CN115767636A - Vehicle-mounted application task unloading decision obtaining method, unloading method and traffic system - Google Patents

Vehicle-mounted application task unloading decision obtaining method, unloading method and traffic system Download PDF

Info

Publication number
CN115767636A
CN115767636A CN202211398518.6A CN202211398518A CN115767636A CN 115767636 A CN115767636 A CN 115767636A CN 202211398518 A CN202211398518 A CN 202211398518A CN 115767636 A CN115767636 A CN 115767636A
Authority
CN
China
Prior art keywords
task
unloading
overall
node
decision
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.)
Pending
Application number
CN202211398518.6A
Other languages
Chinese (zh)
Inventor
权恒友
朱文霞
王洪涛
赵振博
王志斌
邱文利
许忠印
张博
张少波
董立强
张莹
李海冬
徐建声
邱宇
李占强
魏丽松
麻文进
冯兴乐
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Hebei Xiong'an Jingde Expressway Co ltd
Changan University
Original Assignee
Hebei Xiong'an Jingde Expressway Co ltd
Changan University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Hebei Xiong'an Jingde Expressway Co ltd, Changan University filed Critical Hebei Xiong'an Jingde Expressway Co ltd
Priority to CN202211398518.6A priority Critical patent/CN115767636A/en
Publication of CN115767636A publication Critical patent/CN115767636A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • 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

  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention discloses a vehicle-mounted application task unloading decision obtaining method, an unloading method and a traffic system. In order to jointly optimize unloading strategies, channel selection and calculation resource allocation, the vehicle energy consumption minimization target under delay constraint is realized, and the original optimization problem is converted into a two-stage optimization sub-problem; aiming at the coupling between the calculation unloading and the resource allocation, a two-stage joint optimization algorithm is provided, firstly, a candidate mode set strategy is adopted to optimize a set of candidate calculation modes in an initialization stage so as to reduce the complexity of the algorithm, then, a superior combination optimization problem is solved based on an optimal worst ant system algorithm, and channel selection and unloading decisions are determined. The scheme provided by the invention has better convergence, and can effectively improve the task completion rate compared with other algorithms.

