CN115964178A - Internet of vehicles user computing task scheduling method and device and edge service network - Google Patents
Internet of vehicles user computing task scheduling method and device and edge service network Download PDFInfo
- Publication number
- CN115964178A CN115964178A CN202310027977.1A CN202310027977A CN115964178A CN 115964178 A CN115964178 A CN 115964178A CN 202310027977 A CN202310027977 A CN 202310027977A CN 115964178 A CN115964178 A CN 115964178A
- Authority
- CN
- China
- Prior art keywords
- rsu
- task
- vehicle
- max
- calculation
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 32
- 238000004364 calculation method Methods 0.000 claims abstract description 106
- 238000005265 energy consumption Methods 0.000 claims abstract description 30
- 230000005540 biological transmission Effects 0.000 claims description 38
- 230000006855 networking Effects 0.000 claims description 20
- 238000012423 maintenance Methods 0.000 claims description 6
- 230000003595 spectral effect Effects 0.000 claims description 4
- FFBHFFJDDLITSX-UHFFFAOYSA-N benzyl N-[2-hydroxy-4-(3-oxomorpholin-4-yl)phenyl]carbamate Chemical compound OC1=C(NC(=O)OCC2=CC=CC=C2)C=CC(=C1)N1CCOCC1=O FFBHFFJDDLITSX-UHFFFAOYSA-N 0.000 claims description 3
- 230000002068 genetic effect Effects 0.000 claims description 3
- 238000012163 sequencing technique Methods 0.000 claims description 3
- 239000000126 substance Substances 0.000 claims description 3
- 238000012545 processing Methods 0.000 abstract description 12
- 230000008569 process Effects 0.000 abstract description 3
- 230000008447 perception Effects 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 230000033001 locomotion Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012946 outsourcing Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
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
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to the technical field of intelligent transportation, in particular to a method and a device for scheduling calculation tasks of internet of vehicles users and an edge service network. According to the method for scheduling the tasks calculated by the users in the Internet of vehicles, disclosed by the invention, in the cost accounting process, the execution time delay and the execution energy consumption after the tasks are unloaded to each device are simultaneously considered, so that the scheduling strategy is determined, the tasks are unloaded to the device with the lowest execution cost for task processing, and in addition, the game cost is increased in the local execution cost calculation by considering the game relation between the users and an operator, and the actual requirements are better met.
Description
Technical Field
The invention relates to the technical field of intelligent transportation, in particular to a method and a device for scheduling calculation tasks of internet of vehicles users and an edge service network.
Background
The internet of vehicles uses a running vehicle as an information perception object, realizes network connection between the vehicle and other terminals by means of a new generation of information communication technology, improves the overall intelligent driving level of the vehicle, provides safe, comfortable, intelligent and efficient driving feeling and traffic service for users, improves the traffic operation efficiency, and improves the intelligent level of social traffic service. The vehicle networking is limited by the limitations of the computing power and the communication range of the vehicle-mounted equipment, and other computing platforms are needed to assist in processing computing data in the vehicle networking. The traditional data processing of the internet of vehicles is basically carried out with the help of a cloud computing platform, and the cloud computing platform is used for centrally processing data in a remote data center. Due to the development of the internet of things and the rapid increase of the data collection amount of terminal equipment, the requirement for processing multi-source heterogeneous data at large scale edges cannot be met under the linearly increased computing capability of centralized cloud computing, the long-distance transmission of data between a user and a cloud computing platform causes high network delay and computing resource waste, and the protection of data security and privacy in cloud computing faces great challenges in a remote transmission and outsourcing mechanism.
In a cloud computing platform, an end user usually acts as a data consumer, but in the internet of vehicles, the user is both a data consumer and a data producer. The edge computing is to deploy a server with computing and storage capabilities at the edge of a network to perform service processing on an edge terminal, and compared with cloud computing, the edge computing is closer to a physical terminal. The edge computing can complete a large number of business processing processes in a local edge layer, and uploading of a cloud end is not needed, so that the service efficiency of the whole system is improved.
According to the estimates of Internet Business Solution Group (IBSG) and cisco Global Cloud Index (GCI), 5400 thousands of unmanned vehicles were available globally by the year 2035, but only unmanned vehicles could not be used to achieve automatic driving at the L4 or L5 level. Unmanned vehicles need vehicle networking technology (V2X) to realize comprehensive perception of vehicle and road traffic data. More information is obtained than the information of the internal and external sensors of the bicycle, and the perception of the environment in non-line-of-sight is enhanced. For example, real-time data of road driving is acquired through V2X in severe weather, so that the road condition can be intelligently predicted, and accidents are avoided.
The Internet of vehicles has the characteristics of high-speed movement of nodes, dynamic change of a topological structure, coexistence of heterogeneous nodes, lack of important information relay and the like. When designing the service network of the internet of vehicles, the bandwidth of the edge nodes, the transmission path loss, the topological structures of the vehicles and the RSUs, the motion states of the vehicles, and other influences need to be considered. Meanwhile, user vehicles in the internet of vehicles are constantly generating a large amount of data and data computing tasks. The limited computing power and storage capacity of the existing vehicle-mounted equipment cannot meet the requirements of large amount of computation and low time delay, so that the introduction of mobile edge computation into the internet of vehicles is an effective method for solving the problems. In the car networking based on the mobile edge computing, the computing power of a part of core network is moved to the edge of the car networking, and an edge server has strong computing and storing power and is usually equipped on a Road Side Unit (RSU), so that a set of edge service network capable of meeting the requirement of user computing task scheduling is constructed. The computing task may be selected to be performed locally at the vehicle or off-loaded to the nearest edge server for computation based on actual environmental conditions and constraints.
The premise for meeting the scheme is that the calculation task is moved in the Internet of vehicles, which is the concept of task unloading. The core idea of edge computing is to move the corresponding functions of a computing platform from a network core side to a network access side to provide short-distance services for a user, which is also a basic framework of edge services. And after the calculation is finished, returning the task result data to the user, and finishing one-time task unloading.
Therefore, an edge service network is constructed based on the service architecture of the edge computing, and a computing platform is moved from a core network to the network edge, so that the service efficiency of users in the Internet of vehicles can be greatly improved, meanwhile, the time delay of the users is reduced, the privacy of the users is protected, the benefit of the system is improved, and the pressure of the core network is relieved.
In the existing vehicle networking user calculation task scheduling method based on edge calculation, an unloading strategy is selected to schedule a user task and execute the task by calculating the execution cost of the task unloading on different devices, and task result data is returned to a user, so that the time delay of task unloading of an edge calculation network is reduced by the work of part of researchers, however, the influence of energy consumption on efficiency is not considered; the other part of research only considers the energy consumption factor, weakens the influence of time delay and takes the execution cost into account incompletely; in the subjective level of users, calculation of tasks is expected to be processed by an edge server of an operator, users and vehicles of the users only receive results without calling and occupying resources of the vehicles, but in an edge service network, in order to maximize utilization of the calculation resources, a part of tasks are returned to the vehicles of the users for local calculation processing, so that the trust degree of the users on the operator is reduced to a certain extent, and therefore, the existing scheduling method needs to be optimized.
Disclosure of Invention
Therefore, the technical problem to be solved by the invention is to overcome the problems that the execution cost of the scheduling method in the prior art is not fully calculated and the user trust degree is not considered.
In order to solve the technical problem, the invention provides a method for scheduling a calculation task of a vehicle networking user, which comprises the following steps:
when the RSU receives the calculation task of the independent vehicle, the RSU carries out cost accounting, wherein the cost accounting comprises the following steps: respectively calculating the execution cost of a task in a local, RSU and RSU-corresponding high-performance RSU-max, wherein the RSU-max covers a plurality of RSUs, the execution cost comprises task execution time delay and task execution energy consumption, and game cost is increased in the local task execution cost;
determining a scheduling strategy of an independent vehicle calculation task according to a cost accounting result, wherein the scheduling strategy of the independent vehicle calculation task comprises the following steps: when the execution cost of the calculation task in the RSU is the lowest, the RSU is used for executing the calculation task to obtain a task result, the task result is transmitted to the local of the user through the RSU, when the execution cost of the calculation task in the RSU is not the lowest, whether the calculated amount exceeds a threshold value is judged, if yes, the task is unloaded to the RSU-max for execution, the obtained task result is returned to the local of the user, and otherwise, the task result is directly returned to the local of the user for execution.
Preferably, when the RSU receives the set of calculation tasks of the vehicle queue, the RSU forms a pair of calculation tasks two by two, and performs the cost accounting on two calculation tasks of each pair of calculation tasks in sequence;
determining a scheduling strategy of the vehicle queue calculation task according to the cost accounting result, wherein the scheduling strategy of the vehicle queue calculation task comprises the following steps: when the execution cost of the calculation task in the RSU is the lowest, the RSU is used for executing the calculation task to obtain a task result, when the execution cost of the calculation task in the RSU is not the lowest, whether the calculated amount exceeds a threshold value is judged, if yes, the task is unloaded to the RSU-max for execution, the obtained task result is returned to the RSU, otherwise, the task result is directly returned to the user for local execution to obtain the task result, and finally, the task result in the RSU is sent to the user locally in parallel.
Preferably, if the vehicle has moved out of the coverage of the RSU after the RSU is used to execute the calculation task to obtain the task result, the task result is returned to the local user from the RSU-max corresponding to the RSU.
Preferably, the parallel sending of the task results of the vehicle queue computing tasks to the user locally through the RSU comprises:
performing bandwidth allocation on a task result in the RSU according to a genetic algorithm;
and transmitting the task result to each member vehicle according to the bandwidth allocation result, and transmitting the shared part of the task result to the leader vehicle by each member vehicle.
Preferably, the transmitting of the task results of the independent vehicle computing tasks to the user locally by the RSU comprises:
setting an initial time, adding one to the priority of all task results in the current transmission queue of the RSU every time a unit time slice length passes, sequencing all task results in the current transmission queue according to the priority, and preferentially transmitting the task result with the highest priority, wherein the priority is judged according to the residual transmission time of the task result, namely the residual time of the independent vehicle running out of the current RSU, and the shorter the residual time is, the higher the priority of the task result is;
and when the RSU executes the calculation task of a certain independent vehicle to obtain a task result, adding the task result into a transmission queue of the RSU.
Preferably, the execution cost of the tasks at the local, RSU and high performance RSU-max are calculated separately, and the formula is:
wherein the content of the first and second substances,and &>Represents the cost of execution of task i locally, RSU and RSU-max, respectively, <' > based on>And &>Represents the execution time of task i locally in the vehicle, RSU and RSU-max respectively, <' > in conjunction with>And &>Respectively, the execution energy consumption of the task i in the vehicle local, RSU and RSU-max, W game Is the game constant, W max Is the maintenance cost factor, W, of the RSU-max performing a calculation task mid Is the maintenance cost coefficient of RSU executing one calculation task, α is the time preference coefficient, and β is the energy consumption preference coefficient.
Preferably, when the cost accounting of the independent vehicle is carried out, the calculation formula of the execution time of the task i in the vehicle local, the RSU and the RSU-max is as follows:
wherein, t i1 、t i2 、t i3 Calculation times, C, for task i local to the vehicle, RSU and RSU-max, respectively i Is the amount of computation of task i, l is the number of cycles for the CPU to process 1 byte of data, f i1 、f i2 、f i3 The computing resources of task i are allocated to vehicle local, RSU and RSU-max respectively,wait time before calculation for task i, t i,x For the transmission time, t, of task results between RSU and user i,j For the transmission time of the result of task i between RSU-max and user, S i,x For the RSU transmission rate, S, of task result data i,j For RSU-max to the transmission rate of task result data, D i For the size of the task result data volume, B i,x Bandwidth, SNR allocated to task i for RSU i,x Is the signal-to-noise ratio, P, of the user's vehicle and the RSU i Is the transmit power, H, of the RSU i Is the channel gain between the RSU and the user's vehicle, δ is the ambient signal interference, N is the noise spectral density, B i,j Bandwidth, SNR, allocated for task i for RSU-max i,j Is the signal-to-noise ratio, P, of the user vehicle and RSU-max j Is the transmit power of RSU-max, H j Is the channel gain between RSU-max and the user vehicle;
when the cost accounting of the independent vehicle is carried out, the calculation formula of the execution energy consumption of the task i in the vehicle local area, the RSU and the RSU-max is as follows:
where θ is the coefficient of energy consumption for transmitting data, η lo 、η r 、η max The energy consumption coefficient per CPU cycle of the CPU deployed locally in the vehicle, RSU and RSU-max respectively.
Preferably, when the cost accounting of the vehicle queue is carried out, the calculation formula of the execution time of the task i in the vehicle local, the RSU and the RSU-max is as follows:
t i,j =D i /S i,j
wherein, t i,j For the transmission time of the task result between RSU-max and RSU, S i,j Is the transmission rate between RSU-max and RSU, B i,j Bandwidth, SNR, allocated for task i for RSU-max i,j Is the signal-to-noise ratio, P, of RSU and RSU-max j Is the transmit power of RSU-max, H j Is the channel gain between RSU-max and RSU;
when the cost accounting of the vehicle queue is carried out, the calculation formula of the execution energy consumption of the task i in the vehicle local area, the RSU and the RSU-max is as follows:
the invention also provides a device for scheduling the calculation tasks of the users in the Internet of vehicles, which is used for realizing the method for scheduling the calculation tasks of the users in the Internet of vehicles
The invention also provides an edge service network, comprising:
the system comprises a plurality of general-performance Road Side Units (RSUs), wherein each general-performance road side unit comprises the Internet of vehicles user computing task scheduling device;
and each high-performance road side unit is connected with a plurality of common-performance road side units RSU.
Compared with the prior art, the technical scheme of the invention has the following advantages:
according to the method for scheduling the tasks calculated by the users in the Internet of vehicles, disclosed by the invention, in the cost accounting process, the execution time delay and the execution energy consumption after the tasks are unloaded to each device are simultaneously considered, so that the scheduling strategy is determined, the tasks are unloaded to the device with the lowest execution cost for task processing, and in addition, the game cost is increased in the local execution cost calculation by considering the game relation between the users and an operator, and the actual requirements are better met.
Drawings
In order that the manner in which the present invention is more fully understood, a more particular description of the invention will be rendered by reference to specific embodiments thereof which are illustrated in the appended drawings, wherein:
FIG. 1 is a flowchart of an implementation of the Internet of vehicles user computing task scheduling method of the present invention;
FIG. 2 is a flowchart illustrating an implementation of a method for calculating task scheduling for a vehicle networking user according to an embodiment;
FIG. 3 is a schematic diagram of three states after a RSU receives a computing task request;
FIG. 4 is a schematic diagram of a RSU receiving a vehicle queue calculation task request;
fig. 5 is a flowchart of an implementation of the scheduling method after the RSU receives a vehicle queue computing task request.
Detailed Description
The core of the invention is to provide a method and a device for scheduling calculation tasks of internet of vehicles users and an edge service network, which have more comprehensive cost accounting consideration and meet the actual requirements.
In order that those skilled in the art will better understand the disclosure, the invention will be described in further detail with reference to the accompanying drawings and specific embodiments. It is to be understood that the described embodiments are merely exemplary of the invention, and not restrictive of the full scope of the invention. All other embodiments, which can be obtained by a person skilled in the art without making any creative effort based on the embodiments in the present invention, belong to the protection scope of the present invention.
Referring to fig. 1 and fig. 2, fig. 1 is a flowchart illustrating an implementation of a method for scheduling a user calculation task in a car networking system according to the present invention, and fig. 2 is a flowchart illustrating an implementation of a method for scheduling a user calculation task in a car networking system according to an embodiment of the present invention; the specific operation steps are as follows:
s101, when the RSU receives a calculation task of an independent vehicle, performing cost accounting on the RSU, wherein the cost accounting comprises the following steps: respectively calculating the execution cost of a task in a local, RSU and RSU-corresponding high-performance RSU-max, wherein the RSU-max covers a plurality of RSUs, the execution cost comprises task execution time delay and task execution energy consumption, and game cost is increased in the local task execution cost;
the arrival of the independent vehicle random user tasks obeys a poisson distribution:
where λ is the average incidence of computing task requests per unit time, p i (t) is the probability that q computational tasks are generated in time t. During the poisson distribution, the generation of tasks is only related to the time interval.
The execution cost of the tasks in the local, RSU and RSU-max is respectively calculated, and the formula is expressed as follows:
wherein the content of the first and second substances,and &>Represents the cost of execution of task i locally, RSU and RSU-max, respectively, and->And &>Represents the execution time of task i locally in the vehicle, RSU and RSU-max respectively, <' > in conjunction with>And &>Respectively, the execution energy consumption of the task i in the vehicle local, RSU and RSU-max, W game Is the game constant, W max Is the maintenance cost factor, W, of the RSU-max performing a computational task mid Is the maintenance cost coefficient of RSU executing one calculation task, α is the time preference coefficient, and β is the energy consumption preference coefficient.
When the cost accounting of the independent vehicle is carried out, the calculation formula of the execution time of the task i in the vehicle local area, the RSU and the RSU-max is as follows:
wherein, t i1 、t i2 、t i3 Calculation times C for task i locally in the vehicle, RSU and RSU-max, respectively i Is the amount of computation of task i, l is the number of cycles for the CPU to process 1 byte of data, f i1 、f i2 、f i3 The computing resources of task i are allocated to vehicle local, RSU and RSU-max respectively,wait time for task i before calculation, t i,x For the transmission time, t, of task results between RSU and user i,j For the transmission time of the result of task i between RSU-max and user, S i,x For the RSU transmission rate, S, of task result data i,j For RSU-max to the transmission rate of task result data, D i For the size of the task result data volume, B i,x Bandwidth, SNR, allocated to task i for RSU i,x Is the signal-to-noise ratio, P, of the user's vehicle and the RSU i Is the transmit power of the RSU, H i Is the channel gain between the RSU and the user's vehicle, δ is the ambient signal interference, N is the noise spectral density, P i Is the transmit power, H, of the RSU i Is the channel gain between the RSU and the user's vehicle, δ is the external signal interference, N is the noise spectral density, B i,j Bandwidth, SNR, allocated for task i for RSU-max i,j Is the signal-to-noise ratio, P, of the user vehicle and RSU-max j Is the transmit power of RSU-max, H j Is the channel gain between RSU-max and the user vehicle;
when the cost accounting of the independent vehicle is carried out, the calculation formula of the execution energy consumption of the task i in the vehicle local area, the RSU and the RSU-max is as follows:
where θ is the energy consumption coefficient of the transmitted data, η lo 、η r 、η max The energy consumption coefficient per CPU cycle of the CPU deployed locally in the vehicle, RSU and RSU-max respectively.
S102, determining a scheduling strategy of the independent vehicle calculation task according to the cost accounting result, wherein the scheduling strategy of the independent vehicle calculation task comprises the following steps: when the execution cost of the calculation task in the RSU is the lowest, the RSU is used for executing the calculation task to obtain a task result, the task result is transmitted to the local of the user through the RSU, when the execution cost of the calculation task in the RSU is not the lowest, whether the calculated amount exceeds a threshold value is judged, if yes, the task is unloaded to the RSU-max for execution, the obtained task result is returned to the local of the user, and otherwise, the task result is directly returned to the local of the user for execution.
And if the vehicle runs out of the coverage range of the RSU after the RSU is used for executing the calculation task to obtain the task result, returning the task result to the local part of the user from the RSU-max corresponding to the RSU.
The edge service network is established in a relatively independent area, the area comprises a plurality of automobiles and a plurality of RSUs with different computing capabilities, and a computing task request is generated when a user automobile runs, and a large amount of computing data corresponds to the computing task request. After receiving a task request, a RSU performs task offloading according to the size of the task computation amount and the queue congestion condition of the task queue in the current RSU, and the RSU is in the three states shown in fig. 3:
state 1: the RSU verifies the calculation cost, the calculation is carried out at the RSU to be the optimal solution, and the RSU executes the current calculation task and returns the task result data to the user;
state 2: the RSU verifies the calculation cost, the calculation is not the optimal solution in the RSU, and the task is unloaded to the RSU-max or the vehicle is executed locally;
state 3: and the user vehicle drives out of the coverage range of the RSU, and the task result data is returned to the user from the RSU-max of the current area.
The transmitting of the task results of the independent vehicle computing tasks to the user local through the RSU comprises:
the invention selects and uses a dynamic priority preemptive algorithm based on time slice scheduling to transmit task result data; setting an initial time, adding one to the priority of all task results in the current transmission queue of the RSU every time a unit time slice length passes, sequencing all task results in the current transmission queue according to the priority, and preferentially transmitting the task result with the highest priority, wherein the priority is judged according to the residual transmission time of the task result, namely the residual time of the independent vehicle running out of the current RSU, and the shorter the residual time is, the higher the priority of the task result is;
and when the RSU executes the calculation task of a certain independent vehicle to obtain a task result, adding the task result into a transmission queue of the RSU.
The edge service network of the vehicle user is an edge server deployed by an operator at a Road side unit (Road side Un it, RSU for short) at the edge of a Road, so that the edge service network for providing task calculation and task scheduling for the vehicle user is established. The invention designs that RSUs with different performances are deployed in an edge service network, wherein the RSU with common performance is defined as RSU-mi (RSU for short), and the RSU with high performance and wide coverage area is defined as RSU-max. One RSU-max covers a plurality of RSUs, so that stable transmission of vehicle tasks of users and processing of ultra-large computing tasks are guaranteed, and local resources of the vehicles are brought into calling. The system model is modeled from two aspects of time delay and energy consumption, the time delay comprises data transmission time delay and calculation time delay, the energy consumption mainly comprises energy consumption of data transmission and energy consumption of RSU calculation data, and in consideration of game relation between users and operators, game cost is increased in local execution cost calculation, and actual requirements are met better. After the request of the computing task is transmitted to the RSU, the RSU carries out cost evaluation on the computing task, and after the cost evaluation is finished, the RSU unloads the task to a proper device for processing the task and endows dynamic priority to the task. And after the calculation is finished, returning the task result data to the user according to the dynamic priority of the task.
As shown in fig. 4 and 5, based on the above embodiments, when the RSU receives the set of calculation tasks of the vehicle queue, the RSU pairs the calculation tasks two by two based on the two threads of the RSU, and the cost accounting is performed on two calculation tasks in each pair of calculation tasks sequentially and simultaneously;
determining a scheduling strategy of a vehicle queue (according with a front vehicle-pilot type topological structure, and a following vehicle can only communicate with a front vehicle and a leading vehicle) calculation task according to a cost accounting result, wherein the scheduling strategy of the vehicle queue calculation task comprises the following steps: when the execution cost of the calculation task in the RSU is the lowest, the RSU is used for executing the calculation task to obtain a task result, when the execution cost of the calculation task in the RSU is not the lowest, whether the calculated amount exceeds a threshold value or not is judged, if yes, the task is unloaded to the RSU-max for execution, the obtained task result is returned to the RSU, otherwise, the task result is directly returned to the user for local execution to obtain the task result, and finally the task result in the RSU is sent to the user locally in parallel.
The parallel sending of the task results of the vehicle queue computing tasks to the user local through the RSU comprises the following steps:
performing bandwidth allocation on a task result in the RSU according to a genetic algorithm;
and transmitting the task result to each member vehicle according to the bandwidth allocation result, and transmitting the shared part of the task result to the leader vehicle by each member vehicle.
In processing concurrent tasks, the RSU needs to allocate bandwidth for the computation task:
in the case of a serving vehicle queue, the individual task times are scaled to the total transmission time T sum Is represented as follows:
where n is the number of compute tasks in a task set.
When the cost accounting of the vehicle queue is carried out, the calculation formula of the execution time of the task i in the vehicle local area, the RSU and the RSU-max is as follows:
t i,j =D i /S i,j
wherein, t i,j For the transmission time of the task result between RSU-max and RSU, S i,j Is the transmission rate between RSU-max and RSU, B i,j Bandwidth, SNR, allocated for task i for RSU-max i,j Is the signal-to-noise ratio, P, of RSU and RSU-max j Is the transmit power of RSU-max, H j Is the channel gain between RSU-max and RSU;
when the cost accounting of the vehicle queue is carried out, the calculation formula of the execution energy consumption of the task i in the vehicle local area, the RSU and the RSU-max is as follows:
the calculation task scheduling method for the internet of vehicles based on edge calculation, provided by the invention, is improved on the basis of a basic edge calculation mode, not only integrates the main aspects of energy consumption, time delay and the like, but also considers the calculation task request of a multi-scene user vehicle to adapt to the practical requirement, designs two service strategies of an independent calculation task and a concurrent calculation task, improves the service efficiency of an operator, unloads the tasks between an edge server and the local, relieves the calculation pressure of a core network, reduces the task transmission time delay, reduces the operation cost of the operator, and improves the service efficiency of the user.
The invention also provides a device for scheduling the calculation tasks of the users in the Internet of vehicles, which is used for realizing the method for scheduling the calculation tasks of the users in the Internet of vehicles.
The invention also provides an edge service network, comprising:
the system comprises a plurality of general-performance Road Side Units (RSUs), wherein each general-performance road side unit comprises the Internet of vehicles user computing task scheduling device;
and each high-performance road side unit is connected with a plurality of common-performance road side units RSU.
It should be understood that the above examples are only for clarity of illustration and are not intended to limit the embodiments. Other variations and modifications will be apparent to persons skilled in the art in light of the above description. This need not be, nor should it be exhaustive of all embodiments. And obvious variations or modifications therefrom are within the scope of the invention.
Claims (10)
1. A vehicle networking user computing task scheduling method is characterized by comprising the following steps:
when the RSU receives the calculation task of the independent vehicle, the RSU carries out cost accounting, wherein the cost accounting comprises the following steps: respectively calculating the execution cost of a task in a local, RSU and RSU-corresponding high-performance RSU-max, wherein the RSU-max covers a plurality of RSUs, the execution cost comprises task execution time delay and task execution energy consumption, and game cost is increased in the local task execution cost;
determining a scheduling strategy of an independent vehicle calculation task according to a cost accounting result, wherein the scheduling strategy of the independent vehicle calculation task comprises the following steps: when the execution cost of the calculation task in the RSU is the lowest, the RSU is used for executing the calculation task to obtain a task result, the task result is transmitted to the local user through the RSU, when the execution cost of the calculation task in the RSU is not the lowest, whether the calculated amount exceeds a threshold value or not is judged, if the calculated amount exceeds the threshold value, the task is unloaded to the RSU-max for execution, the obtained task result is returned to the local user, and otherwise, the task result is directly returned to the local user for execution.
2. The vehicle networking user computing task scheduling method according to claim 1, wherein when a common performance Road Side Unit (RSU) receives a computing task set of a vehicle queue, computing tasks are paired in pairs, and the cost accounting is performed on two computing tasks in each pair of computing tasks in sequence;
determining a scheduling strategy of the vehicle queue calculation task according to the cost accounting result, wherein the scheduling strategy of the vehicle queue calculation task comprises the following steps: when the execution cost of the calculation task in the RSU is the lowest, the RSU is used for executing the calculation task to obtain a task result, when the execution cost of the calculation task in the RSU is not the lowest, whether the calculated amount exceeds a threshold value is judged, if yes, the task is unloaded to the RSU-max for execution, the obtained task result is returned to the RSU, otherwise, the task result is directly returned to the user for local execution to obtain the task result, and finally, the task result in the RSU is sent to the user locally in parallel.
3. The vehicle networking user calculation task scheduling method according to claim 2, wherein if a vehicle has moved out of the coverage of the RSU after the RSU executes a calculation task to obtain a task result, the task result is returned to the user local by the RSU-max corresponding to the RSU.
4. The vehicle networking user computing task scheduling method according to claim 2, wherein the parallel sending of the task results of the vehicle queue computing tasks to the user locally through the RSU comprises:
performing bandwidth allocation on a task result in the RSU according to a genetic algorithm;
and transmitting the task result to each member vehicle according to the bandwidth allocation result, and transmitting the shared part of the task result to the leader vehicle by each member vehicle.
5. The vehicle networking user computing task scheduling method of claim 1, wherein the transmitting task results of the independent vehicle computing tasks to the user locally through the RSU comprises:
setting an initial time, adding one to the priority of all task results in the current transmission queue of the RSU every time a unit time slice length passes, sequencing all task results in the current transmission queue according to the priority, and preferentially transmitting the task result with the highest priority, wherein the priority is judged according to the residual transmission time of the task result, namely the residual time of the independent vehicle running out of the current RSU, and the shorter the residual time is, the higher the priority of the task result is;
and when the RSU executes the calculation task of a certain independent vehicle to obtain a task result, adding the task result into a transmission queue of the RSU.
6. The vehicle networking user computing task scheduling method according to claim 1, wherein the execution cost of the tasks in the local, RSU and high performance RSU-max is calculated respectively, and the formula is as follows:
wherein the content of the first and second substances,and &>Represents the cost of execution of task i locally, RSU and RSU-max, respectively, and->Andrepresents the execution times of task i locally in the vehicle, RSU and RSU-max, respectively, < >>And &>The energy consumption for the execution of task i locally in the vehicle, RSU and RSU-max, W game Is the game constant, W max Is the maintenance cost factor, W, of the RSU-max performing a computational task mid The RSU is used for executing a maintenance cost coefficient of a calculation task, alpha is a time preference coefficient, and beta is an energy consumption preference coefficient.
7. The vehicle networking user computing task scheduling method according to claim 6, wherein when cost accounting of an independent vehicle is performed, the computing formula of the execution time of the task i in the vehicle local, the RSU and the RSU-max is as follows:
wherein, t i1 、t i2 、t i3 Calculation times C for task i locally in the vehicle, RSU and RSU-max, respectively i Is the amount of computation of task i, l is the number of cycles for the CPU to process 1 byte of data, f i1 、f i2 、f i3 The computing resources of task i are allocated to vehicle local, RSU and RSU-max respectively,wait time before calculation for task i, t i,x For the transmission time of task results between RSU and user, t i,j For the transmission time of the result of task i between RSU-max and user, S i,x For the RSU transmission rate, S, of task result data i,j For RSU-max to the transmission rate of task result data, D i For the size of the data volume of the task result, B i,x Bandwidth, SNR, allocated to task i for RSU i,x Is the signal-to-noise ratio, P, of the user's vehicle and the RSU i Is the transmit power of the RSU, H i Is the channel gain between the RSU and the user's vehicle, δ is the external signal interference, N is the noise spectral density, B i,j Bandwidth, SNR, allocated for task i for RSU-max i,j Is the signal-to-noise ratio, P, of the user vehicle and RSU-max j Is the transmit power of RSU-max, H j Is the channel gain between RSU-max and the user vehicle;
when the cost accounting of the independent vehicle is carried out, the calculation formula of the execution energy consumption of the task i in the vehicle local area, the RSU and the RSU-max is as follows:
where θ is the coefficient of energy consumption for transmitting data, η lo 、η r 、η max The energy consumption coefficient per CPU cycle of the CPU deployed locally in the vehicle, RSU and RSU-max, respectively.
8. The vehicle networking user computing task scheduling method according to claim 7, wherein when the cost accounting of the vehicle queue is performed, the computing formula of the execution time of the task i in the vehicle local, the RSU and the RSU-max is as follows:
t i,j =D i /S i,j
wherein, t i,j For the transmission time of the task result between RSU-max and RSU, S i,j Is the transmission rate between RSU-max and RSU, B i,j Bandwidth, SNR, allocated for task i for RSU-max i,j Is the signal-to-noise ratio, P, of RSU and RSU-max j Is the transmit power of RSU-max, H j Is the channel gain between RSU-max and RSU;
when the cost accounting of the vehicle queue is carried out, the calculation formula of the execution energy consumption of the task i in the vehicle local area, the RSU and the RSU-max is as follows:
9. a vehicle networking user computing task scheduling device, for implementing the vehicle networking user computing task scheduling method according to any one of claims 1 to 8.
10. An edge services network, comprising:
a plurality of generic performance RSUs, each generic performance RSU comprising the internet of vehicles user computing task scheduler of claim 9;
and each high-performance road side unit is connected with a plurality of common-performance road side units RSU.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310027977.1A CN115964178B (en) | 2023-01-09 | 2023-01-09 | Internet of vehicles user computing task scheduling method, device and edge service network |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310027977.1A CN115964178B (en) | 2023-01-09 | 2023-01-09 | Internet of vehicles user computing task scheduling method, device and edge service network |
Publications (2)
Publication Number | Publication Date |
---|---|
CN115964178A true CN115964178A (en) | 2023-04-14 |
CN115964178B CN115964178B (en) | 2024-05-31 |
Family
ID=87357961
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202310027977.1A Active CN115964178B (en) | 2023-01-09 | 2023-01-09 | Internet of vehicles user computing task scheduling method, device and edge service network |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN115964178B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116546429A (en) * | 2023-06-06 | 2023-08-04 | 江南大学 | Vehicle selection method and system in federal learning of Internet of vehicles |
Citations (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111182495A (en) * | 2019-12-25 | 2020-05-19 | 大连理工大学 | 5G internet of vehicles partial calculation unloading method |
CN111831427A (en) * | 2020-05-18 | 2020-10-27 | 南京邮电大学 | Distributed inter-vehicle task unloading method based on mobile edge calculation |
CN111935677A (en) * | 2020-08-10 | 2020-11-13 | 无锡太湖学院 | Internet of vehicles V2I mode task unloading method and system |
CN112367640A (en) * | 2020-11-09 | 2021-02-12 | 中科怡海高新技术发展江苏股份公司 | V2V mode multitask unloading method and system based on mobile edge calculation |
CN112511614A (en) * | 2020-11-20 | 2021-03-16 | 福建师范大学 | Intelligent networking vehicle task unloading method under vehicle-mounted edge environment based on reinforcement learning |
WO2021067140A1 (en) * | 2019-10-04 | 2021-04-08 | Intel Corporation | Edge computing technologies for transport layer congestion control and point-of-presence optimizations based on extended in-advance quality of service notifications |
US20210204101A1 (en) * | 2019-12-26 | 2021-07-01 | Industry Academic Cooperation Foundation Of Yeungnam University | Multichannel access control method in overlapped vehicle networks |
CN113132943A (en) * | 2021-04-18 | 2021-07-16 | 中南林业科技大学 | Task unloading scheduling and resource allocation method for vehicle-side cooperation in Internet of vehicles |
CN114615265A (en) * | 2022-03-09 | 2022-06-10 | 浙江工业大学 | Vehicle-mounted task unloading method based on deep reinforcement learning in edge computing environment |
