CN107317700B - Vehicle-mounted edge computing node selection system and method - Google Patents

Vehicle-mounted edge computing node selection system and method Download PDF

Info

Publication number
CN107317700B
CN107317700B CN201710434272.6A CN201710434272A CN107317700B CN 107317700 B CN107317700 B CN 107317700B CN 201710434272 A CN201710434272 A CN 201710434272A CN 107317700 B CN107317700 B CN 107317700B
Authority
CN
China
Prior art keywords
vehicle
user
task
node
edge computing
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
CN201710434272.6A
Other languages
Chinese (zh)
Other versions
CN107317700A (en
Inventor
章磊
陈玉龙
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.)
Hubei Polytechnic University
Original Assignee
Hubei Polytechnic 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 Hubei Polytechnic University filed Critical Hubei Polytechnic University
Priority to CN201710434272.6A priority Critical patent/CN107317700B/en
Publication of CN107317700A publication Critical patent/CN107317700A/en
Application granted granted Critical
Publication of CN107317700B publication Critical patent/CN107317700B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/14—Network analysis or design
    • H04L41/145—Network analysis or design involving simulating, designing, planning or modelling of a network
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/50—Network service management, e.g. ensuring proper service fulfilment according to agreements
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/01—Protocols
    • H04L67/12—Protocols specially adapted for proprietary or special-purpose networking environments, e.g. medical networks, sensor networks, networks in vehicles or remote metering networks
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W16/00—Network planning, e.g. coverage or traffic planning tools; Network deployment, e.g. resource partitioning or cells structures
    • H04W16/02—Resource partitioning among network components, e.g. reuse partitioning
    • H04W16/10—Dynamic resource partitioning
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W40/00—Communication routing or communication path finding
    • H04W40/02—Communication route or path selection, e.g. power-based or shortest path routing
    • H04W40/04—Communication route or path selection, e.g. power-based or shortest path routing based on wireless node resources
    • 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

Abstract

The invention discloses a vehicle-mounted edge computing node selection system and a method, wherein the system comprises the following steps: the plurality of users send task requests to the wireless access point and complete the task unloading process according to the returned matched vehicle information; each vehicle is used as a vehicle-mounted edge computing node and sends state information to the wireless access point at regular intervals; waiting for successful matching with the user, executing the corresponding task, and assisting the user to finish task unloading; the wireless access point receives and processes data sent by the user and the vehicle edge node, and executes a node selection method to pair the user and the vehicle edge computing node; the wireless access node specifically comprises a communication constraint module, a calculation constraint module, a preliminary matching module and a resource allocation module. According to the method, the total user experience in a certain area can be maximized through the optimal decision-making matching scheme of the vehicle-mounted edge computing node and the user, and meanwhile, the limited spectrum resources of the user-node pair with the maximum efficiency are guaranteed.

Description

Vehicle-mounted edge computing node selection system and method
Technical Field
The invention belongs to the field of wireless communication, relates to an Edge Computing (EC) architecture, and particularly relates to a vehicle-mounted Edge Computing node selection system and a vehicle-mounted Edge Computing node selection method.
Background
Edge Computing (EC) was first proposed by cisco in 2014 with the goal of making efficient use of locally idle Computing and storage resources. Nearest edge computing is seen as an example for achieving low service latency and high data security, privacy security. Notably, edge computing and traditional cloud computing are compatible and may cooperate with each other. In particular, edge computing may utilize limited local resources to pre-process data before sending it to a remote cloud for further processing. On the other hand, the edge calculation can also be used for processing delay-sensitive or geographical-position-related applications independently, and can ensure lower delay and energy consumption. Recently, with the advent of computing-intensive mobile applications such as augmented reality and virtual reality, users are more inclined to offload tasks to edge compute nodes to leverage their computing and storage resources to enhance the user Experience (Quality of Experience, QoE).
Recently, the offloading of tasks to edge compute nodes has also received some attention from researchers. Can be divided into two categories: one is to adopt a logic edge computing node, which is not illustrated by a specific example, but is referred to by the concept of an edge computing node; and the other type of the edge computing node is embodied as a small base station (small cell), an unmanned aerial vehicle (unmanned aerial vehicle), a vehicle (vehicle) and the like, and specific property analysis is combined. Different offloading strategies are proposed for different research modes of the edge computing nodes.