Description

Vehicle-mounted application task unloading decision obtaining method, unloading method and traffic system
Technical Field
The invention belongs to the technical field of vehicle networking edge computing, and particularly relates to a V2X computing unloading and resource allocation scheme based on MEC.
Background
In recent years, an Intelligent Transportation System (ITS) has played an important role in road safety, traffic control, and the like. The internet of vehicles (IoV) is used as an important component of the ITS, information sharing between vehicles (V2V) is carried out through an on-board unit (OBU) integrated by devices such as a computing device, a wireless interface, a sensor and a GPS, information interaction between the vehicles and an infrastructure (V2I) is carried out through Road Side Units (RSU) on two sides of a road, and therefore driving experience is improved, and traffic accident incidence is reduced.
The intelligent vehicle relies on the OBU to provide computing support for novel application programs such as position service, online games and safe driving, however, with continuous progress of scientific technology, the IoV has emerged to be applied in intensive computing and delay sensitive applications such as augmented reality, virtual reality and natural language processing, and the limited vehicle-mounted computing capability cannot continuously meet the computing requirements of the real-time applications.
In IoV, time delay is one of key factors influencing dynamic information interaction, and the longer the time delay is, the greater hidden danger exists in traffic safety. Because the limited vehicle-mounted computing capacity cannot meet the requirements of computing-intensive and delay-sensitive applications on computing capacity and delay, researchers introduce cloud computing into IoV and process vehicle-mounted tasks by means of the powerful computing capacity of a cloud center, but the problem of high delay still exists in long-distance transmission between the cloud center and a vehicle-mounted terminal.
In order to further reduce time delay, mobile Edge Computing (MEC) is introduced into IoV, and an MEC server is deployed near a vehicle-mounted terminal to provide nearby service for a vehicle-mounted task, so that the problem of insufficient computing power of the vehicle-mounted terminal can be solved, and the requirement of low time delay of the task can be met. In addition, the energy consumption problem also occupies a very important position in the vehicle networking communication, and especially for a novel energy vehicle such as an electric vehicle, the energy consumption directly influences the cruising ability of the vehicle.
Based on the MEC, the vehicle unloads the calculation task to the edge server for processing, thereby effectively reducing the calculation time delay and the energy consumption. At present, researches mostly focus on taking the balance of total time delay and energy consumption of a system or the balance of the total time delay and the energy consumption of the system as an optimization target, and an optimal strategy for calculating unloading and resource allocation is solved by constructing a mixed integer nonlinear programming problem and utilizing convex optimization, deep learning and other artificial intelligence methods.
However, the MEC server has limited computing resources, and strong coupling exists between the computing offloading and the resource allocation, which makes direct solution difficult, so that the computing capacity of the vehicle itself should be fully considered, and the computing offloading and resource allocation strategies should be reasonably formulated, thereby improving the resource utilization rate.
Disclosure of Invention
Aiming at the defects or shortcomings of the prior art, the invention provides a decision obtaining method based on unloading of vehicle-mounted application tasks, which is used for obtaining a decision that N tasks are transmitted and unloaded to N +1 nodes through L channels, wherein the N +1 nodes are composed of N vehicles and an MEC (media independent center) server, the N tasks are generated by the N vehicles, and each vehicle generates one task; any task U i ＝{D i ,C i ,T i Deadline ,T i stay In which, N is more than or equal to 2,
Figure BDA0003934069910000021
i＝1,2,3,...,N，D i is the amount of data of any task, C i The computing resource required for any of the tasks, T i Deadline For maximum delay tolerance of any task, T i stay The duration of any task in the cell formed by the N +1 nodes is related to the vehicle position and the vehicle speed, the task needs to be completed within the maximum delay tolerance and the cell duration, otherwise the task fails to be executed.
To this end, the method of the invention comprises:
step 1, respectively generating a feasible candidate mode set of each task, any task U i The feasible candidate mode set comprises a plurality of mode combinations, each mode combination is composed of a channel and a node, any mode combination satisfies the condition of the formula (1), and any mode combination is composed of a channel k and a node j;
Figure BDA0003934069910000022
in formula (1):
Figure BDA0003934069910000023
is a task U i The minimum computational resources required to offload to node j through channel k, j =0,1,2, 3.., N, j =0 represents the MEC server;
Figure BDA0003934069910000031
j = i denotes the task U i Unloading the local vehicle; r is ijk Is a task U i The data transmission rate unloaded to the node j through the channel k can be calculated by a Shannon formula;
Figure BDA0003934069910000032
the maximum computing resource which can be provided by the node j; the value is a specific attribute value of the computing node;
step 2, respectively adopting an ant colony algorithm to obtain an unloading decision of each task in a corresponding feasible candidate mode set; the offload decisions for all tasks constitute the overall offload decision x, x = { x = { ijk }，x ijk Is a task U i Decision to offload to node j via channel k, if task U i Unloaded on node j through channel k, then x ijk =1, otherwise x ijk =0, i =1,2,3,. Cnj =0,1,2,3,. Cndot, N, k =1,2,3,. Cndot, L; initially, taking 0.5-1.5% of all pheromones;
step 3, after the step 2 is executed again, the step 4 is executed;
step 4, adopting the following formula (2) to respectively evaluate the fitness of all the overall unloading decisions obtained in the current iteration process and the previous iteration process, wherein F 1 The overall distribution decision corresponding to the maximum value is an overall optimal solution, F 1 The overall unloading decision corresponding to the minimum value is an overall worst solution;
