algorithm and link information of the M links to obtain S-class links, where each class of link in the S-class links includes at least one link, and S is an integer greater than or equal to 2; and determining a link cost value range of each link according to the maximum link cost and the minimum link cost of each link, wherein the candidate link cost of any link in the M links belongs to one link cost value range in the S link cost value range.
In a third aspect, the present application provides a computing device comprising a memory and a processor. The memory is coupled to the processor. The memory is for storing computer program code comprising computer instructions. When the processor executes the computer instructions, the computing device performs the route optimization method as described in the first aspect and any one of its possible designs.
In a fourth aspect, the present application provides a computer-readable storage medium having stored therein instructions that, when executed on a computer, cause the apparatus to perform the method for route optimization according to the first aspect and any one of its possible design approaches.
In a fifth aspect, the present application provides a computer program product comprising computer instructions which, when run on a computer, cause the computer to perform the method for route optimization according to the first aspect and any one of its possible design approaches.
For a detailed description of the second to fifth aspects and their various implementations in this application, reference may be made to the detailed description of the first aspect and its various implementations; moreover, the beneficial effects of the second aspect to the fifth aspect and the various implementation manners thereof may refer to the beneficial effect analysis of the first aspect and the various implementation manners thereof, and are not described herein again.
These and other aspects of the present application will be more readily apparent from the following description.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed to be 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 only some embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
Fig. 1 is a schematic structural diagram of an IP bearer network according to an embodiment of the present application;
fig. 2 is a schematic flowchart of a route optimization method according to an embodiment of the present application;
fig. 3 is a schematic flowchart illustrating a particle swarm algorithm applied to an embodiment of the present application;
fig. 4 is a schematic structural diagram of a computing device according to an embodiment of the present application;
fig. 5 is a schematic structural diagram of a route optimization device according to an embodiment of the present application.
Detailed Description
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. 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 application.
In the embodiments of the present application, for convenience of clearly describing the technical solutions of the embodiments of the present application, terms such as "first" and "second" are used to distinguish the same items or similar items with substantially the same functions and actions. Those skilled in the art will appreciate that the terms "first," "second," etc. do not denote any order or quantity, nor do the terms "first," "second," etc. denote any order or importance. The technical features described in the first and second descriptions have no sequence or magnitude order.
In the embodiments of the present application, the words "exemplary" or "such as" are used herein to mean serving as an example, instance, or illustration. Any embodiment or design described herein as "exemplary" or "such as" is not necessarily to be construed as preferred or advantageous over other embodiments or designs. Rather, use of the word "exemplary" or "such as" is intended to present relevant concepts in a concrete fashion for ease of understanding.
In the description of the present application, a "/" indicates a relationship in which the objects associated before and after are an "or", for example, a/B may indicate a or B; in the present application, "and/or" is only an association relationship describing an associated object, and means that there may be three relationships, for example, a and/or B, and may mean: a exists alone, A and B exist simultaneously, and B exists alone, wherein A and B can be singular or plural. Also, in the description of the present application, "a plurality" means two or more than two unless otherwise specified. "at least one of the following" or similar expressions refer to any combination of these items, including any combination of the singular or plural items. For example, at least one (one) of a, b, or c, may represent: a, b, c, a-b, a-c, b-c, or a-b-c, wherein a, b, c may be single or multiple.
In the embodiments of the present application, at least one may also be described as one or more, and a plurality may be two, three, four or more, which is not limited in the present application.
The topology of an Internet Protocol (IP) bearer network may be a three-layer network topology or a two-layer network topology, where the three-layer network topology may be divided into a core layer, an aggregation layer and an access layer, and the two-layer network topology may be divided into a core layer (also called a backbone spine layer) and an access layer (also called a leaf layer).
The core layer is a high-speed exchange backbone layer of the IP bearer network and is used for connecting the IP bearer network with equipment (such as external operator equipment) outside the IP bearer network. The core layer may include a router. The core layer has at least one of the following characteristics: reliability, efficiency, redundancy, fault tolerance, manageability, adaptability, low latency, etc. The routing connection of the core layer plays a very critical role in the IP bearer network, and the reliability of the IP bearer network can be generally realized through the redundant connection of a plurality of devices.
The convergence layer is an "intermediary" between the access layer and the core layer to converge data sent by a workstation (e.g., a terminal device or a server) before entering the core layer, thereby reducing the load on the core layer. The convergence layer may include routers.
The access layer is coupled to the workstations for providing access to the workstations for the local network segment. The access stratum may include routers.
Fig. 1 is a schematic structural diagram of an IP bearer network applied in an embodiment of the present application, where the IP bearer network is composed of an access device of an access layer, an aggregation device of an aggregation layer, and a core device of a core layer. The access equipment is connected with a workstation, and the workstation comprises a server and terminal equipment. As shown in fig. 1, the IP bearer network 100 includes a core layer 110, a convergence layer 120, and an access layer 130. Core layer 110 includes router 111 and router 112. Convergence layer 120 includes router 121, router 122, router 123, and router 124. Access stratum 130 includes router 131, router 132, router 133, and router 134.
