CN114928893B - Architecture based on intelligent reflecting surface and task unloading method - Google Patents
Architecture based on intelligent reflecting surface and task unloading method Download PDFInfo
- Publication number
- CN114928893B CN114928893B CN202210699515.XA CN202210699515A CN114928893B CN 114928893 B CN114928893 B CN 114928893B CN 202210699515 A CN202210699515 A CN 202210699515A CN 114928893 B CN114928893 B CN 114928893B
- Authority
- CN
- China
- Prior art keywords
- reflecting surface
- network
- vehicle
- pri
- intelligent
- 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 17
- 238000013507 mapping Methods 0.000 claims abstract description 19
- 230000002787 reinforcement Effects 0.000 claims abstract description 19
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 12
- 238000004891 communication Methods 0.000 claims abstract description 8
- 238000012549 training Methods 0.000 claims description 12
- 238000005457 optimization Methods 0.000 claims description 9
- 238000013527 convolutional neural network Methods 0.000 claims description 7
- 238000005265 energy consumption Methods 0.000 claims description 6
- 230000009466 transformation Effects 0.000 claims description 6
- 230000005540 biological transmission Effects 0.000 claims description 5
- 230000006870 function Effects 0.000 claims description 5
- 238000011478 gradient descent method Methods 0.000 claims description 4
- 238000013468 resource allocation Methods 0.000 claims description 4
- 230000009977 dual effect Effects 0.000 claims description 2
- 230000002776 aggregation Effects 0.000 claims 1
- 238000004220 aggregation Methods 0.000 claims 1
- 238000005516 engineering process Methods 0.000 abstract description 4
- 239000003795 chemical substances by application Substances 0.000 description 5
- 230000008569 process Effects 0.000 description 4
- 238000001210 attenuated total reflectance infrared spectroscopy Methods 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000006855 networking Effects 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000013508 migration Methods 0.000 description 1
- 230000005012 migration Effects 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
- 238000012545 processing Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000012552 review Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 238000013526 transfer learning Methods 0.000 description 1
- 239000013598 vector Substances 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W72/00—Local resource management
- H04W72/50—Allocation or scheduling criteria for wireless resources
- H04W72/54—Allocation or scheduling criteria for wireless resources based on quality criteria
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/30—Services specially adapted for particular environments, situations or purposes
- H04W4/40—Services specially adapted for particular environments, situations or purposes for vehicles, e.g. vehicle-to-pedestrians [V2P]
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W72/00—Local resource management
- H04W72/50—Allocation or scheduling criteria for wireless resources
- H04W72/53—Allocation or scheduling criteria for wireless resources based on regulatory allocation policies
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D30/00—Reducing energy consumption in communication networks
- Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Quality & Reliability (AREA)
- Mobile Radio Communication Systems (AREA)
Abstract
The invention provides an architecture based on an intelligent reflecting surface and a task unloading method, and relates to the field of automatic driving. The architecture and task unloading method based on the intelligent reflecting surface comprises the following steps: s1, establishing a wireless communication channel of a vehicle-intelligent reflecting surface-edge computing server for task unloading; s2, reducing the time delay of the system; s3, mapping the edge equipment into a digital twin body by the mapping plane layer; s4, randomly selecting data from historical data by a Deep Reinforcement Learning (DRL) algorithm; s5, the digital twin network DTN performs unloading decision on the edge network EN through the control plane layer, and the reflection of the intelligent reflecting surface indicates feedback of elements; s6, repeating the steps S1-S6 until the termination condition is reached. By introducing a digital twin technology, the vehicle state and the intelligent reflecting surface state in the Internet of vehicles are dynamically monitored, the Internet of vehicles computing resources are cooperatively scheduled, unloading decisions are reasonably distributed, and the task unloading channel quality is enhanced through the intelligent reflecting surface.
Description
Technical Field
The invention relates to the technical field of automatic driving, in particular to an architecture based on an intelligent reflecting surface and a task unloading method.