Figure BDA0003934069910000033
x in the formula (2) ijk As an element in an evaluated global offload decision;
if there are a plurality of F's of the same size 1 Maximum or multiple of equal-sized F 1 The minimum value, then adopt the equation (3) to carry on the fitness evaluation to all overall distribution decisions obtained in the current and previous iteration process respectively, F 2 The overall distribution decision corresponding to the minimum value is an overall optimal solution, F 2 The overall distribution decision corresponding to the maximum value is an overall worst solution;
Figure BDA0003934069910000041
in formula (3): x is the number of ijk Elements in the evaluated overall offloading decision;
Figure BDA0003934069910000042
is a task U i Calculating the generated energy consumption at the node j;
Figure BDA0003934069910000043
is a task U i The transmission energy consumption generated by unloading to the node j through the channel k, and t is the meaning of transmission;
Figure BDA0003934069910000044
r ijk is a task U i The data transfer rate offloaded to node j through channel k,
step 5, updating ant pheromones on each task unloading path in the overall optimal solution according to the formula (4); updating ant pheromones on each task unloading path in the overall worst solution according to the formula (5):
Figure BDA0003934069910000045
Figure BDA0003934069910000046
in formulae (4) and (5):
Figure BDA0003934069910000047
representing a task U in the current global optimal solution i Unloading to a node j through k;
rho represents the volatilization speed of the ant pheromone, and the value range is 0.1-0.99;
Δτ ijk the ant pheromone increment is used for enhancing the pheromone concentration on the channel path and improving the integral optimizing capability;
Figure BDA0003934069910000048
Figure BDA0003934069910000049
task U for representing and selecting current overall optimal solution i Energy consumption generated by unloading is transmitted to the node j through the channel k;
τ′ ijk the ant pheromone is the ant pheromone before updating;
τ ijk the updated ant pheromone;
Figure BDA0003934069910000051
represents the task U when the current overall worst solution is selected i Energy consumption generated by unloading is transmitted to a computing node j through a channel k;
epsilon is the worst ant pheromone updating parameter, epsilon is more than 0;
and 6, taking the pheromone updated in the step 5, and iteratively executing the steps 2, 4 and 5 until the overall optimal solutions obtained by the previous iteration and the next iteration are the same, wherein the overall optimal solution obtained at the moment is the overall unloading decision of all tasks.
The invention also provides a vehicle-mounted application task unloading method. The task unloading method provided by the invention adopts the method to obtain the whole unloading decision, and then the whole unloading decision is obtained according to the methodThe overall unloading decision unloads each task to the corresponding node through the corresponding channel, and the unloading resource of any task
Figure BDA0003934069910000052
The allocation satisfies the condition of equation (6):
Figure BDA0003934069910000053
the invention also provides a traffic system. The traffic system provided by the invention comprises a plurality of cells, wherein each cell comprises a task unloading system and an MEC server; the task unloading system is used for collecting information of a plurality of vehicles and tasks passing through the cell and unloading the tasks to each node of the cell by adopting the method.
The unloading scheme of the invention improves the computing capability of the network edge by closely combining the cooperative work of the vehicle, other vehicles and the MEC server, solves the problem of minimizing the energy consumption of the vehicle under the delay constraint and optimizes the unloading decision, channel selection and computing resource allocation. The invention ensures that the unloading strategy has better optimizing capability and convergence, and effectively improves the task completion rate.
Drawings
FIG. 1 is a graph showing a comparison of convergence for different numbers of vehicles in example 1 of the present invention using the method of the present invention;
FIG. 2 is a graph comparing the convergence of ACS-WMO in example 2 of the present invention;
FIG. 3 is a comparison of the method of the present invention and an exhaustive algorithm used in example 3 of the present invention;
fig. 4 is a comparison diagram of task completion numbers of the method of the present invention and other algorithms in embodiment 4 of the present invention.
Detailed Description
Unless otherwise defined, scientific and technical terms used herein are to be understood as commonly understood by one of ordinary skill in the relevant art.
Step 2 of the present invention can be carried out by the method disclosed in the document "Huang P Q, wang Y, wang K, et al. A level Optimization Approach for Joint office Decision and Resource Allocation in Cooperative Mobile Edge Computing [ J ]. IEEE Transactions on Cybernetics,2020,50 (10): 4228-4241".
The ant pheromone is a substance left on a path traveled by an ant in the ant colony algorithm when the ant searches for a food source, the substance has volatility and slowly volatilizes along with the passage of time, and the more pheromones, the more food quantity obtained by searching for the food source through the path is represented.
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, 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 some, but not all embodiments of the present invention. The components of the embodiments of the present invention generally described and illustrated in the figures herein may be arranged and designed in a wide variety of different configurations. Thus, the following detailed description of the embodiments of the present invention, as presented in the figures, is not intended to limit the scope of the invention, as claimed, but is merely representative of selected embodiments of the invention. 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.
The total energy consumption of the vehicle in the following embodiments refers to the sum of the energy consumption of task transmission and the energy consumption of calculation of the vehicle, and is implemented by adopting the method disclosed in the documents "Huang P Q, wang Y, wang K, et al. A level Optimization Approach for Joint routing and Resource Allocation in collaborative Mobile Edge Computing [ J ]. IEEE Transactions on Cybernetics,2020,50 (10): 4228-4241".