Both the router 111 and the router 112 are connected to the Internet (Internet). Router 121, router 122, router 123, and router 124 are all connected to router 111. Router 121, router 122, router 123, and router 124 are all connected to router 112. Router 131 is connected to router 121 and router 122, respectively. Router 132 is connected to router 121 and router 122, respectively. Router 133 is connected to router 123 and router 124, respectively. Router 134 is connected to router 123 and router 124, respectively. The workstations 141 and 142 are connected to the router 131, respectively. The workstations 143 and 144 are respectively connected to the router 132. The workstations 145 and 146 are respectively connected to the router 133. Workstation 147 and workstation 148 are each connected to router 134. The workstation may be a server or a terminal device, without limitation.
Fig. 1 is a schematic diagram, and the IP bearer network may further include other devices, such as a device management server. These servers are used to manage all routers to ensure that they are working properly and are not shown in fig. 1. The embodiments of the present application do not limit the number of routers and workstations included in the IP bearer network. In addition, the connection mode between the router and the workstation shown in fig. 1 is only a schematic illustration, and in practical applications, the router and the workstation may also be connected in other modes, which is not limited in this application.
Each device in the IP bearer network can be used as a sending end to send data, and can also be used as a receiving end to receive data. The physical path for a sending end device to send a datagram to a receiving end device includes multiple links. A link is a physical line from one device to another without any other devices in between. Generally, devices in an IP bearer network route datagrams according to the link overhead of the link. The link cost is designed by operation and maintenance personnel according to experience, and the link cost designed manually cannot ensure the accuracy of the routing optimization effect, so that the problem of network resource waste is caused.
In order to solve the above problem, an embodiment of the present application provides a route optimization method, where a computing device obtains a link cost design scheme with a shortest average transmission duration in the entire network according to obtained link information of M links and by combining a heuristic algorithm, so as to ensure accuracy of an IP layer route optimization effect, and compared with a link cost scheme designed manually, the route optimization method can intelligently obtain the link cost design scheme with the shortest average transmission duration in the entire network, reduce overall service transmission duration, and effectively improve speed of an entire network service.
The route optimization method provided by the present application is described in detail below with reference to the accompanying drawings.
As shown in fig. 2, a method for route optimization according to an embodiment of the present application includes the following steps.
S201, the computing equipment acquires link information of the M links.
The link information of the link includes initial link overhead, and at least one of longitude and latitude of the transmitting end, route type of the transmitting end, longitude and latitude of the receiving end, route type of the receiving end, link transmission time length and link bandwidth.
The initial link cost of the link may be manually configured by the operation and maintenance personnel.
Optionally, the M pieces of link information may be network data of an area covered by one base station, or network data of areas covered by multiple base stations. The area may be a residential district, or may be a city or a country, and the range of acquiring the link information in the embodiment of the present application is not limited.
The routing types include an Access Router (AR), a convergence router (BR), a Core Router (CR), and the like. In this embodiment, the sender device may be any one of an access router AR, a convergence router BR, and a core router CR. The receiving end device may also be any one of the access router AR, the aggregation router BR, and the core router CR. Referring to fig. 1, the sending end device may be any one of router 111 and router 112 in core layer 110, may be any one of router 121, router 122, router 123, and router 124 in convergence layer 120, and may be any one of router 131, router 132, router 133, and router 134 in access layer 130. The receiving end device may be any of the router 111 and the router 112 in the core layer 110, the router 121, the router 122, the router 123, and the router 124 in the convergence layer, or any of the router 131, the router 132, the router 133, and the router 134 in the access layer 130.
The transmission duration of a link refers to the time required for a datagram to travel from the sender to the receiver of the link.
The link bandwidth refers to the amount of data that this link can transmit per unit time. The unit time is, for example, 1 second. Link bandwidth may refer to the amount of data transmitted per second.
S202, the computing equipment determines a link cost value range according to the link information of the M links.
In one possible example, the computing device classifies the M links based on a clustering algorithm and link information of the M links to obtain S-class links. Each type of S links comprises at least one link, and S is an integer greater than or equal to 2. And determining the link cost value range of each type of link according to the maximum link cost and the minimum link cost in each type of link.
For example, suppose that M links are classified into class 2, the link overhead of the 1 st link in the class 1 links is the largest link overhead of the links in the class 1 links, and the link overhead of the 1 st link is taken as the largest link overhead of the class 1 links. The link overhead of the 3 rd link in the type 1 link is the smallest link overhead of the links in the type 1 link. And taking the link overhead of the 3 rd link as the minimum link overhead of the 1 st type link, and determining the link overhead value range of the 1 st type link according to the maximum link overhead and the minimum link overhead of the 1 st type link.
For another example, the link overhead of the 2 nd link in the class 2 link is the largest link overhead of the link in the class 2 link, and the link overhead of the 2 nd link is taken as the largest link overhead of the class 2 link. And determining the link overhead value range of the 2 nd type link according to the maximum link overhead and the minimum link overhead of the 2 nd type link by taking the link overhead of the 5 th link as the minimum link overhead of the 2 nd type link, wherein the link overhead of the 5 th link in the 2 nd type link is the minimum link overhead of the link in the 2 nd type link.
And ensuring that the candidate link cost of any one link in the M links belongs to a link cost value range in the S-type link cost value range. The link cost values of multiple candidate links of M links belong to a link cost value range.
Specifically, after obtaining the link information of the M links, the computing device first classifies the M links according to a clustering algorithm.
Optionally, the clustering algorithm includes any one of a k-means clustering algorithm, a mean shift clustering algorithm, a hierarchical clustering algorithm, and the like, and the k-means clustering algorithm is taken as an example in the embodiment of the present application for explanation.