Background
Autopilot cars benefit from the rapid development of internet of vehicles (Internet of Vehicle, ioV) technology that requires high quality radio transmission, intelligent navigation systems, and real-time image, audio, or sensor data analysis capabilities, however, limited by IoV limited communication and computing resources and high mobility channel environments, how to efficiently process internet of vehicles generated data in real-time to efficiently offload decisions for user tasks, and improving resource utilization is critical to research in internet of vehicles.
The Digital Twin (DT) technology can effectively solve the problems of real-time data processing and resource collaborative scheduling, DT is a high-fidelity Digital representation physical reality, integrated simulation and service data are used, the whole life cycle of a physical entity is monitored at the same time, and under an automatic driving scene, the Digital Twin can accurately predict the driving process and the state of a vehicle, including the resource scheduling prediction of vehicle tasks and the maintenance and overhaul evaluation of the vehicle, so that the safe operation of the vehicle and the efficient resource scheduling of a vehicle networking system are ensured.
At present, a task unloading scheme based on digital twin support in the Internet of vehicles is mostly researched from resource allocation, and a task unloading strategy is adaptively allocated to the task by taking the improvement of the resource utilization rate as a target, however, as the mobility of vehicles in the Internet of vehicles is strong, a channel between the vehicles and an edge server is easily interfered by obstacles such as buildings, and the quality of a wireless channel is greatly reduced, and the problem of allocation of unloading decisions is also influenced when the current working scheme considers the limitation of resources and does not consider the interference of the obstacles to the channel.
Disclosure of Invention
Aiming at the defects of the prior art, the invention provides an architecture based on an intelligent reflecting surface and a task unloading method, and solves the problem that the prior scheme does not consider interference of obstacles to a channel while considering resource limitation.
In order to achieve the above purpose, the invention is realized by the following technical scheme: an architecture and task unloading method based on an intelligent reflecting surface comprises the following steps:
s1, establishing a wireless communication channel of a vehicle-intelligent reflecting surface-edge computing server for task unloading;
s2, establishing a target optimization function controlled by the intelligent reflecting surface (Intelligent Reflecting Surface, IRS) to reduce the time delay of the system;
s3, mapping the edge equipment into digital twin body mapping by the mapping plane layer, wherein the mapping represents the running state information of the edge network EN, the digital twin network DTN not only monitors the state of the network in real time, but also records the historical running state of the network, so that the digital twin body mapping is used for task unloading based on deep reinforcement learning DRL and intelligent reflection plane control combined optimization algorithm learning, and digital representation is converted into Tensor form which can be processed by the deep reinforcement learning DRL plane layer;
s4, randomly selecting data from the historical data by a deep reinforcement learning DRL algorithm, and training a model on a deep reinforcement learning DRL plane layer;
s5, the digital twin network DTN performs unloading decision on the edge network EN and feedback of reflection indicating elements of the intelligent reflecting surface through the control plane layer, the edge network EN performs control of the intelligent reflecting surface according to the feedback information, and resource allocation is performed on task unloading of the vehicle;
s6, repeating the steps S1-S6 until reaching a termination condition, wherein the termination condition can be the end of the life cycle of the Internet of vehicles.
Preferably, the operation state information in S3 includes vehicle operation state information Vehicle State Information, VSI), intelligent reflection plane operation state information (IRS State Information, ISI), and the historical operation state in S3 includes real-time rewards review, time delay and energy consumption Energy Consumption of the deep reinforcement learning DRL agent.
The invention provides an architecture based on an intelligent reflecting surface and a task unloading method. The beneficial effects are as follows:
1. according to the invention, by introducing a digital twin technology, the vehicle state and the intelligent reflecting surface state in the Internet of vehicles are dynamically monitored, the Internet of vehicles computing resources are cooperatively scheduled, the unloading decision is reasonably allocated, and the task unloading channel quality is enhanced through the intelligent reflecting surface.
