CN112512018B - Method for dynamically unloading tasks among cooperative vehicles based on mobile edge calculation - Google Patents

Method for dynamically unloading tasks among cooperative vehicles based on mobile edge calculation Download PDF

Info

Publication number
CN112512018B
CN112512018B CN202010727802.8A CN202010727802A CN112512018B CN 112512018 B CN112512018 B CN 112512018B CN 202010727802 A CN202010727802 A CN 202010727802A CN 112512018 B CN112512018 B CN 112512018B
Authority
CN
China
Prior art keywords
vehicle
calculation
task
service
reliability
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
Application number
CN202010727802.8A
Other languages
Chinese (zh)
Other versions
CN112512018A (en
Inventor
田大新
韩旭
段续庭
周建山
郎平
林椿眄
赵元昊
郑坤贤
闫慧文
黄米琪
郝威
龙科军
刘赫
拱印生
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.)
Beihang University
Original Assignee
Beihang 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 Beihang University filed Critical Beihang University
Priority to CN202010727802.8A priority Critical patent/CN112512018B/en
Publication of CN112512018A publication Critical patent/CN112512018A/en
Application granted granted Critical
Publication of CN112512018B publication Critical patent/CN112512018B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • 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]
    • H04W4/46—Services specially adapted for particular environments, situations or purposes for vehicles, e.g. vehicle-to-pedestrians [V2P] for vehicle-to-vehicle communication [V2V]
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W12/00—Security arrangements; Authentication; Protecting privacy or anonymity
    • H04W12/009—Security arrangements; Authentication; Protecting privacy or anonymity specially adapted for networks, e.g. wireless sensor networks, ad-hoc networks, RFID networks or cloud networks
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W12/00—Security arrangements; Authentication; Protecting privacy or anonymity
    • H04W12/02—Protecting privacy or anonymity, e.g. protecting personally identifiable information [PII]
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W24/00—Supervisory, monitoring or testing arrangements
    • H04W24/02—Arrangements for optimising operational condition
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W24/00—Supervisory, monitoring or testing arrangements
    • H04W24/06—Testing, supervising or monitoring using simulated traffic

Abstract

A method for dynamically unloading tasks among cooperative vehicles based on mobile edge calculation is disclosed. The method comprises the following steps: constructing a vehicle cooperative communication reliability evaluation model and a calculation reliability evaluation model of a service vehicle; establishing a virtual queue model based on the service vehicle side computing resources, and integrating the queue length and the privacy perception optimization partition; determining a service request-oriented maximum communication and calculation coupling reliability optimization equation based on the processing time delay constraint of the preset task of the requesting vehicle; task offload decisions are determined based on maximizing coupling reliability. The method considers the communication and the calculation coupling between vehicles adopting cooperative transmission, can divide the calculation tasks of the request vehicles, and realizes the joint processing of the request vehicles and the service vehicles by the mobile edge calculation under the condition that the resources of the service vehicles are limited, so as to improve the utilization rate of the communication and the calculation resources and improve the coupling reliability.

