CN114615705A - Single user resource allocation strategy method based on 5G network - Google Patents
Single user resource allocation strategy method based on 5G network Download PDFInfo
- Publication number
- CN114615705A CN114615705A CN202210237568.XA CN202210237568A CN114615705A CN 114615705 A CN114615705 A CN 114615705A CN 202210237568 A CN202210237568 A CN 202210237568A CN 114615705 A CN114615705 A CN 114615705A
- Authority
- CN
- China
- Prior art keywords
- local processing
- server
- processing cost
- time
- total
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 37
- 238000013468 resource allocation Methods 0.000 title claims abstract description 21
- 238000012545 processing Methods 0.000 claims abstract description 126
- 238000005457 optimization Methods 0.000 claims abstract description 23
- 230000002068 genetic effect Effects 0.000 claims abstract description 7
- 238000004364 calculation method Methods 0.000 claims description 16
- 230000008569 process Effects 0.000 claims description 16
- 230000005540 biological transmission Effects 0.000 claims description 11
- 238000012216 screening Methods 0.000 claims description 6
- 210000000349 chromosome Anatomy 0.000 claims description 5
- 238000004891 communication Methods 0.000 abstract description 6
- 230000006870 function Effects 0.000 description 9
- 238000010295 mobile communication Methods 0.000 description 4
- 238000001228 spectrum Methods 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 230000001419 dependent effect Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 230000035772 mutation Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000002485 combustion reaction Methods 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000008303 genetic mechanism Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000012804 iterative process Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000013439 planning Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000002269 spontaneous effect Effects 0.000 description 1
- 230000004083 survival effect Effects 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W28/00—Network traffic management; Network resource management
- H04W28/02—Traffic management, e.g. flow control or congestion control
- H04W28/0215—Traffic management, e.g. flow control or congestion control based on user or device properties, e.g. MTC-capable devices
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/12—Computing arrangements based on biological models using genetic models
- G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W28/00—Network traffic management; Network resource management
- H04W28/02—Traffic management, e.g. flow control or congestion control
- H04W28/0226—Traffic management, e.g. flow control or congestion control based on location or mobility
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W28/00—Network traffic management; Network resource management
- H04W28/02—Traffic management, e.g. flow control or congestion control
- H04W28/0231—Traffic management, e.g. flow control or congestion control based on communication conditions
- H04W28/0236—Traffic management, e.g. flow control or congestion control based on communication conditions radio quality, e.g. interference, losses or delay
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W28/00—Network traffic management; Network resource management
- H04W28/16—Central resource management; Negotiation of resources or communication parameters, e.g. negotiating bandwidth or QoS [Quality of Service]
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biophysics (AREA)
- Theoretical Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Genetics & Genomics (AREA)
- General Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Physiology (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Quality & Reliability (AREA)
- Mobile Radio Communication Systems (AREA)
Abstract
The invention discloses a single-user resource allocation strategy method based on a 5G network, which comprises the steps of obtaining the total calculated amount of subtasks, obtaining a first local processing cost based on the total calculated amount, and judging the first local processing cost; distributing the total calculated amount to the local processing and server based on the judgment result, and acquiring a second local processing cost and a server unloading processing cost based on the calculated amount distributed to the local processing and server; constructing a total utility function based on the first local processing cost, the second local processing cost and the server offload processing cost, wherein the utility function is a difference between the first local processing cost and a sum of the second local processing cost and the server offload processing; and constructing an optimization model based on the total utility function, and solving the optimization model through a genetic algorithm to obtain a resource allocation strategy. The invention can consider multiple factors, and further perform joint optimization on task unloading decision, communication resource allocation and computing resource allocation.
Description
Technical Field
The invention relates to the technical field of machine learning, in particular to a strategy method for allocating single user resources based on a 5G network.
Background
With the development of 5G, the number of mobile terminals has increased explosively, and a great number of mobile devices have made higher demands on various performances of wireless networks, core networks, computing and storage devices of mobile communication. Therefore, on the basis of meeting the service requirement of 5G diversity, the method is adopted to implement configuration and resource joint optimization of 5G mobile communication network resources so as to achieve the longest standby time of the mobile terminal, that is, to achieve minimization of power consumption of the mobile device, which is a problem to be solved at present.