The k-means clustering algorithm is an indirect clustering method based on similarity measurement among samples, and belongs to an unsupervised learning method. The algorithm takes k as a parameter and divides n objects into k clusters, so that the clusters have higher similarity and the similarity between the clusters is lower. The similarity is calculated based on the average of the objects in a cluster (seen as the center of the cluster). Assuming that the sample data is to be classified into k classes, the algorithm flow is as follows:
(1) the initial center values of the k classes are suitably selected from the sample data, and are generally selected randomly.
(2) In each iteration, the Euclidean distances from any sample data to k centers are respectively calculated, and the sample data is classified into the class with the center value with the shortest distance.
(3) And updating the central values of the k classes by using a mean value method.
(4) And (3) repeating the steps (2) and (3) for the central values of the k classes, and finishing the iteration when the moving distance of the central values of the k classes meets a certain condition, thereby finishing the classification.
In the k-means algorithm, k is preset. In the embodiment of the present application, k may be 10, that is, M links need to be divided into 10 classes. The initial central values of 10 classes are randomly selected, in the iteration process, the Euclidean distance from the link initial link cost to 10 initial central values in the link information of the M links is calculated according to the longitude and latitude of the sending end, the routing type of the sending end, the longitude and latitude of the receiving end, the routing type of the receiving end, the link transmission time length, the link bandwidth and other data in the link information of the M links, and the initial link cost of the link information of the M links is classified into the class where the center with the shortest distance is located. And updating the central values of the 10 classes by using an average method until the initial link overhead in each class has higher similarity, the similarity between the classes is lower, and the iteration is finished.
Specifically, in the embodiment of the present application, first, M links are divided into 10 classes according to a k-means algorithm, and each class includes initial link costs of links with close similarities. And setting a link overhead value range for each type. Assuming that the minimum link overhead in the ith link is min and the maximum link overhead is max, the link overhead value range of the ith link is [ min _ i, max _ i ]. The minimum link overhead min and the link overhead value range min _ i satisfy the following formula (1).
min _ i ═ min x 0.7 formula (1)
The maximum link overhead max and the link overhead max _ i satisfy the following formula (2).
max _ i ═ max 1.3 equation (2)
Illustratively, if the initial link cost set of the first type of link is (100,200,300,400), the link cost value range of the first type of link is [70,520], the updated link cost value is a positive integer and can be evenly divided by 100, that is, the link cost value in the candidate link cost is a positive integer and can be evenly divided by 100.
Repeating the steps to define the link cost value range of the 10 types of links.
The computing equipment divides the M links into 10 classes according to the similarity based on a k-means algorithm, and each class is set with a link cost value range, so that the candidate link cost of any one of the M links is ensured to be in the link cost value range, and the update range of the candidate link cost of the M links is favorably reduced.
S203, the computing equipment obtains transmission duration corresponding to N kinds of candidate link spending of the M links.
Generally, one link may be configured with one link overhead. The N candidate link overheads represent N differently configured candidate link overheads for the M links. Each candidate link cost comprises a candidate link cost for each of the M links. The candidate link cost of at least one link included in any two candidate link costs in the N types of candidate link costs is different, and both N and M are integers greater than or equal to 2.
For example, assuming that N is 3 and M is 3, each of the three candidate link overheads includes link overheads of 3 links. The first candidate link overhead includes 3 links having different link overheads from at least one of the 3 links included in the second candidate link overhead. For example, the first type of candidate link cost includes a link cost of the 1 st link that is different from the link cost of the 1 st link included in the second type of candidate link cost. For another example, the first candidate link cost includes a link cost of the 1 st link that is different from the link cost of the 1 st link included in the second candidate link cost; and the link cost of the 2 nd link contained in the first type of candidate link cost is different from the link cost of the 2 nd link contained in the second type of candidate link cost. For another example, the second candidate link cost includes a link cost of the 2 nd link that is different from a link cost of the 2 nd link included in the third candidate link cost; and the 3 rd link cost contained in the first candidate link cost is different from the 3 rd link cost contained in the third candidate link cost.
In some embodiments, the N candidate link overheads may include an initial link overhead and an updated link overhead. M links are configured with M initial link costs, at least one of the M initial link costs is updated, and the updated link costs can be obtained.
For example, assuming that M is 3, the first candidate link cost includes an initial link cost of 3 links.
And if the initial link cost of part of the links in the 3 links is updated, obtaining a second type of candidate link cost, wherein the second type of candidate link cost comprises the initial link cost and the updated link cost. For example, the initial link cost of the 1 st link and the initial link cost of the 2 nd link in the 3 rd links are updated, and the initial link cost of the 3 rd link is not updated, so the second candidate link cost includes the updated link cost of the 1 st link, the updated link cost of the 2 nd link, and the initial link cost of the 3 rd link.
And if the initial link overheads of the 3 links are updated, obtaining a third candidate link overhead, wherein the third candidate link overhead comprises the updated link overhead. For example, the initial link cost of 3 links is updated, and the third candidate link cost includes the updated link cost of the 1 st link, the updated link cost of the 2 nd link, and the updated link cost of the 3 rd link.
Because the link can be configured with different link costs, the sending end device can select a route to transmit the datagram by using different link costs, and the transmission time of the datagram is different. The transmission duration corresponding to the N candidate link overheads includes an initial transmission duration and an updated transmission duration.