Description

Method for dynamically unloading tasks among cooperative vehicles based on mobile edge calculation
Technical Field
The invention relates to the technical field of wireless communication technology and mobile edge computing technology, in particular to a dynamic task unloading technology between cooperative vehicles based on mobile edge computing.
Background
The Internet of Vehicles (IoV) integrates available computing resources from different fields, constructs a powerful distributed mobile computing system for Vehicles, and makes the most use of potential computing resources around or beside the Vehicles. In order to effectively respond to the challenge of the network communication brought by the mass Computing requirement generated by the network Edge device, a Mobile Edge Computing (MEC) technology is provided, and a vehicle user can fully utilize the underutilized storage and Computing resources of the network Edge device such as a Road Side Unit (RSU) or one or more nearby vehicles to transfer all or part of the Computing tasks, so that the data volume directly transmitted from the vehicle end to the cloud center is reduced. In which a calculation-oriented edge calculation mode between Vehicle-to-Vehicle (V2V) can employ cooperative communication techniques to increase the specific capacity of a Vehicle user using a mobile relay node.
Computation partitioning is one of the core technologies for implementing mobile edge computation. It allows the vehicle terminal to avoid handling all the computational load, but instead divides the entire application into several computational sub-tasks, which are then distributed to the serving vehicle or RSU for handling the computational load and application processing requirements according to the network connection status. The technology of the computation offload is also important for realizing edge computation, and the computation offload process is influenced by different factors, such as the usage habits of users, the quality of radio channels, the performance of mobile devices, and the availability of cloud servers, so the key point of the computation offload is to specify an appropriate offload decision. Meanwhile, the inherent characteristics of the vehicle networking enable the mobile edge computing scheme to be applied to the vehicle networking and need to consider the physical processes of transmission and computing at the same time, so that the partitioning and unloading of the vehicle computing task also need to consider the physical characteristics of the vehicle communication in a dynamic environment and the computing load of the vehicle terminal, and further ensure the coupling reliability of the communication and the computing.
Disclosure of Invention
The invention aims to provide a dynamic task unloading technology between cooperative vehicles based on mobile edge calculation, so as to improve the utilization rate of communication and calculation resources and improve the coupling reliability of communication and calculation.
The application is realized by the following technical scheme:
a dynamic task unloading technology among cooperative vehicles based on mobile edge calculation comprises the following steps:
step 1, acquiring performance parameters of a channel between vehicles based on cooperative transmission and vehicle calculation performance parameters.
Considering the decode-and-forward based three-node cooperative transmission scenario, the requesting vehicle needs to communicate with the target serving vehicle with the help of the relay vehicle. And setting the sampling time slot T as 1,2, … and T according to the preset task life cycle T. In each time slot t, the performance parameter of the channel between the requesting vehicle and the serving vehicle is Xi,t(i＝λ,k,L1,L2,L3SNR), where λ is a mission allocation parameter used to decide on the local computation versus the service vehicle computation mission ratio, i.e.:
Figure GDA0003490773320000011
k is the path loss coefficient, L1,L2,L3The distances between the requesting vehicle and the service vehicle, between the requesting vehicle and the relay vehicle, and between the relay vehicle and the service vehicle, respectively, and the SNRs are signal-to-noise ratios. Requesting the vehicle to calculate a performance parameter of Yj,t(j＝λ,WR,α,β)，WRA CPU cycle for a requesting vehicle, determined by the respective vehicle, for characterizing its computational capabilities; wherein, the CPU period required by calculating and processing each bit of data obeys Gamma distribution with the shape parameter of alpha and the scale parameter of beta, and the probability density function is expressed as:
Figure GDA0003490773320000021