The 5G mobile communication network sinks the calculation task to the edge end of the mobile network, and partial calculation task of the terminal is unloaded to the edge server by utilizing the strong calculation capacity of the edge calculation, so that the electric energy consumed by the terminal and the service delay are reduced. Therefore, the user service under a single user can divide the service into a plurality of dependent subtasks, and whether part of the subtasks are uploaded to the edge server is determined according to the calculated amount, the queue condition and the equipment calculation capacity of the subtasks, so that the purpose of reducing the power consumption of the mobile terminal is achieved, and an energy-saving and efficient resource joint optimization strategy is provided for the user.
For a mobile application composed of multiple subtasks, it is necessary to determine which subtasks are allocated to an edge server for execution, and which subtasks are placed on a mobile terminal for execution, and this offloading process is binary offloading. At present, most of researches are to realize optimization of network and computing resources under the constraint conditions of CPU computing frequency, computing time delay, task queues and the like by formalizing the problems under the assumption that the execution time of an edge server and the time delay of uploading a terminal to the server are ignored. Since the wireless environment is a dynamically changing environment, with the increase of the number of users, the transmission time delay and transmission time of mobile communication have great fluctuation, so that the transmission time and transmission time delay of data from a terminal to an edge server are key factors in a massive terminal environment. Before the subtask executes the unloading strategy, in addition to the conventional factors, the transmission delay and the transmission time of the wireless side must be considered.
Disclosure of Invention
In order to solve the problem of the deficiency of the current task offloading strategy in the prior art, the invention provides a single-user resource allocation strategy method based on a 5G network, which can consider various factors and perform joint optimization on task offloading decisions, communication resource allocation and computing resource allocation.
In order to achieve the technical purpose, the invention provides the following technical scheme: comprises that
Acquiring total calculated amount of the subtasks, acquiring first local processing cost based on the total calculated amount, and judging the first local processing cost, wherein the first local processing cost is the processing cost of the total calculated amount of local processing;
distributing the total calculated amount to a local processing and server based on the judgment result, and acquiring a second local processing cost and a server unloading processing cost based on the calculated amount distributed to the local processing and server, wherein the second local processing cost is the processing cost of the calculated amount distributed to the local processing;
constructing a total utility function based on the first local processing cost, the second local processing cost, and the server offload processing cost, wherein the utility function is a difference between the first local processing cost and a sum of the second local processing cost and the server offload processing;
and constructing an optimization model based on the total utility function, and solving the optimization model through a genetic algorithm to obtain a resource allocation strategy.
Preferably, the process of obtaining the total computation amount of the subtasks includes:
wherein L isiRepresenting the total calculation of the subtasks, etaiRepresenting the ratio of the output result and the input data of the task i; setiRepresenting the in-degree set of subtasks i, liRepresenting the computation data of the subtask i itself.
Preferably, the obtaining of the first local processing cost includes:
calculating a first local processing time t based on the total calculation amounti0And the first local processing time is the time for locally processing the total calculated amount:
obtaining a first local processing cost based on a local processing time
Wherein f isiRepresenting the amount of processing computation per unit time of local processing, eiIndicating the time at which the subtask is expected to complete,
representing the unit processing cost of the subtask.
Preferably, the obtaining of the second local processing cost includes:
calculating second local processing time t 'based on distribution proportion in distribution result'i0And the second local processing time is the processing time of the total local processing calculated amount:
calculating and acquiring second local processing cost based on second local processing time
Wherein λ isiRepresents the ratio of the calculated amount distributed to the server in the subtask to the total calculated amount, and is 0 ≦ λi≤1。
Preferably, the obtaining process of the server offload processing cost includes:
calculating the time for unloading each process by the server based on the distribution proportion in the distribution result, wherein the time for unloading each process by the server comprises connection establishment time, data uploading time, waiting time and unloading task processing time;
summing calculation is carried out on each flow time of unloading processing of the computing server, and unloading processing time of the server is obtained;
calculating server offload processing cost based on server offload processing time
Wherein, tijIn order for the server to offload the processing time,
representing the unit processing cost during data transfer.