(1) The nodes are computed using logical edges. A strategy for consolidating resources in distributed devices to provide them to mobile applications is described in the documents z.m. zhang N, Yang X and S.Y, "crown-funding: a new resource deployment mode for mobile closed computing," PLoS ONE, dec.2016. According to the method, scattered equipment is abstracted into edge computing nodes, and QoE values are obtained by weighting the credit, the availability and the service delay of the nodes, so that unloading nodes are selected. However, the method does not specify a specific edge computing node and has no practical operability.
(2) With specific edge compute nodes, the proposed schemes are:
(2.1) Small base stations as edge computing nodes, see S.Barbarossa, S.Sardelitti, and P.D.Lorenzo, "Joint allocation of computation and communication resources in Multi-user Mobile computing," in IEEE Workshop on Signal processing Advances in Wireless Communications (SPAWC), Darmstadt, Jun.2013, pp.26-30. The method jointly allocates communication and computing resources of the small base station, so that more requests can be processed with less energy consumption. The above approach requires the addition of extra computational functionality at the small base station, which increases costs.
(2.2) unmanned aerial vehicle as edge computing node, see documents N.H.Motlagh, M.Bagaa and T.Taleb, "UAV Selection for A UAV-Based Integrated IoT Platform," IEEE Global communications Conference (GLOBECOM), Washington, DC,2016, pp.1-6. According to the method, the energy consumption and the running time of the unmanned aerial vehicle are optimized by solving the linear integer programming problem, and the energy consumption and the running time are expected to be lower on the premise of completing the same task. But the energy of the unmanned aerial vehicle in the method also needs to support the flight of the unmanned aerial vehicle, and the energy consumption left for calculation can be very limited.
And (2.3) using the vehicle as an edge computing node. Vehicles can now act as nodes in the EC by connecting to the network. The network formed by the interconnection of vehicles is called Internet of vehicles (IoV). Some of the more recent researchers have identified IoV as a very promising EC platform for the following reasons: firstly, the number of vehicles is huge and the vehicles are widely distributed, so that a user can conveniently access the vehicles around; secondly, an On Board Unit (OBU) On the vehicle can provide rich computing and storage resources for a terminal user; finally, the vehicle does not need to consider the electrical problem, since it can be considered as a small generator by itself.
See documents H.Zhang, Q.Zhang, and X.Du, "heated vehicle-associated group computing for smartphones," IEEE Transactions on vehicle Technology, vol.64, No.12, pp.5610-5618, dec.2015. The method provides the method that idle resources in the vehicle are utilized to assist the smart phone in task unloading, and the task can be completed within a time limit by selecting a proper vehicle computing node. But this method does not consider the guarantee of user experience and can only support the offloading of the computing tasks of the handset.
Disclosure of Invention
The invention aims to provide a vehicle-mounted edge computing node selection method, which can maximize the total user experience in a certain area through an optimal decision-making matching scheme of the vehicle-mounted edge computing node and a user, and simultaneously ensures that the user-node pair can use limited spectrum resources with the maximum efficiency.