and 2, constructing a vehicle cooperative communication reliability evaluation model according to the obtained channel performance parameters. And when the task quantity of the request vehicle in the time slot t is s, and the task of (1-lambda) s is calculated by the service vehicle through cooperative transmission, constructing a vehicle cooperative communication reliability evaluation model. The cooperative communication reliability is represented by the interruption probability of the link, and the larger the interruption probability of the link is, the lower the cooperative communication reliability is. The link outage probability is defined as:
Figure GDA0003490773320000022
and 3, constructing a calculation reliability evaluation model of the request vehicle according to the obtained calculation performance parameters of the request vehicle. And when the request vehicle locally calculates the task of lambda s in the time slot t, constructing a request vehicle calculation reliability evaluation model. The more data the CPU needs to compute the processing at a given cutoff duration, the less successful it will be to complete the computation, indicating less reliability of the requesting vehicle to undertake the entire processing task through local computation. Describing this performance from a probabilistic perspective, defining a calculated reliability as the requested vehicle passes through the assignment WRProbability of CPU cycle being able to successfully complete a task before the time slot t expires, if W assigned by the vehicle is requestedRThe CPU period is not enough to support the application program to complete the calculation task, and the success probability of the calculation is as follows:
Figure GDA0003490773320000023
and 4, constructing a virtual queue model based on the service vehicle computing resource limit. Considering that a service vehicle typically includes a processor, data storage components, etc. as memory and cache, the virtual queue is modeled as a single server first-in-first-out queue for storing arriving tasks awaiting execution. According to the litter law, the average queuing delay is proportional to the average queue length, so the virtual queue length of the service vehicle can be used to indirectly characterize the end-to-end delay constraint. An excessively long queue length will eventually undermine the reliability of task computation given the deadline, since incoming tasks will be dropped after overloading the queue buffer of limited size. Defining the historical queue length:
Figure GDA0003490773320000024
wherein WS(i) The computing power allocated for servicing vehicles for computing tasks, characterized by the CPU period, XSThe CPU cycles required to process each bit of data are calculated for the service vehicle. Order to
Figure GDA0003490773320000025
The queue length of the system at the beginning of time slot t is:
Qt＝(1-λt+1)st+1+θt+1Θt+1 (6)
and 5, combining the queue length and the service vehicle privacy perception optimization task distribution parameters. And further defining privacy exposure perception of the service vehicle terminal by referring to privacy entropy, and jointly optimizing task distribution parameters by combining the queue length. The service vehicle defined privacy entropy is:
Figure GDA0003490773320000031
wherein p is the time when the service vehicle receives a service request within the time slot tiValue taking
Figure GDA0003490773320000032
Otherwise piTaking 0; in the case that the service vehicle has a calculation task in each time slot, the system has the maximum privacy entropy
Figure GDA0003490773320000033
Definition PriSTo measure the subjective privacy perception level of the service vehicle:
Figure GDA0003490773320000034
PriSthe increase in value will result in the service vehicle dynamically adjusting its computing power W according to a given threshold, taking into account its privacy exposure rate when assigning computing power to an unloading eventSThereby reducing the likelihood of an attack. At the initial moment of the system, the service vehicle provides a high level of computing power W defined by itselfS(t0)＝WSHThe service vehicle provides a low level of computational power computing power W defined by itself when its subjective privacy perception level is greater than a given thresholdS(t)＝WSLAnd dynamically adjusted as the system evolves. In order to realize load balance of the whole system, a combined request vehicle end queue and a service vehicle end virtual queue are comprehensively considered, and task allocation parameters executed in a time slot t are determined according to the length of a historical queue, wherein the optimized allocation parameters are determined by the following formula:
Figure GDA0003490773320000035
wherein, the system is in an initial state, i.e. t ═ t0Time, random variable
Figure GDA00034907733200000310
Obeying standard uniform distribution
Figure GDA00034907733200000311
Step 6, based on the preset task quantity D of the requested vehicle and the processing time delay T constraint, determining a service request-oriented maximum communication and calculation coupling reliability optimization equation:
Figure GDA0003490773320000036