Preferably, the total utility function is:
wherein,
denotes the total utility, xijRepresenting a server node SjAvailability for subtask i; wherein if the subtask is assigned to the server node and the server node is available for the subtask, xij1, otherwise xij＝0。
Preferably, the optimization model is:
s.t.
0≤λi≤1
max(t'i0,tij)≤ti0
tij≤ei
t'i0≤ei。
preferably, the solving process of the optimization model includes:
obtaining server nodes, carrying out chromosome coding on the server nodes randomly selected by the subtasks to generate an initial population, calculating individual fitness in the initial population through an optimization model, screening individuals with highest fitness as first individuals, and carrying out selection operation on the initial population based on the fitness to obtain a selected population;
carrying out cross variation on the selected population, obtaining individual fitness in the population after the cross variation through calculation, screening the individual with the highest fitness as a second individual, comparing the first individual with the second individual, and replacing the individual with the lowest fitness in the population after the cross variation with the second individual if the first individual is smaller than the second individual fitness to obtain an updated population;
and repeatedly carrying out cross variation, screening, comparison and replacement on the updated population until the cycle number is greater than the iteration number, outputting a statistical result, and obtaining a resource allocation strategy based on the statistical result.
Preferably, the selection operation employs a roulette selection algorithm.
Preferably, the crossing adopts self-adaptive double-point crossing.
The invention has the following technical effects:
the invention can provide a single-user resource allocation strategy method under the constraint conditions of transmission delay, edge server computation frequency, computation delay and the like under the wireless environment, which is a single-user resource allocation strategy for determining which subtasks are allocated to the edge server to be executed and which subtasks are put at a mobile terminal to be executed on the premise that a user meets the user service requirement (specific delay).
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings needed to be used in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings without inventive exercise.
FIG. 1 is a schematic diagram of a system according to an embodiment of the present invention;
FIG. 2 is a diagram of a collaborative edge computing system according to an embodiment of the present invention;
fig. 3 is a flowchart of offloading of a cooperative edge network according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, 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 invention.
In order to solve the problems of the prior art, such as the insufficiency of the current task offloading strategy, as shown in fig. 1, the present invention provides the following scheme:
acquiring total calculated amount of the subtasks, acquiring first local processing cost based on the total calculated amount, and judging the first local processing cost, wherein the first local processing cost is the processing cost of the total calculated amount of local processing;
distributing the total calculated amount to a local processing and server based on the judgment result, and acquiring a second local processing cost and a server unloading processing cost based on the calculated amount distributed to the local processing and server, wherein the second local processing cost is the processing cost of the calculated amount distributed to the local processing;
constructing a total utility function based on the first local processing cost, the second local processing cost, and the server offload processing cost, wherein the utility function is a difference between the first local processing cost and a sum of the second local processing cost and the server offload processing;
and constructing an optimization model based on the total utility function, and solving the optimization model through a genetic algorithm to obtain a resource allocation strategy.
The specific technical method for realizing the steps is as follows:
(1) constructing a network architecture and a task unloading utility model of cooperative edge computing:
the purpose of network architecture setting of cooperative edge computing is to solve the problem of network congestion caused by service delay and massive user service requirements.
The time delay of service processing of a user can be greatly improved by the aid of the edge servers deployed close to the terminal, in a wireless environment, the edge servers often exist, and each user can select a proper edge server as a target unloading server. Since the performance of the servers deployed to the mobile side is inconsistent, if each user selects the edge server with the best performance for unloading, the system is likely to be in a functional failure, and therefore, the system adopts a centralized processing principle to determine where the user tasks are to be processed and whether the user tasks are necessary to be processed according to the resource requirements of the computing tasks, the computing capacity of the terminal, the CPU computing frequency/computing delay/task queue of the edge server, the user mobility and other factors.
Secondly, users in the same region share a section of spectrum resources when unloading the edge server, and the limited spectrum resources can increase the data transmission time when facing the transmission requirement of massive user data volume.
In addition, the user has a mobile feature, and the link between the user and the mobile server is disconnected frequently.
Therefore, when the task is designed to be unloaded, the resource requirement of the computing task, the computing capacity/the terminal queue of the terminal, the CPU computing frequency/the computing delay/the task queue of the edge server, the network spectrum resource, the user mobility and other factors need to be considered, and the task unloading decision, the communication resource allocation and the computing resource allocation are jointly optimized. A system for collaborative edge computation is shown in fig. 2.
As shown in fig. 2, a plurality of edge computing servers are deployed in a certain area, and each edge computing server can perform data interaction with a plurality of users by means of the 5G technology, but each server can only perform one computing task due to the limited computing capability of the edge server. For convenience of representation, each user terminal or each edge server is referred to as a node (a server is referred to as a service node, and a handset terminal is referred to as a terminal node). Each node is provided with an antenna, so that data can be conveniently sent and received. The distributed service nodes provide ubiquitous low-latency computing offload services for users. Due to the difference of communication environments, the coverage area of each edge server is not consistent, and the computing power, bandwidth and storage of the server nodes are different due to different configurations. The invention adopts SDN technology to realize centralized management and cooperative decision of service nodes.
In order to implement centralized management and cooperative decision of service nodes, an offloading flow of a cooperative edge network is performed around links of acquiring task state information, implementing task scheduling decision and scheduling scheme by decision planning, executing instructions, and the like, as shown in fig. 3.
Task unloading utility model:
judging standard sent by the unloading request: assuming that the mobile phone terminal has a computation-intensive application including N subtasks, the present invention is represented by (V, E), where a node set V represents application subtasks and an edge set E represents dependencies between subtasks. Node v for subtask iiDenotes, SetiRepresenting the dependency Set of a subtask i, whose dependency Set is Set before task i is executediMust be completed before it can actually be performed. Binary (L)i,ηi) A dependent parameter, L, representing a subtask iiRepresenting subtasks viInput data amount (available Set)iOutput result to measure) and the amount of computation (computation data l of subtask i itselfi) Sum, ηiRepresenting the ratio of the output result and the input data of task i. If the subtask i depends on the SetiNot null, then task sub-viThe total calculated amount is:
firstly, each subtask is calculated preferentially at the mobile phone terminal side, and the calculation task is executed locally:
if t isi0≥eiWherein e isiRepresenting subtasks viExpect to completeThen means that the local computation incurs a certain delay cost to the user application, which can be expressed as:
wherein
Representing a task viCost per unit delay of if
If the number is more than 0, the mobile phone terminal sends an unloading request to the edge server.
(2) And (5) cooperating with the edge calculation joint optimization model.
Suppose a subtask viThe processing of the entire amount of data is not placed on the service node, but rather a portion of the data is offloaded to the service node. Suppose a subtask viThe ratio of the unloaded data to the total calculated amount is λi,0≤λiLess than or equal to 1. Then task viIs divided into two parts lambdaiLiNot less than 0 and (1-lambda)i)LiIs more than or equal to 0 and respectively represents a subtask viThe amount of data offloaded to the edge server and the amount of data processed at this level.
<1> time of local processing:
the time to offload to the edge server includes:
<2>connection setup time
Mobile phone and edge service node SjThe time required to establish a reliable connection depends on the communication environment, the handset and the node SjRelative position of the mobile phone and velocity of the mobile phone. This connection setup time is usuallyAnd the statistical average value is adopted for measurement (suitable for scenes of highways and high-speed rails).
<3>Time of data upload, task viUploading the offloaded portion of the data to the service node SjIn (1),
wherein r isi uIndicating the data uploading rate and the transmission power P of the mobile phonei tChannel gain h of mobile phone terminal and service node2 service but white Gaussian noise N0The repetition frequency B of the data pulse signal.
<4> latency, associated with the queue of the server, now calculate the queue length of the base station, assuming that q (t) represents the queue length of the serving node at the edge of the timeslot t, which can be expressed as:
wherein F (t-1) represents the amount of computation data that the service node can process at the t-1 slot edge.
Representing the amount of data unloaded by all handsets at time t-1. Then, waiting time
The method comprises the following steps:
<5>offloading task execution time
<6> data result return time, since the result data is small and the downlink data rate is large, the result return time may be ignored.
Then, the time from the whole unloading process to the completion of task processing is:
then the processing cost of offloading the task to the server is:
then the processing cost of the handset local processing is:
based on the above analysis, utility function U before and after unloadingijCan be expressed as the difference between before and after unloading:
in the invention, the subtasks processed between the server nodes need to be returned to the mobile phone side, so that the subtasks received by the server nodes do not need to consider the sequence relation, namely the mobile phone side sequences the subtask unloading sequence before the mobile phone side sends the unloading request. And assuming that the computing performance of each service node is limited and only one subtask can be executed at a time, the task offloading decision and the computing resource allocation optimization of the present invention can be understood as maximizing the total task offloading utility as an objective function.
If subtask viIs distributed to a service node SjFor task viIs availability (a)ij1), then xij1, otherwise xij＝0。aijRepresenting a service node SjAvailability to task i. Then the total utility of the task offload is:
the first item of the formula is that all tasks are placed in local processing, and the second item of the formula is that the unloading processing cost is partially placed in the edge server, namely, the comparison between the local processing scheme and the partial unloading scheme realizes the maximization of the total unloading utility, and the advantage embodied by the unloaded edge service node can be embodied. For computational convenience, we minimize the server offload processing cost, and we can achieve the goal of equation 14:
(3) executing the task unloading decision algorithm to obtain a task unloading decision and scheduling scheme
A genetic algorithm is employed herein to find the optimal value of equation 15. The genetic algorithm is a search optimization algorithm based on natural selection rules and natural genetic mechanisms, and is a basic evolution process which simulates the suitability for survival by using the principle of the superiority and inferiority of a spontaneous combustion selection process. The genetic algorithm comprises a selection operator, a crossover operator and a mutation operator, wherein the crossover operator is dominant. The algorithm is a typical iterative algorithm, which starts from a group of randomly generated solutions, and realizes operations of simulating biological evolution and inheritance in each iterative process, so as to generate a group of new solutions, wherein the new solutions have fitness to give evaluation and are repeated continuously until the algorithm reaches certain required convergence. The method comprises the following specific steps:
step 1: and carrying out chromosome coding on the service node randomly selected by each task, and randomly generating an initial population, wherein the size of the population is P, L is a chromosome length (coding length), and the population is marked as Pop.
Step 2: and converting the binary chromosomes into decimal numbers, and calculating the fitness value of the individuals in the current population.
Step 3: finding the best individual and the best fitness value in the current generation (seeking the minimum value of equation 15), and performing selection operation from the current generation population, this time adopting a roulette selection algorithm, and recording the selected population as S _ pop.
Step 4: and (4) crossing the population S _ pop, wherein the crossing adopts self-adaptive double-point crossing, and then carrying out mutation.
Step 5: and (3) the optimal fitness value after the variation and the corresponding optimal individual are compared with the optimal individual in Step3, if the optimal fitness value after the variation is larger than the optimal fitness value in Step3, no change is made, and if the optimal fitness value after the variation is smaller than the optimal fitness value in Step3, the individual corresponding to the worst fitness value after the variation is replaced by the optimal individual selected from Step3, namely the optimal storage strategy.
Step 6: if the loop times are larger than the maximum iteration times, outputting a statistical result, such as a convergence curve and the like, terminating the whole algorithm, otherwise, returning to step 2.
According to the method, network resources and computing resources are quickly found out to meet requirements after the task unloading decision is executed, and joint optimization of the task unloading decision, communication resource allocation and computing resource allocation is achieved. And then sending the task unloading decision to the mobile phone, sending the task scheduling scheme to the edge service node, sending the task data to the edge service node after the mobile phone receives the task unloading decision, executing the task by the edge service node, and returning the result to the mobile phone side.
The foregoing illustrates and describes the principles, general features, and advantages of the present invention. It will be understood by those skilled in the art that the present invention is not limited to the embodiments described above, which are described in the specification and illustrated only to illustrate the principle of the present invention, but that various changes and modifications may be made therein without departing from the spirit and scope of the present invention, which fall within the scope of the invention as claimed. The scope of the invention is defined by the appended claims and equivalents thereof.
Claims (10)
1. A strategy method for allocating single user resources based on a 5G network is characterized by comprising the following steps:
acquiring total calculated amount of the subtasks, acquiring first local processing cost based on the total calculated amount, and judging the first local processing cost, wherein the first local processing cost is the processing cost of the total calculated amount of local processing;
distributing the total calculated amount to a local processing and server based on the judgment result, and acquiring a second local processing cost and a server unloading processing cost based on the calculated amount distributed to the local processing and server, wherein the second local processing cost is the processing cost of the calculated amount distributed to the local processing;
constructing a total utility function based on the first local processing cost, the second local processing cost, and the server offload processing cost, wherein the utility function is a difference between the first local processing cost and a sum of the second local processing cost and the server offload processing;
and constructing an optimization model based on the total utility function, and solving the optimization model through a genetic algorithm to obtain a resource allocation strategy.
2. The policy method for allocating single-user resources based on 5G network according to claim 1, wherein:
the total calculation amount obtaining formula of the subtasks comprises the following steps:
wherein L isiRepresenting the total calculation of the subtasks, etaiRepresenting the ratio of the output result and the input data of the subtask i;Setirepresenting the in-degree set of subtasks i, liRepresenting the computation data of the subtask i itself.
3. The policy method for allocating single-user resources based on 5G network according to claim 1, wherein:
the obtaining process of the first local processing cost includes:
calculating a first local processing time t based on the total calculation amounti0And the first local processing time is the time for locally processing the total calculated amount:
obtaining a first local processing cost based on a local processing time
Wherein f isiRepresenting the amount of processing computation per unit time of local processing, eiIndicating the time at which the subtask is expected to complete,
representing the unit processing cost of the subtask.
4. The policy method for allocating single-user resources based on 5G network according to claim 1, wherein:
the obtaining process of the second local processing cost includes:
calculating second local processing time t 'based on distribution proportion in distribution result'i0And the second local processing time is the processing time of the total local processing calculated amount:
calculating and acquiring second local processing cost based on second local processing time
Wherein λ isiRepresents the ratio of the calculated amount distributed to the server in the subtask to the total calculated amount, and is 0 ≦ λi≤1。
5. The policy method for allocating single-user resources based on 5G network according to claim 1, wherein:
the acquisition process of the server unloading processing cost comprises the following steps:
calculating the time for unloading each process by the server based on the distribution proportion in the distribution result, wherein the time for unloading each process by the server comprises connection establishment time, data uploading time, waiting time and unloading task processing time;
summing calculation is carried out on each flow time of unloading processing of the computing server, and unloading processing time of the server is obtained;
calculating server offload processing cost based on server offload processing time
Wherein, tijIn order for the server to offload the processing time,
the table represents the unit processing cost during data transmission.
6. The policy method for allocating single-user resources based on 5G network according to claim 1, wherein:
the total utility function is:
wherein,
denotes the total utility, xijRepresenting a server node SjAvailability for subtask i; wherein if the subtask is assigned to the server node and the server node is available for the subtask, xij1, otherwise xij＝0。
7. The policy method for allocating single-user resources based on 5G network according to claim 1, wherein:
the optimization model is as follows:
s.t.
0≤λi≤1
max(t′i0,tij)≤ti0
tij≤ei
t′i0≤ei。
8. the policy method for allocating single-user resources based on 5G network according to claim 1, wherein:
the solving process of the optimization model comprises the following steps:
obtaining server nodes, carrying out chromosome coding on the server nodes randomly selected by the subtasks to generate an initial population, calculating individual fitness in the initial population through an optimization model, screening individuals with highest fitness as first individuals, and carrying out selection operation on the initial population based on the fitness to obtain a selected population;
carrying out cross variation on the selected population, obtaining individual fitness in the population after the cross variation through calculation, screening the individual with the highest fitness as a second individual, comparing the first individual with the second individual, and replacing the individual with the lowest fitness in the population after the cross variation with the second individual if the first individual is smaller than the second individual fitness to obtain an updated population;
and repeatedly carrying out cross variation, screening, comparison and replacement on the updated population until the cycle number is greater than the iteration number, outputting a statistical result, and obtaining a resource allocation strategy based on the statistical result.
9. The policy method for allocating resources of single user based on 5G network according to claim 8, wherein:
the selection operation employs a roulette selection algorithm.
10. The policy method for allocating resources of single user based on 5G network according to claim 8, wherein:
the crossing adopts self-adaptive double-point crossing.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210237568.XA CN114615705B (en) | 2022-03-11 | 2022-03-11 | Single-user resource allocation strategy method based on 5G network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210237568.XA CN114615705B (en) | 2022-03-11 | 2022-03-11 | Single-user resource allocation strategy method based on 5G network |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114615705A true CN114615705A (en) | 2022-06-10 |
| CN114615705B CN114615705B (en) | 2022-12-20 |
Family
ID=81862786
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210237568.XA Active CN114615705B (en) | 2022-03-11 | 2022-03-11 | Single-user resource allocation strategy method based on 5G network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114615705B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2023245835A1 (en) * | 2022-06-20 | 2023-12-28 | 南京邮电大学 | Satellite transmission optimization method based on grid delay prediction |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100082838A1 (en) * | 2008-09-30 | 2010-04-01 | Microsoft Corporation | Isp-friendly rate allocation for p2p applications |
| WO2018209734A1 (en) * | 2017-05-15 | 2018-11-22 | 网宿科技股份有限公司 | Dynamic scheduling and allocation method and system for network traffic |
| CN110035410A (en) * | 2019-03-07 | 2019-07-19 | 中南大学 | Federated resource distribution and the method and system of unloading are calculated in a kind of vehicle-mounted edge network of software definition |
| CN110856215A (en) * | 2019-10-29 | 2020-02-28 | 南京邮电大学 | Resource allocation algorithm for multi-access edge calculation |
| CN110996393A (en) * | 2019-12-12 | 2020-04-10 | 大连理工大学 | Single-edge computing server and multi-user cooperative computing unloading and resource allocation method |
| CN111107566A (en) * | 2019-12-25 | 2020-05-05 | 国网冀北电力有限公司唐山供电公司 | Unloading method based on collaborative content caching in power Internet of things scene |
| CN113873022A (en) * | 2021-09-23 | 2021-12-31 | 中国科学院上海微系统与信息技术研究所 | Mobile edge network intelligent resource allocation method capable of dividing tasks |
| CN113950103A (en) * | 2021-09-10 | 2022-01-18 | 西安电子科技大学 | Multi-server complete computing unloading method and system under mobile edge environment |
| CN113950066A (en) * | 2021-09-10 | 2022-01-18 | 西安电子科技大学 | Single server part calculation unloading method, system and equipment under mobile edge environment |
| CN113992678A (en) * | 2021-10-27 | 2022-01-28 | 嘉兴学院 | Calculation migration method for offshore MEC load balancing and resource allocation joint optimization |
-
2022
- 2022-03-11 CN CN202210237568.XA patent/CN114615705B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100082838A1 (en) * | 2008-09-30 | 2010-04-01 | Microsoft Corporation | Isp-friendly rate allocation for p2p applications |
| WO2018209734A1 (en) * | 2017-05-15 | 2018-11-22 | 网宿科技股份有限公司 | Dynamic scheduling and allocation method and system for network traffic |
| CN110035410A (en) * | 2019-03-07 | 2019-07-19 | 中南大学 | Federated resource distribution and the method and system of unloading are calculated in a kind of vehicle-mounted edge network of software definition |
| CN110856215A (en) * | 2019-10-29 | 2020-02-28 | 南京邮电大学 | Resource allocation algorithm for multi-access edge calculation |
| CN110996393A (en) * | 2019-12-12 | 2020-04-10 | 大连理工大学 | Single-edge computing server and multi-user cooperative computing unloading and resource allocation method |
| CN111107566A (en) * | 2019-12-25 | 2020-05-05 | 国网冀北电力有限公司唐山供电公司 | Unloading method based on collaborative content caching in power Internet of things scene |
| CN113950103A (en) * | 2021-09-10 | 2022-01-18 | 西安电子科技大学 | Multi-server complete computing unloading method and system under mobile edge environment |
| CN113950066A (en) * | 2021-09-10 | 2022-01-18 | 西安电子科技大学 | Single server part calculation unloading method, system and equipment under mobile edge environment |
| CN113873022A (en) * | 2021-09-23 | 2021-12-31 | 中国科学院上海微系统与信息技术研究所 | Mobile edge network intelligent resource allocation method capable of dividing tasks |
| CN113992678A (en) * | 2021-10-27 | 2022-01-28 | 嘉兴学院 | Calculation migration method for offshore MEC load balancing and resource allocation joint optimization |
Non-Patent Citations (1)
| Title |
|---|
| 黄晓舸等: "基于MEC的任务卸载和资源分配联合优化方案", 《系统工程与电子技术》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2023245835A1 (en) * | 2022-06-20 | 2023-12-28 | 南京邮电大学 | Satellite transmission optimization method based on grid delay prediction |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114615705B (en) | 2022-12-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109684075B (en) | Method for unloading computing tasks based on edge computing and cloud computing cooperation | |
| Chen et al. | Energy-efficient task offloading and resource allocation via deep reinforcement learning for augmented reality in mobile edge networks | |
| CN109947545B (en) | Task unloading and migration decision method based on user mobility | |
| CN111445111B (en) | Electric power Internet of things task allocation method based on edge cooperation | |
| CN110493360B (en) | Mobile edge computing unloading method for reducing system energy consumption under multiple servers | |
| CN111372314A (en) | Task unloading method and task unloading device based on mobile edge computing scene | |
| CN108809695A (en) | A kind of distribution uplink unloading strategy towards mobile edge calculations | |
| CN111953758A (en) | Method and device for computing unloading and task migration of edge network | |
| CN111182570A (en) | User association and edge computing unloading method for improving utility of operator | |
| CN111475274A (en) | Cloud collaborative multi-task scheduling method and device | |
| CN111556572A (en) | Spectrum resource and computing resource joint allocation method based on reinforcement learning | |
| CN111988787B (en) | Task network access and service placement position selection method and system | |
| CN113918240A (en) | Task unloading method and device | |
| CN113590279B (en) | Task scheduling and resource allocation method for multi-core edge computing server | |
| CN113573363B (en) | MEC calculation unloading and resource allocation method based on deep reinforcement learning | |
| Kopras et al. | Task allocation for energy optimization in fog computing networks with latency constraints | |
| CN112235387B (en) | Multi-node cooperative computing unloading method based on energy consumption minimization | |
| Wang et al. | Task allocation mechanism of power internet of things based on cooperative edge computing | |
| CN115955479A (en) | Task rapid scheduling and resource management method in cloud edge cooperation system | |
| CN114615705B (en) | Single-user resource allocation strategy method based on 5G network | |
| Singh et al. | Profit optimization for mobile edge computing using genetic algorithm | |
| Chen et al. | Joint optimization of task caching, computation offloading and resource allocation for mobile edge computing | |
| CN113821346A (en) | Computation uninstalling and resource management method in edge computation based on deep reinforcement learning | |
| CN116828534B (en) | Intensive network large-scale terminal access and resource allocation method based on reinforcement learning | |
| CN112148482A (en) | Edge computing task scheduling method based on combination with load balancing |
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 |