Example 1:
taking an MEC server, the number of vehicles and the corresponding number of tasks shown in FIG. 1 as an example, the embodiment adopts the method (TJOA) of the invention to unload a plurality of tasks to the MEC server and a corresponding number of vehicles, and the number of channels is half of the corresponding number of vehicles; in the method, ρ =0.1 and ∈ =20, and initially, the pheromone of each channel is 1.
The results of this example are shown in fig. 1, which shows the convergence comparison of the present invention at different vehicle numbers, and as the number of vehicles increases, the present invention requires more iterations to converge, but as can be seen from fig. 1, even if the number of vehicles reaches 200, the present invention can converge before 50 iterations.
Example 2:
in the embodiment, one MEC server, 50 vehicles and 50 tasks are taken as an example, the method is adopted to unload a plurality of tasks to the MEC server and 50 vehicles, and the number of channels is 25; in the method, ρ =0.1 and ∈ =20, and initially, the pheromone of each channel is 1.
Meanwhile, the same task is unloaded in 51 nodes by adopting the existing ACS-WMO, and the unloading node is an MEC server and a corresponding number of vehicles.
The result of this embodiment is shown in fig. 2, which shows that ACS-WMO gradually converges after 70 iterations, but TJOA can find a better solution and then quickly converges at the beginning, and can converge to an optimal solution before 20 generations, because the TJOA algorithm uses BWAS to solve the computation offload strategy, BWAS increases global pheromone updates to the optimal worst ants in each loop, further weakens the amount of information on the worst path, increases pheromone difference between the optimal path and the worst path, facilitates quick finding of the optimal solution, and thus effectively improves the algorithm convergence speed.
Example 3:
in this embodiment, taking an MEC server, the number of vehicles and the corresponding number of tasks shown in fig. 3 as an example, the method of the present invention is adopted to unload a plurality of tasks to the MEC server and a corresponding number of vehicles, and the number of channels is half the value of the corresponding number of vehicles; in the method, ρ =0.1 and ∈ =20, and initially, the pheromone of each channel is 1.
And simultaneously, unloading the number of vehicles and corresponding tasks shown in the figure 3 by adopting an exhaustive algorithm, wherein the unloading node is an MEC server and a corresponding number of vehicles at the same time.
The results of this example are shown in FIG. 3, which is a graph comparing the method of the present invention with an exhaustive algorithm. The exhaustive algorithm obtains the global optimal solution of the optimization problem by traversing and comparing all feasible solutions, but the complexity of the algorithm increases exponentially along with the increase of the number of vehicles, so that the performance of the algorithm is compared when the number of vehicles is small. Fig. 3 shows that the solution obtained by the present invention has a very small difference from the global optimal solution obtained by the exhaustive method with a small number of vehicles.
Example 4:
in this embodiment, taking an MEC server, the number of vehicles shown in fig. 4, and the corresponding number of tasks as an example, the method of the present invention is adopted to offload a plurality of tasks to the MEC server and a corresponding number of vehicles, and the number of channels is half the value of the corresponding number of vehicles; in the method, ρ =0.1 and ∈ =20, and initially, the pheromone of each channel is 1.
And simultaneously unloading by respectively adopting the number of vehicles and corresponding tasks shown in an LEA, MEA, LMEA and ROA graph 4, wherein the unloading nodes are an MEC server and a corresponding number of vehicles. Wherein:
LEA all Local Execution Algorithms (LEA): tasks generated by the vehicle are not unloaded and are executed locally;
MEA: the MEC server executes an Algorithm (MEC Execution Algorithm, MEA): performing, by the MEC server, all vehicle-generated computing tasks;
LMEA: local or MEC Server Execution (Local or MEC Execution Algorithm, LMEA): the vehicle-mounted task is executed on a local or MEC server;
ROA: random off-loading Algorithm (ROA): the task random decision is executed on the local, cooperating vehicles, MEC server.
The results of this example are shown in FIG. 4, which shows the invention compared to the task completion numbers for LEA, MEA, LMEA, ROA for different vehicle numbers. It can be seen from the figure that when the number of vehicles is small, the invention can completely execute all tasks, and as the number of vehicles increases, the tasks are gradually difficult to complete completely, but compared with other four algorithms, the invention always has the highest task completion number; for the LEA, because the calculation capability of the vehicle-mounted terminal is limited, even when the number of vehicles is small, the requirement of a large amount of calculation resources required by calculation tasks is difficult to meet, the tasks cannot be completely executed, but as the number of vehicles is increased, the number of task completion of the LEA is correspondingly increased because the LEA is not limited by user interference; for the MEA, although the edge server can provide more computing resources than the vehicle, it has a lower task completion number due to the limited bandwidth and the limited computing resources of the edge server, and the limited number of tasks it can accept; the LMEA and ROA provide more choices for the execution of the computing task, and the coordinated vehicle unloading or MEC server unloading can be randomly selected, so that the task completion number is higher than that of the LEA and the MEA.
In summary, the invention provides a scheme for V2X computation offload and resource allocation based on MEC, which can effectively utilize the computation capability at the edge side, and the TJOA algorithm in the invention has better optimization capability and convergence, and can achieve a higher computation task completion rate compared with other algorithms under the same conditions.
The above-mentioned contents are only for illustrating the technical idea of the present invention, and the protection scope of the present invention is not limited thereby, and any modification made on the basis of the technical idea of the present invention falls within the protection scope of the claims of the present invention.