and 7, realizing a coupling reliability-oriented calculation task dynamic unloading method based on each system parameter and the requested vehicle constraint condition, and outputting an optimal coupling reliability index. Wherein, the phase function is defined as:
Figure GDA0003490773320000037
wherein the content of the first and second substances,
Figure GDA0003490773320000038
the values for the recursive definition of the optimal solution are:
Figure GDA0003490773320000039
and finally, calculating the optimal coupling reliability by adopting a bottom-up dynamic programming method, and recording the calculated task quantity s in each time slott。
Compared with the prior art, the invention has the advantages that:
1. in a vehicle networking edge computing scenario with a vehicle as a mobile edge computing server, the invention constructs an evaluation index of communication and computing coupling reliability by stochastic modeling of communication between a requesting vehicle and a serving vehicle and stochastic modeling of vehicle computing capacity. Today, the reliability of data is no longer the biggest difficulty hindering the development of the car networking system, and the reliability of a mobile edge computing scene with communication performance and vehicle computing performance as indexes is ensured.
2. According to the invention, a history queue optimization task allocation strategy is synthesized, a partial unloading method based on dynamic programming is established on the premise of meeting the calculation requirement and time delay constraint of the requested vehicle, and a basis is provided for the optimal calculated amount and the optimal unloading amount in each time slot by a combined calculation mode of local calculation and the vehicle serving as a server, so that the overall coupling reliability of the system is finally improved.
3. Compared with the traditional computing unloading technology, the method has the advantages that the privacy perception grade is calculated for the service vehicle in the implementation process, so that the service vehicle can dynamically adjust the computing capacity for service according to the subjective perception of the service vehicle on the privacy exposure of the computing event, and powerful support is further provided for privacy protection in the marginal computing scene.
Drawings
FIG. 1 is a flowchart of a method for dynamically unloading tasks between cooperating vehicles according to the present invention.
Detailed Description
The invention will be further explained with reference to the drawings.
The invention discloses a dynamic task unloading method among cooperative vehicles based on mobile edge calculation, which is realized by the following steps as shown in figure 1:
step 1, acquiring performance parameters of a channel between vehicles based on cooperative transmission and vehicle calculation performance parameters.
Considering the decode-and-forward based three-node cooperative transmission scenario, the requesting vehicle needs to communicate with the target serving vehicle with the help of the relay vehicle. And setting the sampling time slot T as 1,2, … and T according to the preset task life cycle T. In each time slot t, the performance parameter of the channel between the requesting vehicle and the serving vehicle is Xi,t(i＝λ,k,L1,L2,L3SNR), where λ is a mission allocation parameter used to decide on the local computation versus the service vehicle computation mission ratio, i.e.:
Figure GDA0003490773320000041
k is the path loss coefficient, L1,L2,L3Distances between the requesting vehicle and the service vehicle, between the requesting vehicle and the relay vehicle, and between the relay vehicle and the service vehicle, respectivelyOff, SNR is the signal-to-noise ratio. Requesting the vehicle to calculate a performance parameter of Yj,t(j＝λ,WR,α,β)，WRA CPU cycle for a requesting vehicle, determined by the respective vehicle, for characterizing its computational capabilities; wherein, the CPU period required by calculating and processing each bit of data obeys Gamma distribution with the shape parameter of alpha and the scale parameter of beta, and the probability density function is expressed as:
Figure GDA0003490773320000042
and 2, constructing a vehicle cooperative communication reliability evaluation model according to the obtained channel performance parameters. And when the task quantity of the request vehicle in the time slot t is s, and the task of (1-lambda) s is calculated by the service vehicle through cooperative transmission, constructing a vehicle cooperative communication reliability evaluation model. The cooperative communication reliability is represented by the interruption probability of the link, and the larger the interruption probability of the link is, the lower the cooperative communication reliability is. The link outage probability is defined as:
Figure GDA0003490773320000043
and 3, constructing a calculation reliability evaluation model of the request vehicle according to the obtained calculation performance parameters of the request vehicle. And when the request vehicle locally calculates the task of lambda s in the time slot t, constructing a request vehicle calculation reliability evaluation model. The more data the CPU needs to compute the processing at a given cutoff duration, the less successful it will be to complete the computation, indicating less reliability of the requesting vehicle to undertake the entire processing task through local computation. Describing this performance from a probabilistic perspective, defining a calculated reliability as the requested vehicle passes through the assignment WRProbability of CPU cycle being able to successfully complete a task before the time slot t expires, if W assigned by the vehicle is requestedRThe CPU period is not enough to support the application program to complete the calculation task, and the success probability of the calculation is as follows:
Figure GDA0003490773320000051
and 4, constructing a virtual queue model based on the service vehicle computing resource limit. Considering that a service vehicle typically includes a processor, data storage components, etc. as memory and cache, the virtual queue is modeled as a single server first-in-first-out queue for storing arriving tasks awaiting execution. According to the litter law, the average queuing delay is proportional to the average queue length, so the virtual queue length of the service vehicle can be used to indirectly characterize the end-to-end delay constraint. An excessively long queue length will eventually undermine the reliability of task computation given the deadline, since incoming tasks will be dropped after overloading the queue buffer of limited size. Defining the historical queue length:
Figure GDA0003490773320000052
wherein WS(i) The computing power allocated for servicing vehicles for computing tasks, characterized by the CPU period, XSThe CPU cycles required to process each bit of data are calculated for the service vehicle. Order to
Figure GDA0003490773320000053
The queue length of the system at the beginning of time slot t is:
Qt＝(1-λt+1)st+1+θt+1Θt+1 (6)
and 5, combining the queue length and the service vehicle privacy perception optimization task distribution parameters. And further defining privacy exposure perception of the service vehicle terminal by referring to privacy entropy, and jointly optimizing task distribution parameters by combining the queue length. The service vehicle defined privacy entropy is:
Figure GDA0003490773320000054
wherein p is the time when the service vehicle receives a service request within the time slot tiValue taking
Figure GDA0003490773320000055
Otherwise piTaking 0; in the case that the service vehicle has a calculation task in each time slot, the system has the maximum privacy entropy
Figure GDA0003490773320000056
Definition PriSTo measure the subjective privacy perception level of the service vehicle:
Figure GDA0003490773320000057
PriSthe increase in value will result in the service vehicle dynamically adjusting its computing power W according to a given threshold, taking into account its privacy exposure rate when assigning computing power to an unloading eventSThereby reducing the likelihood of an attack. At the initial moment of the system, the service vehicle provides a high level of computing power W defined by itselfS(t0)＝WSHThe service vehicle provides a low level of computational power computing power W defined by itself when its subjective privacy perception level is greater than a given thresholdS(t)＝WSLAnd dynamically adjusted as the system evolves. In order to realize load balance of the whole system, a combined request vehicle end queue and a service vehicle end virtual queue are comprehensively considered, and task allocation parameters executed in a time slot t are determined according to the length of a historical queue, wherein the optimized allocation parameters are determined by the following formula:
Figure GDA0003490773320000061
wherein, the system is in an initial state, i.e. t ═ t0Time, random variable
Figure GDA0003490773320000062
Obeying standard uniform distribution
Figure GDA0003490773320000063
Step 6, based on the preset task quantity D of the requested vehicle and the processing time delay T constraint, determining a service request-oriented maximum communication and calculation coupling reliability optimization equation:
Figure GDA0003490773320000064
and 7, realizing a coupling reliability-oriented calculation task dynamic unloading method based on each system parameter and the requested vehicle constraint condition, and outputting an optimal coupling reliability index. Wherein, the phase function is defined as:
Figure GDA0003490773320000065
wherein the content of the first and second substances,
Figure GDA0003490773320000066
the values for the recursive definition of the optimal solution are:
Figure GDA0003490773320000067
and finally, calculating the optimal coupling reliability by adopting a bottom-up dynamic programming method, and recording the calculated task quantity s in each time slott。