2. According to the invention, task unloading and IRS regulation are simultaneously carried out through a two-stage algorithm based on (Double Deep Q Network, DDQN), the dynamic change of the environment is self-adapted, and the overall task unloading time delay of the system is reduced.
3. The invention combines the deep reinforcement learning and the transfer learning innovatively, improves the reusability of the model, and further improves the training efficiency of the model.
Drawings
FIG. 1 is a diagram of the structure of a DTVIF of the present invention;
FIG. 2 is a schematic diagram of the steps of the present invention.
Detailed Description
The following description of the embodiments of the present invention 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 invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The present invention proposes a Digital Twin driven vehicle mission offloading and intelligent reflector configuration framework (Digital Twin-Driven Vehicular Task Offloading and IRS Configuration Framework, DTVIF) to effectively monitor, learn and manage IoV, employ mobile edge computing MEC and intelligent reflector IRS to provide enhanced computing power for the vehicle and improve transmission performance when the vehicle communicates with the mobile edge computing MEC server, the Digital Twin DT being used to implement real-time data acquisition and Digital representation of IoV physical operating environments to better support decisions, the DTVIF comprising, with reference to fig. 1, three components: the core cloud network CCN, the edge network EN and the digital twin network DTN, the DTN and the EN are controlled by the CCN, the DT controller is installed in the CCN, the CCN provides DTN management in the cloud server, and the DTN consists of three planes: the details of the mapping plane, DRL plane and control plane are described below:
mapping planes: EN consists of communication devices such as vehicles, base station BSs, IRSs, MEC servers, etc., MEC servers are placed in BSs, IRSs may be installed on the outside walls of the building, MEC servers help provide enhanced resources to support the vehicles due to their limited computing power, in a car networking scenario, vehicles may select the nearest infrared to communicate, and IRSs may select the nearest MEC server to establish a tunnel.
DRL plane: on the plane, real-time state is fused and input into a deep reinforcement learning DRL-based algorithm, a DRL-based task offloading and intelligent reflection surface control joint optimization algorithm is deployed on the DRL plane, historical data and real-time data can be analyzed and learned so as to carry out task offloading decision and channel quality required by task transmission, and in DTVIF, the aim is to minimize energy consumption and delay by adjusting reflection element coefficients and offloading decision.
When personalized services such as car radio, GPS navigation, route planning and the like are needed, a task unloading and intelligent reflection surface control joint optimization algorithm based on the DRL can make decisions of task unloading and IRS configuration according to the environment, and whether an MEC server is needed for task execution can be determined by environment feedback.
Examples:
as shown in fig. 1-2, an embodiment of the present invention provides an architecture and task offloading method based on an intelligent reflection plane, including the following steps:
s1, establishing a wireless communication channel of a vehicle-intelligent reflecting surface-edge computing server for task unloading;
in this step of the present embodiment, the wireless communication channel of the task offloaded vehicle-intelligent reflection-edge computing server is established by:
h k ＝h k,m +G H Θ H h r,m (1)
h k for the channel gain of vehicle k to the edge server. Wherein G is H Represents the channel gain, Θ, between the IRS and the vehicle k H Representing the reflection coefficient of IRS, and h r,m Representing the channel gain of the IRS to the edge server.
R k,m (t) transfer rate between vehicle k and edge server, wherein W k,m (t) is the bandwidth at time t;beamforming vectors, and must satisfy: />
S2, establishing a target optimization function for task unloading and intelligent reflection surface control so as to reduce the time delay of the system;
in this step of this embodiment, since the vehicle has a certain computing power, when the vehicle directly executes the task locally, the generated time delay is:
vehicle k sets the task size to X k When the data of (t) is unloaded to the edge server for execution, the transmission delay is as follows:
wherein f s (t) is the frequency of the CPU of the MEC server, θ (t) is the rotation speed of the CPU, f l And (t) is the frequency of the local CPU.