Claims (3)

1. A method for obtaining unloading decisions based on vehicle-mounted application tasks is used for obtaining decisions of N tasks which are transmitted and unloaded to N +1 nodes through L channels, wherein the N +1 nodes are composed of N vehicles and an MEC server, the N tasks are generated by the N vehicles, and each vehicle generates one task; any task
Figure FDA0003934069900000011
Wherein, N is more than or equal to 2,
Figure FDA0003934069900000012
i＝1,2,3,...,N，D i is the amount of data of any task, C i The computing resources required for any of these tasks, T i Deadline For the maximum delay tolerance of any task, T i stay The duration of any task in a cell formed by N +1 nodes;
the method is characterized by comprising the following steps:
step 1, respectively generating a feasible candidate mode set of each task, any task U i The feasible candidate mode set comprises a plurality of mode combinations, each mode combination is composed of a channel and a node, any mode combination satisfies the condition of the formula (1), and any mode combination is composed of a channel k and a node j;
Figure FDA0003934069900000013
in formula (1):
Figure FDA0003934069900000014
is a task U i Minimum computational resources required to offload to node j through channel k, j =0,1,2, 3.., N, j =0 represents the MEC server;
Figure FDA0003934069900000015
j = i denotes a task U i Unloading the local vehicle; r is ijk Is a task U i A data transmission rate offloaded to node j through channel k;
Figure FDA0003934069900000016
the maximum computing resource which can be provided by the node j;
step 2, respectively adopting an ant colony algorithm to obtain an unloading decision of each task in a corresponding feasible candidate mode set; the offload decisions for all tasks constitute the overall offload decision x, x = { x = { ijk }，x ijk For task U i Decision to offload to node j via channel k, if task U i Unloaded on node j via channel k, then x ijk =1, otherwise x ijk ＝0，i＝1,2,3,.A plurality of channels, nj =0,1,2,3, ·, N, k =1,2,3, ·, L; initially, taking 0.5-1.5% of all pheromones;
step 3, after the step 2 is executed again, the step 4 is executed;
step 4, adopting formula (2) to respectively evaluate the fitness of all the overall unloading decisions obtained in the current iteration process and the previous iteration process, wherein F 1 The overall distribution decision corresponding to the maximum value is an overall optimal solution, F 1 The overall unloading decision corresponding to the minimum value is an overall worst solution;
Figure FDA0003934069900000021
x in the formula (2) ijk Elements in the evaluated overall offloading decision;
if there are a plurality of F's of the same size 1 Maximum or multiple of equal-sized F 1 And (3) evaluating the fitness of all overall distribution decisions obtained in the current iteration process and the current iteration process respectively by adopting the formula (3), wherein F 2 The overall distribution decision corresponding to the minimum value is an overall optimal solution, F 2 The overall distribution decision corresponding to the maximum value is an overall worst solution;
Figure FDA0003934069900000022
in formula (3): x is a radical of a fluorine atom ijk As an element in an evaluated global offload decision;
Figure FDA0003934069900000023
for task U i Calculating the generated energy consumption at the node j;
Figure FDA0003934069900000024
for task U i Offloading the resulting transmission energy to node j via channel k;
step 5, updating ant pheromones on each task unloading path in the current overall optimal solution according to the formula (4); updating ant pheromones on each task unloading path in the current overall worst solution according to the formula (5):
Figure FDA0003934069900000031
Figure FDA0003934069900000032
in formulae (4) and (5):
Figure FDA0003934069900000038
representing a task U in the current global optimal solution i Unloading to the node j through k;
rho represents the volatilization speed of the ant pheromone, and the value range is 0.1-0.99;
Δτ ijk for the increment of the ant pheromone, the quantity of the ant pheromone is increased,
Figure FDA0003934069900000033
Figure FDA0003934069900000034
task U for representing and selecting current overall optimal solution i Energy consumption generated by unloading is transmitted to the node j through the channel k;
τ′ ijk the ant pheromone is the ant pheromone before updating;
τ ijk the updated ant pheromone;
Figure FDA0003934069900000035
represents the task U when selecting the current overall worst solution i Energy consumption generated by unloading is transmitted to a computing node j through a channel k;
epsilon is the worst ant pheromone updating parameter, epsilon is more than 0;
and 6, taking the pheromone updated in the step 5, and iteratively executing the steps 2, 4 and 5 until the overall optimal solutions obtained by the previous iteration and the next iteration are the same, wherein the overall optimal solution obtained at the moment is the overall unloading decision of all tasks.
2. A vehicle-mounted application task unloading method is characterized in that the method in claim 1 is adopted to obtain an overall unloading decision, then each task is unloaded to a corresponding node through a corresponding channel according to the overall unloading decision, and unloading resources of any task
Figure FDA0003934069900000036
The allocation satisfies the condition of equation (6):
Figure FDA0003934069900000037
3. a transportation system comprising a plurality of cells, each cell comprising a task offload system and an MEC server; the task offloading system is configured to collect information of a plurality of vehicles and tasks passing through the local cell, and offload the plurality of tasks to each node of the local cell by using the method according to claim 3.
CN202211398518.6A 2022-11-09 2022-11-09 Vehicle-mounted application task unloading decision obtaining method, unloading method and traffic system Pending CN115767636A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202211398518.6A CN115767636A (en) 2022-11-09 2022-11-09 Vehicle-mounted application task unloading decision obtaining method, unloading method and traffic system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202211398518.6A CN115767636A (en) 2022-11-09 2022-11-09 Vehicle-mounted application task unloading decision obtaining method, unloading method and traffic system