Claims (1)

1. A method for dynamically unloading tasks among cooperative vehicles based on mobile edge calculation is characterized in that: the method is realized by the following steps:
step 1, acquiring performance parameters of a channel between vehicles based on cooperative transmission and vehicle calculation performance parameters;
step 2, constructing a vehicle cooperative communication reliability evaluation model;
step 3, constructing a model for requesting the calculation of the reliability of the vehicle;
step 4, constructing a virtual queue model based on service vehicle computing resource limitation;
step 5, combining the queue length and service vehicle privacy perception optimization task allocation parameters;
step 6, determining a service request-oriented maximum communication and calculation coupling reliability optimization equation based on the processing time delay constraint of the preset task of the request vehicle;
step 7, realizing a coupling reliability-oriented calculation task dynamic unloading method, and outputting an optimal coupling reliability index;
in the step 1, a decoding-forwarding-based three-node cooperative transmission scene is considered, a request vehicle needs to communicate with a target service vehicle under the help of a relay vehicle, a sampling time slot T is set to be 1,2i，t(i＝λ，k，L1，L2，L3SNR), where λ is a mission allocation parameter used to decide on the local computation versus the service vehicle computation mission ratio, i.e.:
Figure FDA0003416148390000011
k is the path loss coefficient, L1，L2，L3Respectively the distances between the request vehicle and the service vehicle, between the request vehicle and the relay vehicle and between the relay vehicle and the service vehicle, the SNR is the signal-to-noise ratio, and the calculation performance parameter of the request vehicle is Yj，t(j＝λ，WR，α，β)，WRA CPU cycle for a requesting vehicle, determined by the respective vehicle, for characterizing its computational capabilities; wherein, the CPU period required by calculating and processing each bit of data obeys Gamma distribution with the shape parameter of alpha and the scale parameter of beta, and the probability density function is expressed as:
Figure FDA0003416148390000012
wherein τ (α) is a Gamma distribution function; x is a random variable;
according to the channel performance parameters, when the task quantity of a request vehicle in a time slot t is s, and the task of (1-lambda) s is calculated by a service vehicle through cooperative transmission, a vehicle cooperative communication reliability evaluation model is constructed, wherein the cooperative communication reliability is represented by the interruption probability of a link, the larger the link interruption probability is, the lower the cooperative communication reliability is, and the link interruption probability is defined as:
Figure FDA0003416148390000013
according to the obtained performance parameter of the requested vehicle calculation, when the requested vehicle locally calculates the task of lambda s in the time slot t, a requested vehicle calculation reliability evaluation model is constructed, under the condition of a given cutoff duration, the more data quantity needing calculation processing of a CPU, the smaller the success probability of the calculation completion of the CPU, the smaller the reliability of the requested vehicle for bearing the whole processing task through local calculation is, the performance is described from the aspect of probability, and the calculation reliability is defined as the W distribution of the requested vehicleRProbability of CPU cycle being able to successfully complete a task before the time slot t expires, if W assigned by the vehicle is requestedRThe CPU period is not enough to support the application program to complete the calculation task, and the success probability of the calculation is as follows:
Figure FDA0003416148390000014
the step of constructing the virtual queue model based on the service vehicle computing resource limit is as follows: the service vehicle comprises a processor, the data storage component is used as a memory and a cache, the virtual queue is modeled as a single server first-in first-out queue and is used for storing arriving tasks waiting to be executed, according to the Litter law, the average queuing delay is in proportion to the average queue length, therefore, the virtual queue length of the service vehicle can be used for indirectly representing end-to-end delay constraint, under the condition of a given cutoff time, the reliability of task calculation is finally destroyed due to the overlong queue length, and after a queue buffer with a limited size is overloaded, an incoming task is discarded, and the historical queue length is defined as follows:
Figure FDA0003416148390000021
wherein WS(i) The computing power allocated for servicing vehicles for computing tasks, characterized by the CPU period, XSCalculating a CPU cycle required for processing each bit of data for the service vehicle; order to
Figure FDA0003416148390000022
The queue length of the system at the beginning of time slot t is:
Qt＝(1-λt+1)st+1+θt+1Θt+1 (6)；
the service vehicle defined privacy entropy is:
Figure FDA0003416148390000023
wherein the content of the first and second substances,
Figure FDA0003416148390000024
representing a probability distribution, p when the service vehicle receives a request for a transaction within a time slot tiValue taking
Figure FDA0003416148390000025
Otherwise piTaking 0; in the case that the service vehicle has a calculation task in each time slot, the system has the maximum privacy entropy
Figure FDA0003416148390000026
Definition PriSTo measure the subjective privacy perception level of the service vehicle:
Figure FDA0003416148390000027
PriSthe increase in value will result in the service vehicle dynamically adjusting its computing power W according to a given threshold, taking into account its privacy exposure rate when assigning computing power to an unloading eventSThereby reducing the possibility of attack, and at the initial moment of the system, the service vehicle provides a high level of computing power W defined by itselfS(t0)＝WSHThe service vehicle provides a low level of computational power computing power W defined by itself when its subjective privacy perception level is greater than a given thresholdS(t)＝WSLAnd with the dynamic adjustment of the system evolution, in order to realize the load balance of the whole system, the combined request vehicle end queue and the service vehicle end virtual queue are comprehensively considered, and the task allocation parameter executed in the time slot t is determined according to the historical queue length, and the optimized allocation parameter is determined by the following formula:
Figure FDA0003416148390000028
wherein, the system is in an initial state, i.e. t ═ t0Time, random variable
Figure FDA0003416148390000029
Obeying standard uniform distribution
Figure FDA00034161483900000210
Step 7 provides a dynamic programming algorithm based on steps 1-6, and a dynamic unloading method for the calculation task facing the coupling reliability is realized; the phase function is defined as:
Figure FDA00034161483900000211
wherein the content of the first and second substances,
Figure FDA00034161483900000212
the values for the recursive definition of the optimal solution are:
Figure FDA0003416148390000031
wherein, P (l)T) For the task quantity to be l in T-T time slotTDefines the value of the optimal solution; and finally, calculating the optimal coupling reliability by adopting a bottom-up dynamic programming method, and recording the calculated task quantity s in each time slott。
CN202010727802.8A 2020-07-24 2020-07-24 Method for dynamically unloading tasks among cooperative vehicles based on mobile edge calculation Active CN112512018B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010727802.8A CN112512018B (en) 2020-07-24 2020-07-24 Method for dynamically unloading tasks among cooperative vehicles based on mobile edge calculation