When all links in a candidate link overhead are configured as the initial link overhead, the transmission duration corresponding to the candidate link overhead may be the initial transmission duration. If the initial link cost of at least one link in the candidate link costs is updated, the transmission duration corresponding to the candidate link costs may be an updated transmission duration. Understandably, the transmission duration corresponding to the candidate link overhead may refer to a duration used by the sending end device to send the datagram to the receiving end device based on the candidate link overhead. In this embodiment, the transmission duration may be an average transmission duration of the datagram transmission in the link overhead scheme. It is understood that the network-wide average transmission time may refer to an average value of transmission time after at least one datagram is transmitted through the M links.
S204, the computing device obtains the shortest transmission duration from the transmission durations corresponding to the N kinds of candidate link overheads based on a heuristic algorithm.
Optionally, the heuristic algorithm includes any one of a particle swarm algorithm, a simulated annealing algorithm, a genetic algorithm, an artificial neural network algorithm, and the like. In the embodiments of the present application, a particle swarm algorithm is taken as an example for explanation.
The Particle Swarm Optimization (PSO) algorithm is a global random search algorithm based on Swarm intelligence and provided by simulating migration and clustering behaviors in the foraging process of a bird Swarm, is the same as other evolutionary algorithms and is based on the concepts of population and evolution, and the search of the optimal solution of a complex space is realized through cooperation and competition among individuals. Particle swarm optimization simulates birds in a bird flock by designing a particle without mass, which has two attributes: the position represents the direction of the particle movement and the speed represents the speed of the particle movement.
The PSO algorithm is initialized to a population of random particles (random solution) and then iteratively finds the optimal solution. In each iteration, the particle updates itself by tracking two extrema; the first extreme is the optimal solution found by the particle itself, this solution being called the individual extreme (pbest) i ) (ii) a The other extreme is the best solution currently found for the entire population, which is the global extreme (gbest) i )。
All the particles in the particle swarm adjust their positions and speeds according to the found current individual extremum and the global extremum in the whole particle swarm, and the positions are substituted into the objective function to obtain the fitness value (FitNum) of the particles i ) And determining the position according to the size of the fitness value.
Suppose that in a D-dimensional target search space, N particles form a cluster, where the ith particle is represented as a D-dimensional vector, and is denoted as:
x i ＝(x 1 ,x 2 ,x 3 ,...,x N ),i＝1,2,3,...,N。
the velocity of the ith particle is also a vector in D dimension, and is noted as:
v i ＝(v 1 ,v 2 ,v 3 ,...,v N ),i＝1,2,3,...,N。
the optimal position searched by the ith particle so far is called an individual extremum and is recorded as:
pbest i ＝(p 1 ,p 2 ,p 3 ,...,p N ),i＝1,2,3,...,N。
the optimal position searched by the whole particle swarm so far is called as a global extremum, and is recorded as:
gbest i ＝(g 1 ,g 2 ,g 3 ,...,g N ),i＝1,2,3,...,N。
when this optimum value is found, the particle updates its velocity and position by the following equations (3) and (4).
v i ＝ω (t) ×v i-1 +c 1 ×rand( )×(pbest i-1 -x i-1 )+c 2 ×rand( )×(gbest i-1 -x i-1 ) Formula (3)
x i ＝x i-1 +v i Formula (4)
Wherein, c 1 And c 2 For learning factors, also called acceleration constants, c in general 1 ＝c 2 2. rand () is between [0,1 ]]Uniform random number, x, within a range i Is the position of this iteration of the particle, i.e. the current position, x, of the particle i-1 Is the position of the last iteration of the particle. v. of i Is the velocity of the current iteration of the particle, i.e. the current velocity, v, of the particle i-1 Is the speed of the last iteration of the particle. pbest i-1 Is the individual extremum of the last iteration of the particle, gbest i-1 Is the global extremum of the last iteration of the particle swarm.
ω (t) The inertia factor has a large influence on the convergence of the particle swarm algorithm. Omega (t) Can be taken from [0,1]The random number of the interval may be a fixed value.
The flow of the particle swarm optimization is as follows:
1) initializing a population of particles comprising a population size N, a position x of each particle i And velocity v i ；
2) Calculating the fitness value FitNum of each particle i ；
3) For each particle, the fitness value FitNum of each particle is used i And individual extremum pbest per particle i By comparison, if FitNum i Less than pbest i Using FitNum i Replacing pbest i Otherwise pbest i This is done for each particle without updating, updating the pbest of each particle i 。
4) For each particle, using the particle fitness value FitNum i And global extreme of this particle, gbest i By comparison, if FitNum i Less than gbest i Using FitNum i Replacement of gbest i Otherwise gbest i And not updated. For each granuleDoing this, the global gbest is updated i 。
5) Updating the position x of the particle according to the formulas (3) and (4) i And velocity v i ；
6) Exit if the end condition is met (fitness value is small enough or maximum number of cycles is reached), otherwise return to 2).
The iteration termination condition is generally selected to reach the maximum iteration times and the optimal position searched by the particle swarm so far meets the preset minimum adaptation threshold according to specific problems. Referring to fig. 3, a schematic flow chart of a particle swarm algorithm applied to an embodiment of the present application is shown.
Specifically, a population of particles is initialized, including the number of particles, the dimensions, the range of values for each dimension, and the position and velocity of each particle.