CN114995990A (en) * | 2022-04-18 | 2022-09-02 | 国网河南省电力公司信息通信公司 | Method and device for unloading computing tasks, electronic equipment and computer storage medium |
CN115065683A (en) * | 2022-07-27 | 2022-09-16 | 多伦科技股份有限公司 | Vehicle edge network task allocation and unloading method based on vehicle clustering |
-
2023
- 2023-01-09 CN CN202310027977.1A patent/CN115964178B/en active Active
Patent Citations (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2021067140A1 (en) * | 2019-10-04 | 2021-04-08 | Intel Corporation | Edge computing technologies for transport layer congestion control and point-of-presence optimizations based on extended in-advance quality of service notifications |
CN111182495A (en) * | 2019-12-25 | 2020-05-19 | 大连理工大学 | 5G internet of vehicles partial calculation unloading method |
US20210204101A1 (en) * | 2019-12-26 | 2021-07-01 | Industry Academic Cooperation Foundation Of Yeungnam University | Multichannel access control method in overlapped vehicle networks |
CN111831427A (en) * | 2020-05-18 | 2020-10-27 | 南京邮电大学 | Distributed inter-vehicle task unloading method based on mobile edge calculation |
CN111935677A (en) * | 2020-08-10 | 2020-11-13 | 无锡太湖学院 | Internet of vehicles V2I mode task unloading method and system |
CN112367640A (en) * | 2020-11-09 | 2021-02-12 | 中科怡海高新技术发展江苏股份公司 | V2V mode multitask unloading method and system based on mobile edge calculation |
CN112511614A (en) * | 2020-11-20 | 2021-03-16 | 福建师范大学 | Intelligent networking vehicle task unloading method under vehicle-mounted edge environment based on reinforcement learning |
CN113132943A (en) * | 2021-04-18 | 2021-07-16 | 中南林业科技大学 | Task unloading scheduling and resource allocation method for vehicle-side cooperation in Internet of vehicles |
CN114615265A (en) * | 2022-03-09 | 2022-06-10 | 浙江工业大学 | Vehicle-mounted task unloading method based on deep reinforcement learning in edge computing environment |
CN114995990A (en) * | 2022-04-18 | 2022-09-02 | 国网河南省电力公司信息通信公司 | Method and device for unloading computing tasks, electronic equipment and computer storage medium |
CN115065683A (en) * | 2022-07-27 | 2022-09-16 | 多伦科技股份有限公司 | Vehicle edge network task allocation and unloading method based on vehicle clustering |
Non-Patent Citations (2)
Title |
---|
YUANZHI NI等: "Hybrid RSU Management in Cybertwin-IoV for Temporal and Spatial Service Coverage", 《IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY》, vol. 71, no. 5, 31 May 2022 (2022-05-31) * |
田大新;: "智能车联网***未来展望", 信息通信技术与政策, no. 08, 15 August 2020 (2020-08-15) * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116546429A (en) * | 2023-06-06 | 2023-08-04 | 江南大学 | Vehicle selection method and system in federal learning of Internet of vehicles |
CN116546429B (en) * | 2023-06-06 | 2024-01-16 | 杭州一诺科创信息技术有限公司 | Vehicle selection method and system in federal learning of Internet of vehicles |
Also Published As
Publication number | Publication date |
---|---|
CN115964178B (en) | 2024-05-31 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108541027B (en) | Communication computing resource replacement method based on edge cloud network | |
CN110928658B (en) | Cooperative task migration system and algorithm of vehicle edge cloud cooperative framework | |
CN110109745B (en) | Task collaborative online scheduling method for edge computing environment | |
Qi et al. | Mobile edge computing-assisted admission control in vehicular networks: The convergence of communication and computation | |
CN111475274B (en) | Cloud collaborative multi-task scheduling method and device | |
CN111182495B (en) | 5G internet of vehicles partial calculation unloading method | |
CN113543074B (en) | Joint computing migration and resource allocation method based on vehicle-road cloud cooperation | |
Wang et al. | Dynamic offloading scheduling scheme for MEC-enabled vehicular networks | |
CN112188627B (en) | Dynamic resource allocation strategy based on state prediction | |
CN111107566A (en) | Unloading method based on collaborative content caching in power Internet of things scene | |
CN110149401B (en) | Method and system for optimizing edge calculation task | |
CN112153728B (en) | Optimization method for base station association and module dormancy | |
CN115964178B (en) | Internet of vehicles user computing task scheduling method, device and edge service network | |
CN114928611B (en) | IEEE802.11p protocol-based energy-saving calculation unloading optimization method for Internet of vehicles | |
CN111212108B (en) | Multi-user parallel migration method based on non-orthogonal multiple access and mobile edge computing | |
CN116204315A (en) | Track-based dynamic task unloading method for vehicle resource pool in Internet of vehicles | |
CN108200601A (en) | A kind of multilink selection method based on data priority, link state and traffic current density | |
CN113709249B (en) | Safe balanced unloading method and system for driving assisting service | |
US12010596B2 (en) | Control device configured for and method of determining a data format | |
Hong et al. | An optimal resource allocation mechanism in vehicular MEC systems | |
CN111431950B (en) | Task unloading method and device, mobile terminal, fog node and storage medium | |
CN113613206A (en) | Wireless heterogeneous Internet of vehicles edge unloading scheme based on reinforcement learning | |
CN108833486B (en) | Hybrid dynamic task scheduling method for complex vehicle-mounted fog computing system environment | |
CN115100898B (en) | Collaborative computing task unloading method of urban intelligent parking management system | |
CN115103376A (en) | Dynamic resource slicing method for air-ground integrated Internet of vehicles |
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 |