Wherein C1 is a discrete variable, equation (5) is a non-convex optimization problem, and constraint C2 ensures channel power p between MEC server and vehicle k Not greater than maximum power P max Constraint C3 ensures channel bandwidth W between MEC server and vehicle k k,m Not greater than the maximum bandwidth W in the life cycle max C4 limits the maximum number of vehicles V and the life cycle T.
S3, mapping the edge equipment into digital twin body mapping by the mapping plane layer, wherein the mapping represents the running state information of the edge network EN, the digital twin network DTN not only monitors the state of the network in real time, but also records the historical running state of the network, so that the digital twin body mapping is used for task unloading based on deep reinforcement learning DRL and intelligent reflection plane control combined optimization algorithm learning, and digital representation is converted into Tensor form which can be processed by the deep reinforcement learning DRL plane layer;
in this step of this embodiment, in each time slot t, the state of the current vehicle is acquired by the agent: the position P, the computing resource C and the communication resource W are then fused into a Tensor form of [ P, C, W ] and sent to an unloading decision ODM module for training, and meanwhile, the IRS configuration ICM module also needs to acquire [ P, C, W ] and the intelligent reflecting surface state I for training,
S4, randomly selecting data from the historical data by a deep reinforcement learning DRL algorithm, and training a model on a deep reinforcement learning DRL plane layer;
in this step of this embodiment, the present embodiment uses ψ in the ODM Pri Extracting features at ψ Pri Extracting spatial correlation from convolutional neural network CNN, performing feature transformation by using fully connected network FC layer to conform to strategy output dimension, and performing training of dual depth Q network DDQN in the first stage by using the following formula:
wherein ψ is Tar Network parameters in (a)Synchronizing at regular intervals the signals from ψ Pri R is a prize, i.e. r= -L (t), γ s For discounting factor, use L 1 Solving the error and updating the ψ by using a gradient descent method Pri Parameter of (1) and ψ Pri The output of (i) is task offloading policy, [ P, C, W ]]' is the state of the intelligent agent [ P, C, W ]]The state after policy y is taken down, so ψ Tar The function of (1) is to let ψ Pri Is more stable, in the second phase, the network parameters of the first phase are migrated to phi Pri Structurally identical parts, at Φ Pri In the method, [ P, C and W ] are respectively extracted by using a convolutional neural network CNN]And the spatial correlation of the state I of the intelligent reflecting surface, and finally, respectively utilizing the fully connected FC neural network to perform characteristic transformation and aggregate to enable the characteristic transformation to conform to the surface output dimension of the reflecting element, and then, utilizing the following formula to perform the training of the DDQN of the second stage:
wherein phi is Tar Network parameters in (a)Synchronizing at regular intervals from phi Pri R is a prize, i.e. r= -L (t), γ p For discounting factor, use L 2 Solving the error and updating phi by using a gradient descent method Pri And phi is the parameter in (a) Pri The output of (a) is the output of the reflection indicating element [ P, C, W ]]'and I' are the conditions of the agent in the vehicle [ P, C, W ]]And intelligent reflector operating state I is taken such that ψ Pri Strategically placed vehicle with maximum output valueVehicle state status and intelligent reflector operating state.
S5, the digital twin network DTN performs unloading decision on the edge network EN and feedback of reflection indicating elements of the intelligent reflecting surface through the control plane layer, the edge network EN performs control of the intelligent reflecting surface according to the feedback information, and resource allocation is performed on task unloading of the vehicle;
in this step of this embodiment, the resulting offloading decisions y and IRS indicate element coefficientsRespectively fed back to the vehicles in EN and the intelligent reflecting surface.
S6, repeating the steps S1-S6 until reaching a termination condition, wherein the termination condition can be the end of the life cycle of the Internet of vehicles.
In this embodiment, steps S1-S6 are repeated until the maximum lifecycle T is reached.