In the embodiment of the present application, each particle represents an optimized link overhead design scheme, which includes M links, i.e., a link group. The particle group represents a collection of several kinds of link groups, i.e. a collection of link groups.
The number of particles may be represented by N, which may be set to 40, representing 40 different link overhead designs, i.e., 40 link groups. The number of iterations may be represented by t, which may be set to 500. The dimension number is in one-to-one correspondence with the number of links, and the number of links and the dimension number are equal. The link cost value range corresponding to each link is the value range of each dimension.
Illustratively, a first link of the M links may be a first dimension and a second link may be a second dimension. Through the classification of the k-means algorithm, a first link in the M links is classified into a first class, a second link is classified into a third class, the value range of the first dimension adopts the link cost value range of the first class, and the value range of the second dimension adopts the link cost value range of the third class.
In this embodiment of the present application, the position of the particle represents a link cost design scheme of the link group at the g-th iteration of the link group set, and the speed of the particle represents a variation of the link cost of at least one link in the link group updating process. Setting the speed coefficient c 1 ＝c 2 0.5, the speed coefficient is mainlyFor controlling the refresh rate of the particles. In the embodiment of the present application, the inertia factor ω satisfies the following formula (5).
ω (t) ＝(ω ini -ω end )(G k -g)/G k +ω end Formula (5)
G k Is set to 500 for the maximum number of iterations, g is the current number of iterations, ω ini The initial inertia weight is set to 0.9. Omega end The inertia weight when iterating to the maximum evolution algebra is set to 0.4.
In some embodiments, the computing device may calculate the network-wide average transmission time length of each link group, and use the calculated network-wide average transmission time length as the fitness value of the link group.
In other embodiments, the average transmission time length of the whole network under each link group scheme may be calculated by other devices, and then the calculation result is transmitted to the calculation device by the other devices.
It can be understood that the link group set is not updated after initialization, and the current average transmission duration of each link group in the whole network is the link group extremum of each link group. And comparing the average transmission time lengths of the whole networks of the plurality of link groups in the link group set, determining the shortest average transmission time length of the whole networks in the average transmission time lengths of the whole networks of the plurality of link groups, and taking the shortest average transmission time length of the whole networks as the extreme value of the link group set. It should be understood that the set of extrema for the link group is a global extremum and the extrema for the link group is an individual extremum.
Further, each link group is updated, and the link cost of each link in the link group is updated in the value range corresponding to each dimension.
Optionally, if the computing power of the computing device or other devices is sufficient, the 40 link groups may be updated simultaneously, and the computing device or other devices may calculate the 40 link groups in real time to obtain the updated average transmission duration of each link group in the whole network.
Optionally, if the computational power of the computing device or other devices is insufficient, the 40 link groups may be updated sequentially, and the computing device or other devices sequentially calculate the updated link groups, and sequentially obtain the updated average transmission duration of each link group in the whole network.
And when the updated average transmission time length of the whole network of all the link groups is calculated, comparing the updated average transmission time length of each link group with the extreme value of the link group of each link group, and if the updated average transmission time length of the whole network is less than the extreme value of the link group, taking the updated average transmission time length of the whole network as the extreme value of the link group. And if the updated average transmission time length of the whole network is greater than or equal to the extreme value of the link group, the extreme value of the link group is not changed.
Further, the updated average transmission time length of the whole network of each link group is compared with the set extreme value of the link group, and if the updated average transmission time length of any link group is less than the set extreme value of the link group, the updated average transmission time length of the whole network of the link group is used as the set extreme value of the link group. If the updated average transmission time of any link group in the whole network is still greater than or equal to the set extreme value of the link group, the set extreme value of the link group is not changed. The extreme value of the link group set refers to the shortest average transmission time length of the whole network in the link group set during the g iteration in the iteration process of the link group set.
For example, if the set of link groups includes a first link group, the computing device calculates an initial transmission duration of the first link group as a first transmission duration, and an initial link cost of the first link group as a first candidate link cost. The updated link cost after the first candidate link cost is updated is the second candidate link cost, and the updated transmission time length after the first transmission time length is updated is the second transmission time length. If the first transmission duration is less than the second transmission duration, it represents that the first candidate link overhead is better than the second candidate link overhead. And taking the first transmission duration as the transmission duration of the first link group and the link group set transmission duration, that is, the first transmission duration is the first link group extreme value and the link group set extreme value.
If the link group set includes the first link group and the second link group, the computing device calculates the initial transmission duration of the second link group as a third transmission duration, and the initial link cost of the second link group is a third candidate link cost. The updated link cost after the update of the third candidate link cost is the fourth candidate link cost, and the updated transmission time length after the update of the third transmission time length is the fourth transmission time length. If the third transmission duration is less than the fourth transmission duration, it represents that the third candidate link overhead is better than the fourth candidate link overhead. The third transmission duration is taken as the transmission duration of the second link group, that is, the third transmission duration is the extreme value of the second link group.
And comparing the first transmission duration with the third transmission duration, wherein if the third transmission duration is less than the first transmission duration, the third candidate link overhead is superior to the first candidate link overhead, and the third transmission duration corresponding to the third candidate link overhead is the shortest transmission duration of the link group set, namely the extreme value of the link group set.
S205, the computing device takes the candidate link cost corresponding to the shortest transmission duration as the link cost of the M links.