In order to achieve the purpose, the adopted technical scheme is as follows:
a vehicle-mounted edge computing node selection system is provided, comprising a plurality of users, a wireless access node and a plurality of vehicles;
the plurality of users send task requests to the wireless access point and complete the task unloading process according to the returned matched vehicle information;
each vehicle is used as a vehicle-mounted edge computing node and sends own state information to the wireless access point at regular intervals; waiting for successful matching with the user, executing the corresponding task, and assisting the user to finish task unloading;
the wireless access point receives and processes data sent by the user and the vehicle edge node, executes a node selection method, and pairs the user and the vehicle edge computing node;
the wireless access node specifically comprises a communication constraint module, a calculation constraint module, a preliminary matching module and a resource allocation module;
the communication constraint module calculates a time constraint condition according to the mobility of the vehicle and determines the number of sub-channels required by pairing between the user and the vehicle; dividing a channel into a plurality of sub-channels with equal bandwidth through orthogonal frequency division multiple access, wherein the average data transmission rate of each sub-channel of the same link is equal; calculating the maximum communication time length of the vehicle according to the vehicle state information, and taking the smaller of the time constraint of the task and the communication time length of the vehicle as the actual total time length of the task, wherein the total time length comprises the time of the task sending and processing process; calculating task processing time and task sending time according to the size of each task, calculating required data transmission rate, and then calculating actual data transmission rate according to a Shannon formula, noise power and a channel gain matrix between a user and a vehicle; the required data transmission rate is equal to the actual data transmission rate, and the number of sub-channels distributed between each pair of users and the vehicle is obtained;
the calculation constraint module is used for modeling the energy consumed by the processing task and is divided into two cases for modeling according to whether unloading is carried out or not: 1) if not, the task is processed locally; under the condition that the task execution time is just the time constraint of the task, the CPU frequency of the user equipment takes a smaller value between the maximum frequency and the task required frequency according to the dynamic voltage scaling technology; then calculating local consumed energy; 2) if unloading is carried out, the task is processed at the vehicle-mounted edge node; calculating energy consumed by the user equipment for transmitting and receiving data and static power; if the data volume of the task result is very small, the consumed energy and time are ignored; calculating the energy consumed by the user equipment when the task is unloaded by combining the sending time;
the preliminary matching module is used for carrying out preliminary matching on the user-vehicle edge computing nodes based on a bilateral auction theory according to the number of the user-vehicle required sub-channels output by the communication constraint module, the energy consumption required by the processing tasks output by the computing constraint module and a user experience evaluation model; measuring user experience of a user through three aspects of node availability, node reputation and communication overhead, and pairing the user and the vehicle-mounted edge computing node based on a bilateral auction theory;
and the resource allocation module is used for allocating sub-channel resources based on linear programming according to the user-vehicle pairing scheme output by the primary matching module and outputting a final matching and spectrum resource allocation result.
According to the technical scheme, the task request sent by the user comprises the task size, the QoE value to be achieved, time constraint, sending power, static power, maximum CPU frequency and user coordinates.
According to the technical scheme, the state information of the vehicle comprises the coordinates, the speed, the movement direction, the maximum CPU frequency, the credit and the availability of the vehicle.
According to the technical scheme, the node availability refers to the probability that the node can provide services after receiving the tasks, the node reputation refers to the probability that the node can provide services according to quality after receiving the tasks, and the communication overhead represents the energy consumed by a user for sending and receiving data.
According to the technical scheme, the user and the vehicle respectively send the task request and the state information to the wireless access point through the WiFi network, and the user and the vehicle-mounted edge computing node are in short-distance communication based on the orthogonal frequency division multiple access technology.
The invention also provides a vehicle-mounted edge computing node selection method, which comprises the following steps:
a plurality of users send task requests to the wireless access point, and the task unloading process of the users is completed according to the returned matched vehicle information;
each vehicle is used as a vehicle-mounted edge computing node, and the plurality of vehicles send own state information to the wireless access point at regular intervals; waiting for successful matching with the user, executing the corresponding task, and assisting the user to finish task unloading;
the wireless access point receives and processes data sent by a user and a vehicle edge node, executes a node selection method, pairs the user and the vehicle edge computing node, and specifically comprises the following steps:
calculating a time constraint condition according to the mobility of the vehicle, and determining the number of sub-channels required by pairing between the user and the vehicle; dividing a channel into a plurality of sub-channels with equal bandwidth through orthogonal frequency division multiple access, wherein the average data transmission rate of each sub-channel of the same link is equal; calculating the maximum communication time length of the vehicle according to the vehicle state information, and taking the smaller of the time constraint of the task and the communication time length of the vehicle as the actual total time length of the task, wherein the total time length comprises the time of the task sending and processing process; calculating task processing time and task sending time according to the size of each task, calculating required data transmission rate, and then calculating actual data transmission rate according to a Shannon formula, noise power and a channel gain matrix between a user and a vehicle; the required data transmission rate is equal to the actual data transmission rate, and the number of sub-channels distributed between each pair of users and the vehicle is obtained;
modeling energy consumed by processing tasks, and modeling under two conditions according to whether unloading is carried out: 1) if not, the task is processed locally; under the condition that the task execution time is just the time constraint of the task, the CPU frequency of the user equipment takes a smaller value between the maximum frequency and the task required frequency according to the dynamic voltage scaling technology; then calculating local consumed energy; 2) if unloading is carried out, the task is processed at the vehicle-mounted edge node; calculating energy consumed by the user equipment for transmitting and receiving data and static power; if the data volume of the task result is very small, the consumed energy and time are ignored; calculating the energy consumed by the user equipment when the task is unloaded by combining the sending time;
according to the number of sub-channels required by the user and the vehicle and output by the communication constraint module, the energy consumption required by the processing task and output by the calculation constraint module and the constructed user experience evaluation model, and based on a bilateral auction theory, performing preliminary matching of the user and the vehicle-mounted edge calculation node; measuring user experience of a user through three aspects of node availability, node reputation and communication overhead, and pairing the user and the vehicle-mounted edge computing node based on a bilateral auction theory;
and according to the user-vehicle pairing scheme output by the primary matching module, performing sub-channel resource allocation based on linear programming, and outputting a final matching and spectrum resource allocation result.