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010727802.8A CN112512018B (en) 2020-07-24 2020-07-24 Method for dynamically unloading tasks among cooperative vehicles based on mobile edge calculation

Publications (2)

Publication Number Publication Date
CN112512018A CN112512018A (en) 2021-03-16
CN112512018B true CN112512018B (en) 2022-03-04

Family

ID=74953344

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010727802.8A Active CN112512018B (en) 2020-07-24 2020-07-24 Method for dynamically unloading tasks among cooperative vehicles based on mobile edge calculation

Country Status (1)

Country Link
CN (1) CN112512018B (en)

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113423091B (en) * 2021-05-24 2022-07-29 西安电子科技大学 Multidimensional resource intelligent joint optimization method and system of vehicle-mounted computing power network
CN113316116B (en) * 2021-05-27 2022-08-16 南京邮电大学 Vehicle calculation task unloading method
CN113254219B (en) * 2021-06-28 2021-10-15 国汽智控(北京)科技有限公司 Cooperative processing method, cooperative processing device, electronic device, and storage medium
CN113626107B (en) * 2021-08-20 2024-03-26 中南大学 Mobile computing unloading method, system and storage medium
CN113993108B (en) * 2021-12-27 2022-03-18 江苏移动信息系统集成有限公司 Cache content placement method and system based on vehicle-mounted network edge

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109067842A (en) * 2018-07-06 2018-12-21 电子科技大学 Calculating task discharging method towards car networking
CN109391681A (en) * 2018-09-14 2019-02-26 重庆邮电大学 V2X mobility prediction based on MEC unloads scheme with content caching

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10705884B2 (en) * 2018-07-12 2020-07-07 Toyota Jidosha Kabushiki Kaisha Managing computational tasks in vehicle context
CN108990016B (en) * 2018-08-17 2019-11-26 电子科技大学 A kind of calculating task unloading of more vehicles collaboration and transmission method
CN109951821B (en) * 2019-02-26 2022-03-11 重庆邮电大学 Task unloading scheme for minimizing vehicle energy consumption based on mobile edge calculation
CN110928691B (en) * 2019-12-26 2021-07-09 广东工业大学 Traffic data-oriented edge collaborative computing unloading method
CN111314889B (en) * 2020-02-26 2023-03-31 华南理工大学 Task unloading and resource allocation method based on mobile edge calculation in Internet of vehicles

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109067842A (en) * 2018-07-06 2018-12-21 电子科技大学 Calculating task discharging method towards car networking
CN109391681A (en) * 2018-09-14 2019-02-26 重庆邮电大学 V2X mobility prediction based on MEC unloads scheme with content caching

Also Published As

Publication number Publication date
CN112512018A (en) 2021-03-16

Similar Documents

Publication Publication Date Title
CN112512018B (en) Method for dynamically unloading tasks among cooperative vehicles based on mobile edge calculation
CN108509276B (en) Video task dynamic migration method in edge computing environment
US8887217B2 (en) Systems and methods for quality of experience aware joint scheduling of buffered video on demand and best effort flows
CN109218355B (en) Load balancing engine, client, distributed computing system and load balancing method
JP3803712B2 (en) Dynamic control method of bandwidth limit value for non-real-time communication
US6560243B1 (en) System and method for receiver based allocation of network bandwidth
CN111400001B (en) Online computing task unloading scheduling method facing edge computing environment
US7177271B2 (en) Method and system for managing admission to a network
CN112737823A (en) Resource slice allocation method and device and computer equipment
CN108092908B (en) Method for controlling flow and sending end equipment
CN113141590B (en) Industrial Internet of things-oriented wireless communication scheduling method and device
CN110113140B (en) Calculation unloading method in fog calculation wireless network
US20140281034A1 (en) System and Method for Compressing Data Associated with a Buffer
CN113452566A (en) Cloud edge side cooperative resource management method and system
CN111935783A (en) Edge cache system and method based on flow perception
CN110996390B (en) Wireless access network computing resource allocation method and network system
CN111211996B (en) Flow scheduling method
US9621438B2 (en) Network traffic management
CN111949408A (en) Dynamic allocation method for edge computing resources
AU2021106293A4 (en) Dynamic task unloading method between cooperative vehicles based on mobile edge computing
CN116302578A (en) QoS (quality of service) constraint stream application delay ensuring method and system
CN113271221B (en) Network capacity opening method and system and electronic equipment
CN114936089A (en) Resource scheduling method, system, device and storage medium
CN113300973B (en) Scheduling method and device for multi-queue transmission, electronic equipment and storage medium
WO2024051424A1 (en) Network resource allocation method, apparatus, and storage medium

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