It can be understood that the transmission duration of the candidate link overhead corresponding to the link group set extremum is the shortest transmission duration. In this embodiment of the present application, the particle swarm algorithm terminating condition may be that after the update times of the 40 link groups reach the maximum iteration time of 500 times, the candidate link cost of the link group corresponding to the shortest transmission duration in the link group set at the 500 th iteration is used as the link cost of the M links.
Optionally, in this embodiment of the application, the particle swarm algorithm may be terminated under the condition that, in the updating process of the 40 types of link groups, the link group set extremum determined in a certain updating is unchanged after continuously updating for 10 times. And taking the candidate link cost of the link group corresponding to the extreme value of the link group set as the link cost of the M links.
In the embodiment of the application, the computing device classifies the link information of the M links through a clustering algorithm, and sets the value ranges of the link expenses according to the classification result, so that the link expenses of the M links have corresponding value ranges and are updated in the value ranges, and the convergence rate of the particle swarm algorithm is increased. Determining multiple link overhead design schemes through a particle swarm algorithm, calculating the average transmission time length of the whole network under each scheme, and continuously updating the link overhead of the links in each design scheme until the link overhead design scheme with the shortest average transmission time length of the whole network is found. Compared with the prior art that operation and maintenance personnel manually design link spending according to related experience, the method and the device realize intelligent acquisition of the link spending design scheme with the shortest average transmission time of the whole network, improve the accuracy of the link spending design scheme, reduce resource waste caused by the fact that the link spending scheme designed by the operation and maintenance personnel due to judgment errors is not the scheme with the shortest average transmission time of the whole network, reduce the transmission time of the whole service and effectively improve the speed of the whole network service.
The scheme provided by the embodiment of the application is mainly introduced from the perspective of a method. To implement the above functions, it includes hardware structures and/or software modules for performing the respective functions. Those of skill in the art would readily appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as hardware or combinations of hardware and computer software. Whether a function is performed as hardware or computer software drives hardware depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.
As shown in fig. 4, an embodiment of the present application provides a computing device 400. The computing device may include at least one processor 401, communication lines 402, memory 403, and communication interfaces 404.
In particular, processor 401 is configured to execute computer-executable instructions stored in memory 403 to implement steps or actions of a computing device. In this embodiment, the processor 401 may be configured to determine a plurality of link overhead schemes and calculate a transmission duration for each link overhead scheme.
The processor 401 may be a chip. For example, the Field Programmable Gate Array (FPGA) may be a Field Programmable Gate Array (FPGA), an Application Specific Integrated Circuit (ASIC), a system on chip (SoC), a Central Processing Unit (CPU), a Network Processor (NP), a digital signal processing circuit (DSP), a Micro Controller Unit (MCU), a Programmable Logic Device (PLD) or other integrated chips.
A communication line 402 for transmitting information between the processor 401 and the memory 403. In this embodiment, the communication line 402 may be configured to transmit the multiple link overhead design schemes determined by the processor 401 and the transmission durations corresponding to the multiple link overhead design schemes to the memory 403.
A memory 403 for storing and executing computer-executable instructions, and controlled by the processor 401. In the embodiment of the present application, the memory 403 may be used to store various link overhead designs and corresponding transmission durations.
The memory 403, which may be separate, is connected to the processor via the communication line 402. The memory 403 may be either volatile memory or nonvolatile memory, or may include both volatile and nonvolatile memory. The non-volatile memory may be a read-only memory (ROM), a Programmable ROM (PROM), an Erasable PROM (EPROM), an Electrically Erasable PROM (EEPROM), or a flash memory. Volatile memory can be Random Access Memory (RAM), which acts as external cache memory. By way of example, and not limitation, many forms of RAM are available, such as Static Random Access Memory (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (enhanced SD RAM). It should be noted that the memory of the systems and devices described herein is intended to comprise, without being limited to, these and any other suitable types of memory.
A communication interface 404 for communicating with other devices or a communication network. The communication network may be an ethernet, a Radio Access Network (RAN), or a Wireless Local Area Network (WLAN). In an embodiment of the present application, the communication interface 404 may be used for the computing device 400 to communicate with other devices.
It should be noted that the configuration shown in FIG. 4 does not constitute a limitation of the computing device, which may include more or less components than those shown, or some components in combination, or a different arrangement of components than those shown in FIG. 4.
As shown in fig. 5, an embodiment of the present application provides a route optimization device 50. The apparatus may comprise a communication unit 51 and a processing unit 52.
The communication unit 51 is configured to obtain transmission durations corresponding to N types of candidate link overheads of the M links, where the candidate link overheads of at least one link included in any two types of candidate link overheads of the N types of candidate link overheads are different, and N and M are integers greater than or equal to 2.
The processing unit 52 is configured to obtain the shortest transmission duration among the transmission durations corresponding to the N candidate link overheads based on a heuristic algorithm. And the candidate link overhead corresponding to the shortest transmission duration is used as the link overhead of the M links.
It should be understood that, in the various embodiments of the present application, the sequence numbers of the above-mentioned processes do not mean the execution sequence, and the execution sequence of each process should be determined by its function and inherent logic, and should not constitute any limitation to the implementation process of the embodiments of the present application.
In actual implementation, the communication unit 51 and the processing unit 52 may be implemented by the processor 401 shown in fig. 4 calling the program code in the memory 403, and are not described here again.
Another embodiment of the present application further provides a computer-readable storage medium, which stores computer instructions, and when the computer instructions are executed on a computer, the computer is caused to execute the steps in the method flow shown in the above method embodiment.