The running state information in S3 includes the vehicle running state information VSI and the intelligent reflection plane running state information ISI, and the historical running state in S3 includes the real-time rewards Reward, the time delay Energy Consumption of the deep reinforcement learning DRL agent.
In the embodiment, the running state of the internet of vehicles is dynamically monitored through digital twinning, the problems of limited resources and channel quality in the calculation and unloading process are comprehensively considered, a task unloading channel between the vehicle and the intelligent reflecting surface and the MEC server is established, a double Q learning network (Double Deep Q Network, DDQN) reinforcement learning algorithm embedded in the digital twinning network can perform calculation and unloading strategy distribution and IRS regulation and control, the self-adaption mobility internet of vehicles environment is realized, a convolution layer is adopted in the double Q learning network (Double Deep Q Network, DDQN) to extract the spatial similarity of the vehicle and the intelligent reflecting surface, and the training efficiency and stability are improved through parameter synchronization and parameter migration in the training process.
Although embodiments of the present invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made therein without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.
Claims (2)
1. An architecture and task unloading method based on an intelligent reflecting surface is characterized in that: the method comprises the following steps:
s1, establishing a wireless communication channel of a vehicle-intelligent reflecting surface-edge computing server for task unloading:
h k ＝h k,m +G H Θ H h r,m
h k channel gain for vehicle k to edge server, where G H Represents the channel gain, Θ, between the IRS and the vehicle k H Representing the reflection coefficient of IRS, and h r,m Representing the channel gain of the IRS to the edge server;
s2, establishing a target optimization function for task unloading and intelligent reflector control to reduce the time delay of the system, wherein the time delay generated when the vehicle locally executes the task is as follows:
vehicle k sets the task size to X k When the data of (t) is unloaded to the edge server for execution, the transmission delay is as follows:
wherein f s (t) is the frequency of the CPU of the MEC server, θ (t) is the rotation speed of the CPU, f l (t) is the frequency of the local CPU;
s3, mapping the edge equipment into digital twin body mapping by the mapping plane layer, wherein the mapping represents the running state information of the edge network EN, the digital twin network DTN not only monitors the state of the network in real time, but also records the historical running state of the network, so that the digital twin body mapping is used for task unloading based on deep reinforcement learning DRL and intelligent reflection plane control combined optimization algorithm learning, and digital representation is converted into Tensor form which can be processed by the deep reinforcement learning DRL plane layer;
s4, randomly selecting data from historical data by a deep reinforcement learning DRL algorithm, training a model on a deep reinforcement learning DRL plane layer, and utilizing ψ in an ODM Pri Extracting features at ψ Pri Extracting spatial correlation from convolutional neural network CNN, performing feature transformation by using fully connected network FC layer to conform to strategy output dimension, and performing training of dual depth Q network DDQN in the first stage by using the following formula:
wherein ψ is Tar Network parameters in (a)Synchronizing at regular intervals the signals from ψ Pri R is a prize, i.e. r= -L (t), γ s For discounting factor, use L 1 Solving the error and updating the ψ by using a gradient descent method Pri Parameter of (1) and ψ Pri The output of (i) is task offloading policy, [ P, C, W ]]' is the state of the intelligent agent [ P, C, W ]]The state after policy y is taken down, so ψ Tar The function of (1) is to let ψ Pri Is more stable, in ICM, the network parameters of the first stage are migrated to phi Pri Structurally identical parts, at Φ Pri In which [ P, C, W ] are extracted by CNN]And the spatial correlation of the state I of the intelligent reflecting surface, finally, performing characteristic transformation and aggregation by using FC to enable the characteristic transformation to conform to the surface output dimension of the reflecting element, and then performing the training of the DDQN of the second stage by using the following formula:
wherein phi is Tar Network parameters in (a)Synchronizing at regular intervals from phi Pri R is a prize, i.e. r= -L (t), γ p For discounting factor, use L 2 Solving the error and updating phi by using a gradient descent method Pri And phi is the parameter in (a) Pri The output of (a) is the output of the reflection indicating element [ P, C, W ]]'and I' are the conditions of the agent in the vehicle [ P, C, W ]]And intelligent reflector operating state I is taken such that ψ Pri Outputting the vehicle state with the largest value after the strategy and the intelligent reflecting surface running state;
s5, the digital twin network DTN performs unloading decision on the edge network EN and feedback of reflection indicating elements of the intelligent reflecting surface through the control plane layer, the edge network EN performs control of the intelligent reflecting surface according to the feedback information, and resource allocation is performed on task unloading of the vehicle;
s6, repeating the steps S1-S6 until reaching a termination condition, wherein the termination condition can be the end of the life cycle of the Internet of vehicles.