According to the technical scheme, the task request sent by the user comprises the task size, the QoE value to be achieved, time constraint, sending power, static power, maximum CPU frequency and user coordinates.
According to the technical scheme, the state information of the vehicle comprises the coordinates, the speed, the movement direction, the maximum CPU frequency, the credit and the availability of the vehicle.
According to the technical scheme, the node availability refers to the probability that the node can provide services after receiving the tasks, the node reputation refers to the probability that the node can provide services according to quality after receiving the tasks, and the communication overhead represents the energy consumed by a user for sending and receiving data.
According to the technical scheme, the user and the vehicle respectively send the task request and the state information to the wireless access point through the WiFi network, and the user and the vehicle-mounted edge computing node are in short-distance communication based on the orthogonal frequency division multiple access technology.
The invention has the following beneficial effects: the vehicle-mounted edge computing node selection method provided by the invention considers that the edge computing node is set as a vehicle and considered by combining the mobility of the edge computing node. Modeling is carried out on the user experience of the users, and the sum of the user experience obtained by all the users is enabled to be as large as possible through the selection method. Meanwhile, spectrum resource allocation is carried out on each pair of matched users and nodes, so that resource utilization is maximized. The invention can be applied to the field of future vehicle networking, effectively utilizes the idle resources of vehicles and has wide application range.
Drawings
The invention will be further described with reference to the accompanying drawings and examples, in which:
FIG. 1 is a system block diagram of a task offloading process provided by the present invention;
fig. 2 is a system block diagram of modules within a wireless access point.
Fig. 3 is a flowchart of the in-vehicle edge calculation.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
As shown in fig. 1, the present invention involves multiple users, multiple on-board edge computing nodes, and wireless access points.
A plurality of users (user 1, user 2) can send task requests (the request content includes task size, QoE value to be achieved, time constraint, transmission power, static power, maximum CPU frequency and user coordinate) to the wireless access point 3, obtain matched vehicle information, and thus continue to complete the task offloading process.
The plurality of vehicle-mounted edge computing nodes (the vehicle-mounted edge computing nodes 4 and the vehicle-mounted edge computing nodes 5) send own state information (the state information comprises the coordinates, the speed, the movement direction, the maximum CPU frequency, the credit and the availability of the vehicle) to the wireless access point at regular intervals. And waiting for successful matching with the user so as to assist in completing task unloading.
The wireless access point 3 is a task allocation end, processes received user and vehicle edge node data, executes a node selection method, and pairs multiple users and multiple vehicle-mounted edge computing nodes.
Fig. 2 is a system block diagram of internal modules of the wireless access point, including a communication constraint module 31, a calculation constraint module 32, a preliminary matching module 33, and a resource allocation module 34. The communication constraint module and the calculation constraint module preprocess the collected information of the user and the vehicle edge node, respectively output results to the primary matching module, then the output results of the primary matching module enter the resource allocation module, and output a final user-vehicle edge node pairing and spectrum resource allocation scheme.
The wireless access point receives the state information sent by the nearby vehicles at regular time intervals (e.g., 100 ms). When a user has a task to offload, they also send their request information to the wireless access point. The wireless access point will buffer the information, and then perform a node selection method to match the user and the node every fixed period (e.g. 100ms) in combination with the information of the user and the vehicle, and output the matching result.