Publications (1)

Publication Number Publication Date
CN115767636A true CN115767636A (en) 2023-03-07

Family

ID=85368533

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202211398518.6A Pending CN115767636A (en) 2022-11-09 2022-11-09 Vehicle-mounted application task unloading decision obtaining method, unloading method and traffic system

Country Status (1)

Country Link
CN (1) CN115767636A (en)

Similar Documents

Publication Publication Date Title
Hu et al. Twin-timescale artificial intelligence aided mobility-aware edge caching and computing in vehicular networks
CN113296845B (en) Multi-cell task unloading algorithm based on deep reinforcement learning in edge computing environment
CN110557732B (en) Vehicle edge computing network task unloading load balancing system and balancing method
Wu et al. Load balance guaranteed vehicle-to-vehicle computation offloading for min-max fairness in VANETs
CN113727306B (en) Decoupling C-V2X network slicing method based on deep reinforcement learning
Sethi et al. FedDOVe: A Federated Deep Q-learning-based Offloading for Vehicular fog computing
CN115134242B (en) Vehicle-mounted computing task unloading method based on deep reinforcement learning strategy
CN113542376A (en) Task unloading method based on energy consumption and time delay weighting
CN116566838A (en) Internet of vehicles task unloading and content caching method with cooperative blockchain and edge calculation
CN113992677A (en) MEC calculation unloading method for delay and energy consumption joint optimization
Qin et al. User‐Edge Collaborative Resource Allocation and Offloading Strategy in Edge Computing
He et al. Computation offloading and resource allocation based on DT-MEC-assisted federated learning framework
CN116185523A (en) Task unloading and deployment method
Chen et al. An intelligent task offloading algorithm (iTOA) for UAV network
Zhao et al. Adaptive multi-UAV trajectory planning leveraging digital twin technology for urban IIoT applications
Li et al. Collaborative optimization of edge-cloud computation offloading in internet of vehicles
CN117156493A (en) QoE-based vehicle networking task unloading and resource allocation method
Selvamanju et al. Deep learning based mobile traffic flow prediction model in 5G cellular networks
Hossain et al. Edge orchestration based computation peer offloading in MEC-enabled networks: a fuzzy logic approach
CN115767636A (en) Vehicle-mounted application task unloading decision obtaining method, unloading method and traffic system
Zheng et al. Cooperative‐routing mechanism based on node classification and task allocation for opportunistic social networks
Alghamdi et al. Optimized Contextual Data Offloading in Mobile Edge Computing
Li et al. Cooperative computation offloading combined with data compression in mobile edge computing system
CN112203309A (en) Joint task unloading and caching method based on server cooperation
CN115051999B (en) Energy consumption optimal task unloading method, device and system based on cloud edge cooperation

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