2. The architecture and task offloading method of claim 1, wherein: the operation state information in the S3 comprises vehicle operation state information VSI and intelligent reflection surface operation state information ISI, and the historical operation state in the S3 comprises real-time rewards Reward, time delay and energy consumption Energy Consumption of the deep reinforcement learning DRL agent.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210699515.XA CN114928893B (en) | 2022-06-20 | 2022-06-20 | Architecture based on intelligent reflecting surface and task unloading method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210699515.XA CN114928893B (en) | 2022-06-20 | 2022-06-20 | Architecture based on intelligent reflecting surface and task unloading method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114928893A CN114928893A (en) | 2022-08-19 |
| CN114928893B true CN114928893B (en) | 2024-04-16 |
Family
ID=82814857
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210699515.XA Active CN114928893B (en) | 2022-06-20 | 2022-06-20 | Architecture based on intelligent reflecting surface and task unloading method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114928893B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117793754B (en) * | 2023-11-28 | 2024-06-25 | 山东大学 | Intelligent reflecting surface communication system deduction optimization method and system based on digital twinning |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| IT7825102A0 (en) * | 1977-07-08 | 1978-06-29 | Ibm | EQUIPMENT FOR PROCESSING TASKS FOR A COMPUTING SYSTEM. |
| CN112118601A (en) * | 2020-08-18 | 2020-12-22 | 西北工业大学 | Method for reducing task unloading delay of 6G digital twin edge computing network |
| CN113163497A (en) * | 2021-03-29 | 2021-07-23 | 南京航空航天大学 | Computing efficiency optimization method in millimeter wave mobile edge computing system based on reconfigurable intelligent surface |
| CN113543176A (en) * | 2021-07-08 | 2021-10-22 | 中国科学院深圳先进技术研究院 | Unloading decision method of mobile edge computing system based on assistance of intelligent reflecting surface |
| CN113726471A (en) * | 2021-07-20 | 2021-11-30 | 西安交通大学 | Intelligent reflection surface auxiliary MIMO covert communication system and parameter optimization method |
| WO2022027776A1 (en) * | 2020-08-03 | 2022-02-10 | 威胜信息技术股份有限公司 | Edge computing network task scheduling and resource allocation method and edge computing system |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US12017352B2 (en) * | 2020-10-29 | 2024-06-25 | Nvidia Corporation | Transformation of joint space coordinates using machine learning |
-
2022
- 2022-06-20 CN CN202210699515.XA patent/CN114928893B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| IT7825102A0 (en) * | 1977-07-08 | 1978-06-29 | Ibm | EQUIPMENT FOR PROCESSING TASKS FOR A COMPUTING SYSTEM. |
| WO2022027776A1 (en) * | 2020-08-03 | 2022-02-10 | 威胜信息技术股份有限公司 | Edge computing network task scheduling and resource allocation method and edge computing system |
| CN112118601A (en) * | 2020-08-18 | 2020-12-22 | 西北工业大学 | Method for reducing task unloading delay of 6G digital twin edge computing network |
| CN113163497A (en) * | 2021-03-29 | 2021-07-23 | 南京航空航天大学 | Computing efficiency optimization method in millimeter wave mobile edge computing system based on reconfigurable intelligent surface |
| CN113543176A (en) * | 2021-07-08 | 2021-10-22 | 中国科学院深圳先进技术研究院 | Unloading decision method of mobile edge computing system based on assistance of intelligent reflecting surface |