In another embodiment of the present application, there is also provided a computer program product comprising instructions which, when executed on a computer, cause the computer to perform the steps of the method flow illustrated in the above method embodiment.
Those of ordinary skill in the art will appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the technical solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.
It can be clearly understood by those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described systems, devices and units may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
In the several embodiments provided in the present application, it should be understood that the disclosed system, apparatus, and method may be implemented in other ways. For example, the above-described embodiments of the apparatus are merely illustrative, and for example, a division of a unit is merely a logical division, and an actual implementation may have another division, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form.
Units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, functional units in the embodiments of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit.
The above description is only an embodiment of the present application, but the scope of the present application is not limited thereto, and any changes or substitutions within the technical scope of the present disclosure should be covered by the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.
Claims (9)
1. A method of route optimization, the method being performed by a computing device, the method comprising:
acquiring transmission time lengths corresponding to N kinds of candidate link overheads of M links, wherein the candidate link overheads of at least one link included in any two kinds of candidate link overheads of the N kinds of candidate link overheads are different, and both N and M are integers greater than or equal to 2;
acquiring the shortest transmission duration in the transmission durations corresponding to the N kinds of candidate link overheads based on a heuristic algorithm;
and taking the candidate link cost corresponding to the shortest transmission time length as the link cost of the M links.
2. The method of claim 1, wherein the N candidate link overheads include an initial link overhead and an updated link overhead, and wherein the transmission durations corresponding to the N candidate link overheads include an initial transmission duration and an updated transmission duration.
3. The method of claim 2, wherein the heuristic algorithm comprises a particle swarm algorithm.
4. The method of claim 3, wherein the N candidate link overheads include a first candidate link overhead associated with a first transmission duration of a first link group, a second candidate link overhead associated with a second transmission duration of the first link group, a third candidate link overhead associated with a third transmission duration of a second link group, and a fourth candidate link overhead associated with a fourth transmission duration of the second link group; the link group set includes the first link group and the second link group, the first link group includes the M links, and the second link group includes the M links;
the obtaining the shortest transmission duration of the transmission durations corresponding to the N candidate link overheads based on a heuristic algorithm includes:
comparing the first transmission duration with the second transmission duration;
if the first transmission duration is less than the second transmission duration, determining the first transmission duration as the transmission duration of the first link group and the transmission duration of the link group set;
Comparing the third transmission duration to the fourth transmission duration;
if the third transmission duration is less than the fourth transmission duration, determining the third transmission duration as the transmission duration of the second link group;
comparing the third transmission duration to the first transmission duration;
and if the third transmission time length is smaller than the first transmission time length, determining the third transmission time length as the transmission time length of the link group set.
5. The method according to any one of claims 1 to 4, wherein before said obtaining the transmission durations corresponding to the multiple candidate link overheads for the M links, the method further comprises:
acquiring link information of the M links, wherein the link information of each link comprises initial link overhead and at least one of longitude and latitude of a sending end, route type of the sending end, longitude and latitude of a receiving end, route type of the receiving end, link transmission time and link bandwidth;
and determining a link cost value range according to the link information of the M links, wherein the link cost values of multiple candidate links of the M links belong to the link cost value range.
6. The method of claim 5, wherein the determining a link cost range according to the link information of the M links comprises:
Classifying the M links based on a clustering algorithm and link information of the M links to obtain S links, wherein each link in the S links comprises at least one link, and S is an integer greater than or equal to 2;
and determining a link cost value range of each link according to the maximum link cost and the minimum link cost of each link, wherein the candidate link cost of any link in the M links belongs to one link cost value range in the S-type link cost value range.
7. An apparatus for route optimization, the apparatus comprising:
the communication unit is used for acquiring transmission duration corresponding to N kinds of candidate link overheads of M links, wherein the candidate link overheads of at least one link included in any two kinds of candidate link overheads of the N kinds of candidate link overheads are different, and both N and M are integers greater than or equal to 2;
the processing unit is used for acquiring the shortest transmission duration from the transmission durations corresponding to the N kinds of candidate link overheads based on a heuristic algorithm;
the processing unit is further configured to use the candidate link cost corresponding to the shortest transmission duration as the link cost of the M links.
8. A computing device comprising one or more processors and one or more memories;
The one or more memories coupled with the one or more processors for storing computer program code comprising computer instructions which, when executed by the one or more processors, cause the computing device to perform the route optimization method of any of claims 1-6.