The communication constraint module 31 calculates its time constraint considering the mobility of the vehicle, thereby determining the communication bandwidth between the user and the node. Suppose that the user set is U, the number of users is M, the vehicle-mounted edge computing node set is V, and the number of vehicle-mounted edge computing nodes is N. We consider Orthogonal Frequency Division Multiple Access (OFDMA), dividing the channel into several equal bandwidths (W;)0) The average data transmission rates of the subchannels of (1) are equal. L is the total number of channels, LijThe smaller of the time constraint of the task itself and the communication time of the vehicle is taken as the actual total time of the task, including the time of the task sending and processing processes
Figure BDA0001318211130000081
Join task size (user i's task size is denoted D)i) The required data transmission rate can be calculated and then the noise power N is calculated according to the Shannon formula0And the actual data transmission rate calculated from the channel gain matrix H between the user and the vehicle (H is the channel gain matrix between user i and vehicle j)ij). The number of sub-channels distributed between each pair of users and nodes can be obtained by making the two equal;
the computational constraint module 32 models the energy consumed by the processing task and considers two cases:
1) if not, the task is processed locally. Under the condition that the task execution time is just the time constraint of the task, the CPU frequency of the user equipment takes a smaller value between the maximum frequency and the task demand frequency according to a Dynamic Voltage Scaling (DVS) technology. The local energy consumption is then calculated in combination with the correlation formula (the local energy consumption of user i is expressed as
Figure BDA0001318211130000082
)。
2) If the unloading is performed, the task is processed at the edge node. It is necessary to calculate the energy consumed by the ue to transmit and receive data and static power (the task off-load of user i consumes energy of
Figure BDA0001318211130000083
A transmission power of
Figure BDA0001318211130000084
). Since the data amount of the task result is very small, the consumed energy and time can be ignored, and are described in M.Halpern, Y.Zhu, and V.J.Reddi, "Mobile cpu's rise to power," Quantifying the impact of genetic engineering cpu design on Performance, energy, and user utilization, in IEEEInternational Symposium on High Performance Computer Architecture (HPCA), Barcelona, Mar.2016, pp.64-76. The energy consumed by the user equipment when the task is unloaded can be calculated by combining the sending time.
The computational constraint module, which considers both the unloaded and un-unloaded cases, has not been able to decide whether to unload (the non-paired vehicle is considered not to unload) as a result of the final resource allocation module.
The preliminary matching module 33 performs preliminary matching of the user-vehicle edge computing node based on the user experience evaluation model and based on the bilateral auction theory. The user experience evaluation model is realized by three parties of node availability, node reputation and communication overheadThe user experience of a user is measured (node availability refers to the probability that a node can provide a service after receiving a task, node reputation refers to the probability that a node can provide a service according to quality after receiving a task, and communication overhead represents the energy consumed by the user to send and receive data). The user has a preference coefficient for each index, using rxIndicating that (x ∈ {1,2,3}), taking values between (0,1), the three values are added to be 1 (using three values of
Figure BDA0001318211130000091
The value representing the xth aspect of the vehicle j) is normalized to (0,1) and then weighted according to the preference coefficient to obtain the corresponding user experience (the user experience provided by the vehicle j for the user i is QoE)ij) Use αijIndicating whether user i matches vehicle j.
According to the communication constraint, the calculation constraint and the spectrum resource constraint limitation, the following optimization problem is established for improving the user experience. The optimization problem can output a user-node matching result that improves the overall user experience. The optimization problem is established as follows:
Figure BDA0001318211130000092
limited by: s.t are provided.
Figure BDA0001318211130000093
Figure BDA0001318211130000094
Figure BDA0001318211130000095
Figure BDA0001318211130000096
Figure BDA0001318211130000097
Wherein M is the number of users, which can generally be 1-50, and N is the number of vehicles, which can generally be 10-100.