| CN113726471A (en) * | 2021-07-20 | 2021-11-30 | 西安交通大学 | Intelligent reflection surface auxiliary MIMO covert communication system and parameter optimization method |
Non-Patent Citations (3)
| Title |
|---|
| A Mobility Aware Task Offloading Scheme For Vehicle Edge Computing;Jing Tang;《WCSP》;20211231;全文 * |
| 基于DQN的车载边缘网络任务分发卸载算法;赵海涛;张唐伟;陈跃;赵厚麟;朱洪波;;通信学报;20201231(第10期);全文 * |
| 雾计算网络中基于移动感知的任务卸载和资源分配;陈雷;《武汉大学学报》;20211231;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114928893A (en) | 2022-08-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Zhang et al. | Energy-efficient trajectory optimization for UAV-assisted IoT networks | |
| CN112104502B (en) | Time-sensitive multitask edge computing and cache cooperation unloading strategy method | |
| CN113286329B (en) | Communication and computing resource joint optimization method based on mobile edge computing | |
| CN111915142B (en) | Unmanned aerial vehicle auxiliary resource allocation method based on deep reinforcement learning | |
| CN113254188B (en) | Scheduling optimization method and device, electronic equipment and storage medium | |
| CN115297171B (en) | Edge computing and unloading method and system for hierarchical decision of cellular Internet of vehicles | |
| CN115209426B (en) | Dynamic deployment method for digital twin servers in edge car networking | |
| CN114928893B (en) | Architecture based on intelligent reflecting surface and task unloading method | |
| CN114650567A (en) | Unmanned aerial vehicle-assisted V2I network task unloading method | |
| CN115037751B (en) | Unmanned aerial vehicle-assisted heterogeneous Internet of vehicles task migration and resource allocation method | |
| Sha et al. | DRL-based task offloading and resource allocation in multi-UAV-MEC network with SDN | |
| CN114626298A (en) | State updating method for efficient caching and task unloading in unmanned aerial vehicle-assisted Internet of vehicles | |
| CN115659803A (en) | Intelligent unloading method for computing tasks under unmanned aerial vehicle twin network mapping error condition | |
| CN113573363A (en) | MEC calculation unloading and resource allocation method based on deep reinforcement learning | |
| CN116916337A (en) | Task unloading resource allocation method for unmanned aerial vehicle mobile edge computing network system based on intelligent reflection surface assistance | |
| CN114363803A (en) | Energy-saving multi-task allocation method and system for mobile edge computing network | |
| CN114866979A (en) | Unmanned aerial vehicle-assisted edge calculation time delay minimization method | |
| Wang et al. | Task offloading and resource allocation for UAV-assisted mobile edge computing with imperfect channel estimation over Rician fading channels | |
| CN111930435B (en) | Task unloading decision method based on PD-BPSO technology | |
| CN116723548A (en) | Unmanned aerial vehicle auxiliary calculation unloading method based on deep reinforcement learning | |
| CN116737391A (en) | Edge computing cooperation method based on mixing strategy in federal mode | |
| CN117528649A (en) | Method for establishing end-edge cloud system architecture, task unloading and resource allocation optimization method and end-edge cloud system architecture | |
| CN116137724A (en) | Task unloading and resource allocation method based on mobile edge calculation | |
| CN115134370A (en) | Multi-unmanned-aerial-vehicle-assisted mobile edge calculation unloading method | |
| CN114980160A (en) | Unmanned aerial vehicle-assisted terahertz communication network joint optimization method and device |
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 |