9. A computer storage medium comprising computer instructions that, when executed on a computing device, cause the computing device to perform the route optimization method of any one of claims 1-6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110831817.3A CN113507413B (en) | 2021-07-22 | 2021-07-22 | Route optimization method and device and computing equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110831817.3A CN113507413B (en) | 2021-07-22 | 2021-07-22 | Route optimization method and device and computing equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113507413A CN113507413A (en) | 2021-10-15 |
| CN113507413B true CN113507413B (en) | 2022-07-29 |
Family
ID=78013528
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110831817.3A Active CN113507413B (en) | 2021-07-22 | 2021-07-22 | Route optimization method and device and computing equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113507413B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116156592B (en) * | 2022-12-27 | 2023-08-22 | 深圳市宇通联发科技有限公司 | Low-delay wireless transmission method, device, communication management equipment and storage medium |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2007103837A1 (en) * | 2006-03-09 | 2007-09-13 | Firetide, Inc. | Effective bandwidth path metric and path computation method for wireless mesh networks with wired links |
| CN102036130A (en) * | 2009-09-24 | 2011-04-27 | 中国电信股份有限公司 | Quantification method for searching optimal path for circuit in ASON (automatic switched optical network) network |
| CN105763467A (en) * | 2016-03-25 | 2016-07-13 | 杭州华三通信技术有限公司 | Flow switching method and device |
| CN108200623A (en) * | 2017-12-29 | 2018-06-22 | 华南理工大学 | A kind of centralized path computation and power-economizing method based on genetic algorithm |
| CN111884927A (en) * | 2020-07-16 | 2020-11-03 | 中盈优创资讯科技有限公司 | Link overhead obtaining method and device based on ospf link database |
| CN112242950A (en) * | 2019-07-18 | 2021-01-19 | 华为技术有限公司 | Method for determining path and related equipment |
| CN112923940A (en) * | 2021-01-11 | 2021-06-08 | 珠海格力电器股份有限公司 | Path planning method, device, processing equipment, mobile equipment and storage medium |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9967166B2 (en) * | 2015-06-30 | 2018-05-08 | Cisco Technology, Inc. | Soft constrained shortest-path first tunneling |
| US11438243B2 (en) * | 2019-04-12 | 2022-09-06 | EMC IP Holding Company LLC | Adaptive adjustment of links per channel based on network metrics |
-
2021
- 2021-07-22 CN CN202110831817.3A patent/CN113507413B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2007103837A1 (en) * | 2006-03-09 | 2007-09-13 | Firetide, Inc. | Effective bandwidth path metric and path computation method for wireless mesh networks with wired links |
| CN102036130A (en) * | 2009-09-24 | 2011-04-27 | 中国电信股份有限公司 | Quantification method for searching optimal path for circuit in ASON (automatic switched optical network) network |
| CN105763467A (en) * | 2016-03-25 | 2016-07-13 | 杭州华三通信技术有限公司 | Flow switching method and device |
| CN108200623A (en) * | 2017-12-29 | 2018-06-22 | 华南理工大学 | A kind of centralized path computation and power-economizing method based on genetic algorithm |
| CN112242950A (en) * | 2019-07-18 | 2021-01-19 | 华为技术有限公司 | Method for determining path and related equipment |
| CN111884927A (en) * | 2020-07-16 | 2020-11-03 | 中盈优创资讯科技有限公司 | Link overhead obtaining method and device based on ospf link database |
| CN112923940A (en) * | 2021-01-11 | 2021-06-08 | 珠海格力电器股份有限公司 | Path planning method, device, processing equipment, mobile equipment and storage medium |
Non-Patent Citations (2)
| Title |
|---|
| Harkirat Kaur.Analysis of Metrics:Improved Hybrid ACO-PSO Based Routing Algorithm for Mobile Ad-hoc Network .《PDGC》.2017, * |
| 移动IP中基于遗传算法的优化路由算法;杨建军;《浙江大学学报(工学版)》;20041130;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113507413A (en) | 2021-10-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Chen et al. | When UAV swarm meets edge-cloud computing: The QoS perspective | |
| CN109583665B (en) | Unmanned aerial vehicle charging task scheduling method in wireless sensor network | |
| CN104115165B (en) | Machine learning is used to carry out the method for mapped media assembly | |
| CN110247793B (en) | Application program deployment method in mobile edge cloud | |
| CN110968426B (en) | Edge cloud collaborative k-means clustering model optimization method based on online learning | |
| Jiang et al. | Fedmp: Federated learning through adaptive model pruning in heterogeneous edge computing | |
| Al-Janabi et al. | Efficient whale optimisation algorithm-based SDN clustering for IoT focused on node density | |
| CN111988225A (en) | Multi-path routing method based on reinforcement learning and transfer learning | |
| CN111953547B (en) | Heterogeneous base station overlapping grouping and resource allocation method and device based on service | |
| WO2021008675A1 (en) | Dynamic network configuration | |
| CN113507413B (en) | Route optimization method and device and computing equipment | |
| CN111885493B (en) | Micro-cloud deployment method based on improved cuckoo search algorithm | |
| CN113328953B (en) | Method, device and storage medium for network congestion adjustment | |
| CN113485409A (en) | Unmanned aerial vehicle path planning and distribution method and system for geographic fairness | |
| CN109495565A (en) | High concurrent service request processing method and equipment based on distributed ubiquitous computation | |
| CN113676357B (en) | Decision method for edge data processing in power internet of things and application thereof | |
| CN112770256B (en) | Node track prediction method in unmanned aerial vehicle self-organizing network | |
| CN116720132A (en) | Power service identification system, method, device, medium and product | |
| CN115545106A (en) | AoI sensitive data collection method and system in multiple unmanned aerial vehicles | |
| CN114997422A (en) | Grouping type federal learning method of heterogeneous communication network | |
| CN110933728B (en) | Mapping method, device and storage medium of virtualized wireless sensor network | |
| CN114785692A (en) | Virtual power plant aggregation regulation and control communication network flow balancing method and device | |
| CN114629769A (en) | Traffic map generation method of self-organizing network | |
| CN114595052A (en) | Distributed communication load balancing method based on graph partitioning algorithm | |
| Jeong et al. | Optimal Partitioning of Distributed Neural Networks for Various Communication Environments |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |