CN112600706B - Task unloading method based on 5G network online learning - Google Patents
Task unloading method based on 5G network online learning Download PDFInfo
- Publication number
- CN112600706B CN112600706B CN202011470057.XA CN202011470057A CN112600706B CN 112600706 B CN112600706 B CN 112600706B CN 202011470057 A CN202011470057 A CN 202011470057A CN 112600706 B CN112600706 B CN 112600706B
- Authority
- CN
- China
- Prior art keywords
- task
- base station
- small base
- value
- tasks
- 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
- 238000000034 method Methods 0.000 title claims abstract description 19
- 230000006870 function Effects 0.000 claims abstract description 18
- 238000010276 construction Methods 0.000 claims abstract description 5
- 238000012545 processing Methods 0.000 claims description 5
- 230000001413 cellular effect Effects 0.000 abstract description 5
- 238000010586 diagram Methods 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 238000004891 communication Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 239000002360 explosive Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/14—Network analysis or design
- H04L41/145—Network analysis or design involving simulating, designing, planning or modelling of a network
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W24/00—Supervisory, monitoring or testing arrangements
- H04W24/02—Arrangements for optimising operational condition
-
- 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
- H04W28/08—Load balancing or load distribution
Abstract
The utility model discloses a task unloading method based on 5G network online learning, which comprises the following steps: s1: modeling the problem with integer linear programming→s2: introducing a Lagrangian multiplier construction auxiliary function to S3: estimated prize value for task offload completion → S4: offloading decision to complete multiple tasks→s5: and updating the state table according to the feedback result, selecting a task for unloading according to the unloading scheme in the step S4 by all the small base station nodes, feeding back the result after completing the task, and updating the relevant parameters of the auxiliary function in the step S2 according to the feedback value of the task completion for calculating the reward value of the next time slot. In order to ensure the performance of the network system, the utility model requires each small base station node in each time slot t to successfully complete at least 15 tasks, and the task is unloaded and decided in the network system under the special structure of the 5G cellular network so as to maximize the whole effective rewarding value of the network system.
Description
Technical Field
The utility model relates to the technical field of edge computing, in particular to a task unloading method based on 5G network online learning.
Background
With the explosive growth of mobile services, the requirements of users are more and more difficult to meet by the conventional macro base station, the 5G technology is generated, the 5G-age edge computing becomes a great feature of network architecture change, an edge network appears in a wireless network system for the first time, the edge network promotes revolutionary reconstruction of network construction, support and operation chains, and future internet of things services are highly dependent on the deployment capability of the edge to realize the requirements of high broadband, low delay, high-density connection and the like;
the small base station is used as an access scene with the most characteristic of 5G, is highly matched with the requirements of an edge computing platform in the aspects of high density, easy deployment, freeness and low cost, compared with the macro base station, the deployment of the small base station is closer to users and scenes, the edge computing coverage area deployed on the small base station is within the coverage area of a single site, the small base station is very suitable for service processing and forwarding within the coverage area, a core network is not needed, and the time delay of the users can be reduced to milliseconds.
The prior art has the following defects: a calculation server is deployed on each small base station node in a zone, because the communication resources and the server calculation resources of each small base station node are limited, the number of devices which can be connected by each small base station node at the same time is limited, and millimeter waves have weak penetrability, so that signals are easy to be blocked by obstacles (such as large-scale mobile trains), when tasks to be offloaded in the coverage area of each small base station node are too many, the offloading decision needs to be carried out, the tasks in each time period are online, the task situation which will appear at the next moment is not known at the current moment, the task offloading decision is required to be carried out immediately, and the problem becomes more difficult.
Disclosure of Invention
The utility model provides a task unloading method based on 5G network online learning, which is used for ensuring the performance of a network system, requiring each small base station node to successfully complete at least 15 tasks in each time slot t, and carrying out unloading decision on the tasks in the network system under the special structure of a 5G cellular network so as to maximize the overall effective rewarding value of the network system, thereby solving the problems in the background art.
In order to achieve the above purpose, the present utility model provides the following technical solutions: a method for task offloading based on 5G network online learning, the method comprising the steps of:
s1: the problem is modeled with an integer linear program, the algorithm is as follows,
subjectto：
wherein ,indicating the effective rewards value brought by the task i unloading to the small base station node m at the time t after being processed; />Indicating whether task i is offloaded to the small base station node m during time t, +.>Indicating the probability that the small base station node m can successfully complete phi-i type tasks at time t,/>The resources occupied by the processing task i of the small base station node m in the time slot t are represented, and beta represents the total amount of computing resources of each small base station node;
s2: a lagrangian multiplier construction auxiliary function is introduced,
estimating potential rewarding values of each task unloading completion according to the modeling in the step S1, marking the difference (remorse value) between the total effective rewarding value finally brought by the algorithm and the effective rewarding value of the theoretical optimal algorithm as R (T) according to the objective function of the integer linear programming (1),
wherein ,representing the offloading decision of the algorithm, km being the maximum value of the devices that can occur in the coverage area of each small base station node, +.>Then the decision result of the theoretical optimal algorithm is represented;
minimizing the satisfaction of the constraints (1 c) and (1 d) of R (T) above defines two violations V 1(T) and V2 (T) measuring whether the algorithm satisfies constraint conditions (1 c) and (1 d):
wherein ,[·]+ =max (·, 0), algorithm achieves V while satisfying the minimization of R (T) 1(T) and V2 (T) minimization, to achieve equilibrium of both, lagrangian multiplier is introduced and />Constructing an auxiliary function:
wherein Rm (T), V m.1(T) and Vm.2 (T) represents the remorse value and the violation value of the small base station node m at T, respectively;
s3: the estimated task offloads the completed prize value,
according to the function formula in step S2Estimating in the formula (5), calculating the probability of each small base station node unloading the tasks in the coverage area of the small base station node in an online learning mode, and unloading the tasksIs relaxed from the value constraint of {0,1} to interval [0,1]]Dividing all task space into a plurality of intervals according to the context information of the tasks, setting a weight value for each interval, and calculating the probability of the task i to the small base station node m by adopting an online algorithm for each small base station node m based on the weight value>
S4: the decision to offload a plurality of tasks is completed,
constructing a weighted bipartite graph between the task and the small base station node based on the probability of offloading each task to one or more small base station nodes calculated in step S3, wherein,the method comprises the steps that the weight of an edge (m, i) in a weighted bipartite graph is obtained, an edge with the largest weight is selected in the weighted bipartite graph in a greedy iteration mode, tasks of the corresponding edge are unloaded to small base station nodes corresponding to the edge, and meanwhile the edge is deleted from the bipartite graph;
s5: the state table is updated according to the feedback result,
and (3) selecting tasks to unload according to the unloading scheme in the step (S4), feeding back results after the tasks are completed, updating relevant parameters of the auxiliary function in the step (S2) according to feedback values of the completed tasks, calculating the reward value of the next time slot, and updating Lagrange multipliers according to the following formula:
wherein the auxiliary parameterγ m ∈(0，1]。
Preferably, in the step S2, a task i value of 1 indicates that the task i is to be offloaded onto the small cell node m at t, and is not offloaded at 0.
Preferably, in the step S2, (1 a) the algorithm constraint condition ensures that the device connected by each small cell node in each time slot does not exceed its maximum connection upper limit, (1 b) the algorithm constraint condition ensures that each task cannot be repeatedly unloaded, (1 c) the algorithm ensures that the number of tasks successfully completed by each small cell node in each time slot is not less than α, (1 d) the constraint condition ensures that the resources occupied by the task on each small cell node in each time slot does not exceed its resource upper limit, (1 e) the constraint condition defines a decision variableThe value range of (2) is equal to 0 or 1.
Preferably, in the step S3, the weight value is initialized to 1, context information of all tasks is acquired in each time slot, the tasks are classified into corresponding intervals according to the context information, and the weight value of the tasks is estimated by the weight value of the intervals.
Preferably, in the step S4, when the task to be offloaded on a small base station node reaches the maximum, all edges taking the small base station node as the vertex are deleted from the bipartite graph, and the decision of offloaded of a plurality of tasks to a plurality of small base station nodes is completed according to the greedy iterative manner.
Preferably, in the step S5, the feedback result is that and />Is a real value of (c).
The utility model has the technical effects and advantages that:
in order to ensure the performance of the network system, the utility model requires each small base station node in each time slot t to successfully complete at least 15 tasks, and the task is unloaded and decided in the network system under the special structure of the 5G cellular network so as to maximize the whole effective rewarding value of the network system.
Drawings
In order to more clearly illustrate the embodiments of the present utility model or the technical solutions in the prior art, the drawings that are required to be used in the embodiments will be briefly described below, and it is apparent that the drawings in the following description are only some embodiments described in the present utility model, and other drawings may be obtained according to these drawings for a person having ordinary skill in the art.
Fig. 1 is a conceptual flow diagram of an embodiment of the present utility model.
FIG. 2 is a graph showing experimental results of an embodiment of the present utility model.
FIG. 3 is a graph of the experimental results of a prior algorithm.
FIG. 4 is a second diagram of the experimental results of the prior algorithm.
FIG. 5 is a third diagram of the experimental results of the prior algorithm.
Detailed Description
The following description of the embodiments of the present utility model will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present utility model, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the utility model without making any inventive effort, are intended to be within the scope of the utility model.
Referring to fig. 1-5 of the specification, the utility model provides a task unloading method based on 5G network online learning, which comprises the following steps:
s1: the problem is modeled with an integer linear program, the algorithm is as follows,
subjectto：
wherein ,indicating the effective rewards value brought by the task i unloading to the small base station node m at the time t after being processed; />Indicating whether task i is offloaded to the small base station node m during time t, +.>Indicating the probability that the small base station node m can successfully complete phi-i type tasks at time t,/>The resources occupied by the processing task i of the small base station node m in the time slot t are represented, and beta represents the total amount of computing resources of each small base station node;
s2: a lagrangian multiplier construction auxiliary function is introduced,
estimating potential rewarding values of each task unloading completion according to the modeling in the step S1, marking the difference (remorse value) between the total effective rewarding value finally brought by the algorithm and the effective rewarding value of the theoretical optimal algorithm as R (T) according to the objective function of the integer linear programming (1),
wherein ,representing the offloading decision of the algorithm, km being the maximum value of the devices that can occur in the coverage area of each small base station node, +.>Then the decision result of the theoretical optimal algorithm is represented;
minimizing the satisfaction of the constraints (1 c) and (1 d) of R (T) above defines two violations V 1(T) and V2 (T) measuring whether the algorithm satisfies constraint conditions (1 c) and (1 d):
wherein ,[·]+ =max (·, 0), algorithm achieves V while satisfying the minimization of R (T) 1(T) and V2 (T) minimization, to achieve equilibrium of both, lagrangian multiplier is introduced and />Constructing an auxiliary function:
wherein Rm (T), V m.1(T) and Vm.2 (T) represents the remorse value and the violation value of the small base station node m at T, respectively;
s3: the estimated task offloads the completed prize value,
estimating according to the function formula (5) in the step S2, and calculating the probability of each small base station node unloading the tasks in the coverage range of the small base station node in an online learning mode, wherein the probability of unloading the tasksIs relaxed from the value constraint of {0,1} to interval [0,1]]Dividing all task space into a plurality of intervals according to the context information of the tasks, setting a weight value for each interval, and calculating the probability of the task i to the small base station node m by adopting an online algorithm for each small base station node m based on the weight value>
S4: the decision to offload a plurality of tasks is completed,
constructing a weighted bipartite graph between the task and the small base station node based on the probability of offloading each task to one or more small base station nodes calculated in step S3, wherein,the method comprises the steps that the weight of an edge (m, i) in a weighted bipartite graph is obtained, an edge with the largest weight is selected in the weighted bipartite graph in a greedy iteration mode, tasks of the corresponding edge are unloaded to small base station nodes corresponding to the edge, and meanwhile the edge is deleted from the bipartite graph;
s5: the state table is updated according to the feedback result,
and (3) selecting tasks to unload according to the unloading scheme in the step (S4), feeding back results after the tasks are completed, updating relevant parameters of the auxiliary function in the step (S2) according to feedback values of the completed tasks, calculating the reward value of the next time slot, and updating Lagrange multipliers according to the following formula:
wherein the auxiliary parameterγ m ∈(0，1]。
Example 1: multiple users may be present in the coverage area of the 5G network area per time slot, with one task to be offloaded in each user's wireless device. Considering the limited resources, it is not possible to offload all tasks to the small base station nodes, in order to maximize the overall prize value (efficiency) of the network system while satisfying the system constraints, it is then necessary to decide by means of an algorithm which tasks to offload to which small base station nodes. Firstly modeling a problem (describing the problem with linear programming); then, lagrangian multipliers are introduced to construct a reward function, and a reward value of each task is calculated by using an online learning method and used for assisting in calculating the unloading probability of each task; then, for each small base station node, calculating the probability that all tasks in the coverage range of the small base station node are unloaded; finally, because a task may be covered by a plurality of small base station nodes, in order to avoid the problem of repeated unloading, a weighted bipartite graph is constructed, and the task unloading is performed by adopting a greedy algorithm according to the weighted bipartite graph, so that the overall unloading scheme of the 5G network system is obtained.
Preferably, in the step S2, a task i value of 1 indicates that the task i is to be offloaded onto the small cell node m at t, and is not offloaded at 0.
Preferably, in the step S2, (1 a) algorithm constraintThe conditions ensure that the equipment connected by each small base station node in each time slot does not exceed the maximum connection upper limit, (1 b) the algorithm constraint condition ensures that each task cannot be repeatedly unloaded, (1 c) the algorithm ensures that the number of tasks successfully completed by each small base station node in each time slot is not less than alpha, (1 d) the constraint condition ensures that the resources occupied by the tasks on each small base station node in each time slot do not exceed the resource upper limit, and (1 e) the constraint condition limits the decision variableThe value range of (2) is equal to 0 or 1.
Preferably, in the step S3, the weight value is initialized to 1, context information of all tasks is acquired in each time slot, the tasks are classified into corresponding intervals according to the context information, and the weight value of the tasks is estimated by the weight value of the intervals.
Preferably, in the step S4, when the task to be offloaded on a small base station node reaches the maximum, all edges taking the small base station node as the vertex are deleted from the bipartite graph, and the decision of offloaded of a plurality of tasks to a plurality of small base station nodes is completed according to the greedy iterative manner.
Preferably, in the step S5, the feedback result is that and />Is a real value of (c).
Example 2: in a cellular network area, there are m small base station nodes (small base station nodes) connected to the same Macro Base Station (MBS) by optical cables, and in which small cellular network Wireless Devices (WD) are distributed, each device possibly having tasks with offloading, and a computing server is disposed on each small base station node to handle the tasks from the wireless device, taking into account T time slots, symbolizingRepresentation ofTime gap set, i.e.)>Similarly, the symbol->Representing the set of small base station nodes, the task set of the wireless device that appears at small base station m at time slot t is denoted +.>All task sets in the network area at time slot t are then denoted +.>At the beginning of each time slot, each device transmits the context information (including the input parameter size, the output parameter size and the like) of the task to the small base station node, the context information of the task i is recorded as phi, the number of devices which can be connected by each small base station node is assumed to be c, and the total computing resource of each small base station node is assumed to be beta.
Example 3: setting a 5G network area comprising 30 small base station nodes and a macro base station, assuming that the number of infinite devices (i.e. tasks to be offloaded) in the coverage area of each small base station node is 35-300, and the number of devices which can be connected in the same time slot of one small base station node is 20, the rewarding value brought by each small base station node processing one task is distributed between [0,1], the weak penetrability of millimeter waves can make the connection between the devices and the small base station nodes unstable, so that even if one task is offloaded to the small base station node, the situation that the task is not successfully processed can occur, the probability of each small base station node successfully completing each task is distributed between [0,1], and in order to ensure the performance of the network system, each small base station node is required to successfully complete at least 15 tasks in each time slot t.
The last points to be described are: first, in the description of the present utility model, it should be noted that, unless otherwise specified and defined, the terms "mounted," "connected," and "connected" are to be construed broadly, and may be mechanical or electrical, or may be a direct connection between two elements, and "upper," "lower," "left," "right," etc. are merely used to indicate relative positional relationships, which may be changed when the absolute position of the object being described is changed;
secondly: in the drawings of the disclosed embodiments, only the structures related to the embodiments of the present disclosure are referred to, and other structures can refer to the common design, so that the same embodiment and different embodiments of the present disclosure can be combined with each other under the condition of no conflict;
finally: the foregoing description of the preferred embodiments of the utility model is not intended to limit the utility model to the precise form disclosed, and any such modifications, equivalents, and alternatives falling within the spirit and principles of the utility model are intended to be included within the scope of the utility model.
Claims (6)
1. A task unloading method based on 5G network online learning is characterized in that: the method comprises the following steps:
s1: the problem is modeled with an integer linear program, the algorithm is as follows,
subjectto：
wherein ,indicating the effective rewards value brought by the task i unloading to the small base station node m at the time t after being processed;indicating whether task i is offloaded to the small base station node m during time t, +.>Indicating the probability that the small base station node m can successfully complete phi-i type tasks at time t,/>The resources occupied by the processing task i of the small base station node m in the time slot t are represented, and beta represents the total amount of computing resources of each small base station node;
s2: a lagrangian multiplier construction auxiliary function is introduced,
estimating potential rewarding values of each task unloading completion according to the modeling in the step S1, marking the difference (remorse value) between the total effective rewarding value finally brought by the algorithm and the effective rewarding value of the theoretical optimal algorithm as R (T) according to the objective function of the integer linear programming (1),
wherein ,representing the offloading decision of the algorithm, km being the maximum value of the devices that can occur in the coverage area of each small base station node, +.>Then the decision result of the theoretical optimal algorithm is represented;
minimizing the satisfaction of the constraints (1 c) and (1 d) of R (T) above defines two violations V 1(T) and V2 (T) measuring whether the algorithm satisfies constraint conditions (1 c) and (1 d):
wherein ,[·]+ =max (·, 0), algorithm achieves V while satisfying the minimization of R (T) 1(T) and V2 (T) minimization, to achieve equilibrium of both, lagrangian multiplier is introduced and />Constructing an auxiliary function:
wherein Rm (T),V m.1(T) and Vm.2 (T) represents the remorse value and the violation value of the small base station node m at T, respectively;
s3: the estimated task offloads the completed prize value,
estimating according to the function formula (5) in the step S2, and calculating the probability of each small base station node unloading the tasks in the coverage range of the small base station node in an online learning mode, wherein the probability of unloading the tasksIs relaxed from the value constraint of {0,1} to interval [0,1]]Dividing all task space into a plurality of intervals according to the context information of the tasks, setting a weight value for each interval, and calculating the probability of the task i to the small base station node m by adopting an online algorithm for each small base station node m based on the weight value>
S4: the decision to offload a plurality of tasks is completed,
constructing a weighted bipartite graph between the task and the small base station node based on the probability of offloading each task to one or more small base station nodes calculated in step S3, wherein,the method comprises the steps that the weight of an edge (m, i) in a weighted bipartite graph is obtained, an edge with the largest weight is selected in the weighted bipartite graph in a greedy iteration mode, tasks of the corresponding edge are unloaded to small base station nodes corresponding to the edge, and meanwhile the edge is deleted from the bipartite graph;
s5: the state table is updated according to the feedback result,
and (3) selecting tasks to unload according to the unloading scheme in the step (S4), feeding back results after the tasks are completed, updating relevant parameters of the auxiliary function in the step (S2) according to feedback values of the completed tasks, calculating the reward value of the next time slot, and updating Lagrange multipliers according to the following formula:
wherein the auxiliary parameter
2. The task offloading method based on 5G network online learning of claim 1, wherein: in the step S2, a task i value of 1 indicates that the task i is to be offloaded onto the small base station node m at t, and is not to be offloaded at 0.
3. The task offloading method based on 5G network online learning of claim 1, wherein: in the step S2, (1 a) the algorithm constraint condition ensures that the device connected by each small cell node in each time slot does not exceed the maximum connection upper limit, (1 b) the algorithm constraint condition ensures that each task cannot be repeatedly unloaded, (1 c) the algorithm ensures that the number of tasks successfully completed by each small cell node in each time slot is not less than α, (1 d) the constraint condition ensures that the resources occupied by the task on each small cell node in each time slot does not exceed the resource upper limit, and (1 e) the constraint condition defines the decision variableThe value range of (2) is equal to 0 or 1.
4. The task offloading method based on 5G network online learning of claim 1, wherein: in the step S3, the weight value is initialized to 1, and in each time slot, the context information of all tasks is obtained, and classified into the corresponding section according to the context information, and the weight value of the task is estimated by the weight value of the section.
5. The task offloading method based on 5G network online learning of claim 1, wherein: in the step S4, when the task to be unloaded on a small base station node reaches the maximum, all edges taking the small base station node as the vertex are deleted from the bipartite graph, and the unloading decision from the tasks to the small base station nodes is completed according to the greedy iterative mode.
6. The task offloading method based on 5G network online learning of claim 1, wherein: in the step S5, the feedback result is that and />Is a real value of (c).
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011470057.XA CN112600706B (en) | 2020-12-14 | 2020-12-14 | Task unloading method based on 5G network online learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011470057.XA CN112600706B (en) | 2020-12-14 | 2020-12-14 | Task unloading method based on 5G network online learning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112600706A CN112600706A (en) | 2021-04-02 |
| CN112600706B true CN112600706B (en) | 2023-10-27 |
Family
ID=75195602
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011470057.XA Active CN112600706B (en) | 2020-12-14 | 2020-12-14 | Task unloading method based on 5G network online learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112600706B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109951873A (en) * | 2019-02-28 | 2019-06-28 | 华北电力大学 | A kind of Internet of Things mist calculate in information asymmetry do not know lower task unloading mechanism |
| CN110928654A (en) * | 2019-11-02 | 2020-03-27 | 上海大学 | Distributed online task unloading scheduling method in edge computing system |
| CN111918311A (en) * | 2020-08-12 | 2020-11-10 | 重庆邮电大学 | Vehicle networking task unloading and resource allocation method based on 5G mobile edge computing |
-
2020
- 2020-12-14 CN CN202011470057.XA patent/CN112600706B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109951873A (en) * | 2019-02-28 | 2019-06-28 | 华北电力大学 | A kind of Internet of Things mist calculate in information asymmetry do not know lower task unloading mechanism |
| CN110928654A (en) * | 2019-11-02 | 2020-03-27 | 上海大学 | Distributed online task unloading scheduling method in edge computing system |
| CN111918311A (en) * | 2020-08-12 | 2020-11-10 | 重庆邮电大学 | Vehicle networking task unloading and resource allocation method based on 5G mobile edge computing |
Non-Patent Citations (1)
| Title |
|---|
| 张文献 ; 杜永文 ; 张希权 ; .面向多用户移动边缘计算轻量任务卸载优化.小型微型计算机系统.2020,(10),全文. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112600706A (en) | 2021-04-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111726826B (en) | Online task unloading method in base station intensive edge computing network | |
| Al-Turjman et al. | Enhanced deployment strategy for the 5G drone-BS using artificial intelligence | |
| US10805931B2 (en) | Resource allocation method and apparatus for wireless backhaul network based on reinforcement learning | |
| CN102056180B (en) | Method and system for acquiring deployment scheme of wireless local area network (WLAN) access point (AP) | |
| US10866836B1 (en) | Method, apparatus, device and storage medium for request scheduling of hybrid edge computing | |
| CN113346944B (en) | Time delay minimization calculation task unloading method and system in air-space-ground integrated network | |
| JP2005318603A (en) | Method and equipment for forming signal strength model about access point located at arbitrary position | |
| US20080075051A1 (en) | Methods, apparatus and articles for radio frequency planning | |
| CN111915142B (en) | Unmanned aerial vehicle auxiliary resource allocation method based on deep reinforcement learning | |
| Khoramnejad et al. | On joint offloading and resource allocation: A double deep q-network approach | |
| Lakew et al. | Intelligent offloading and resource allocation in hap-assisted mec networks | |
| CN114090108B (en) | Method and device for executing computing task, electronic equipment and storage medium | |
| CN109618285B (en) | Base station control method, device and equipment based on coverage tree | |
| CN112600706B (en) | Task unloading method based on 5G network online learning | |
| CN115278698A (en) | Unmanned aerial vehicle base station dynamic deployment method and device based on dynamic user distribution prediction | |
| Awan et al. | A robust machine learning method for cell-load approximation in wireless networks | |
| CN113821346A (en) | Computation uninstalling and resource management method in edge computation based on deep reinforcement learning | |
| US11265731B1 (en) | Small cell telecommunications network design | |
| Wang et al. | Optimization of deployable base stations with guaranteed QoE in disaster scenarios | |
| CN114461299B (en) | Unloading decision determining method and device, electronic equipment and storage medium | |
| CN103648105A (en) | Method and system for obtaining wireless local area network (WLAN) access point (AP) disposition scheme | |
| CN116185523A (en) | Task unloading and deployment method | |
| CN114599102A (en) | Method for unloading linear dependent tasks of edge computing network of unmanned aerial vehicle | |
| CN114281544A (en) | Electric power task execution method and device based on edge calculation | |
| CN113468819A (en) | Energy consumption optimization method based on genetic algorithm and adopting unmanned aerial vehicle to assist edge calculation |
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 | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20231226 Address after: Room 1109, 11th Floor, Building 4, Yinhu Huayuan, Yinhu Street, Fuyang District, Hangzhou City, Zhejiang Province, 311422 Patentee after: Zhejiang Yisu Network Technology Co.,Ltd. Address before: Room 1914, xieying building, 99 Tuanjie Road, yanchuangyuan, Nanjing area, China (Jiangsu) pilot Free Trade Zone, Nanjing, Jiangsu, 211800 Patentee before: Nanjing Wanban Shangpin Information Technology Co.,Ltd. |