The problem can be solved by two subproblems, namely a user and vehicle-mounted edge computing node pairing problem based on a bilateral auction theory, and a limited spectrum resource allocation problem. The preliminary matching module solves the first problem. The theory of bilateral auctions is described in detail in A.L.jin, W.Song, P.Wang, D.Niyato, and P.Ju, "Automation mechanics aware efficacy sharing for sharing in mobile sharing Computing," IEEE Transactions on Services Computing, vol.9, No.6, pp.895-909, Nov.2016. The solving algorithm is as follows:
QNS preliminary matching stage
_____________________________________
1. Sequencing the set Q, and enabling the serial number of the user i to be 1;
2. for user i ∈ U, if i ≦ M, perform the following steps:
3. definition of Q-iDeleting Q of user i for set QiThe formed set. Find the set Q-iAnd is defined as
Figure BDA0001318211130000101
4. If node j successfully matches user i, we delete node j from set V;
5. define C as satisfying QoE in set Vij≥QiThe number of nodes of (c);
6. if C is 0, then
Figure BDA0001318211130000102
User i mismatches and executes i ═ i +1, returning to Step 2;
7. if C is 1, the following steps are executed:
8. if node x is QoE compliantix≥QiThen define the corresponding QoE as Qselect；
9. Find set V0Satisfy QoEij≤QselectIf so, selecting the node y with the highest QoE, wherein the QoE provided by the node y is the QoEiy；
10. Defining a balanced QoE of
Figure BDA0001318211130000103
11. If there is
Figure BDA0001318211130000104
And
Figure BDA0001318211130000105
user i and node x are successfully matched, wherein,
Figure BDA0001318211130000106
and j is not equal to x,
Figure BDA0001318211130000107
executing i to i +1, and returning to Step 2;
12. if C is more than or equal to 2, the following steps are executed:
13. will satisfy QoE in set Vij≥QiThe set of nodes of (a) is defined as I. Finding the node x with the highest QoE in the set I, wherein the QoE provided by the node x is defined as Qselect；
14. Find set V0Satisfy QoEij≤QselectNode(s) of (2) selecting node(s) having the highest QoEijNode y of (a);
15. defining a balanced QoE of
Figure BDA0001318211130000111
16. If there is
Figure BDA0001318211130000112
And
Figure BDA0001318211130000113
user i and node x are successfully matched, wherein,
Figure BDA0001318211130000114
and j is not equal to x,
Figure BDA0001318211130000115
executing i to i +1, and returning to Step 2;
17. otherwise, deleting the node x from the set I, and selecting the highest QoE in the set I againijThe node of (2) defines the QoE provided by the node as QselectAnd returns to Step 14;
the resource allocation module 34 solves the second problem, the problem of communication resource allocation based on Linear programming, see a. schrijver, the Theory of Linear and Integer programming, john Wiley and Sons, 1998. And outputs the final matching result. The optimization problem solved by this process is as follows:
Figure BDA0001318211130000116
limited by:
Figure BDA0001318211130000117
Figure BDA0001318211130000118
the existing solution is solved using the 0-1 knapsack problem.
As shown in fig. 3, based on the above system, the present invention further provides a method for selecting a vehicle-mounted edge computing node, including the following steps:
a plurality of users send task requests to the wireless access point, and the task unloading process of the users is completed according to the returned matched vehicle information;
each vehicle is used as a vehicle-mounted edge computing node, and the plurality of vehicles send own state information to the wireless access point at regular intervals; waiting for successful matching with the user, executing the corresponding task, and assisting the user to finish task unloading;
the wireless access point receives and processes data sent by a user and a vehicle edge node, executes a node selection method, pairs the user and the vehicle edge computing node, and specifically comprises the following steps:
calculating a time constraint condition according to the mobility of the vehicle, and determining the number of sub-channels required by pairing between the user and the vehicle; dividing a channel into a plurality of sub-channels with equal bandwidth through orthogonal frequency division multiple access, wherein the average data transmission rate of each sub-channel of the same link is equal; calculating the maximum communication time length of the vehicle according to the vehicle state information, and taking the smaller of the time constraint of the task and the communication time length of the vehicle as the actual total time length of the task, wherein the total time length comprises the time of the task sending and processing process; calculating task processing time and task sending time according to the size of each task, calculating required data transmission rate, and then calculating actual data transmission rate according to a Shannon formula, noise power and a channel gain matrix between a user and a vehicle; the required data transmission rate is equal to the actual data transmission rate, and the number of sub-channels distributed between each pair of users and the vehicle is obtained;
modeling energy consumed by processing tasks, and modeling under two conditions according to whether unloading is carried out: 1) if not, the task is processed locally; under the condition that the task execution time is just the time constraint of the task, the CPU frequency of the user equipment takes a smaller value between the maximum frequency and the task required frequency according to the dynamic voltage scaling technology; then calculating local consumed energy; 2) if unloading is carried out, the task is processed at the vehicle-mounted edge node; calculating energy consumed by the user equipment for transmitting and receiving data and static power; if the data volume of the task result is very small, the consumed energy and time are ignored; calculating the energy consumed by the user equipment when the task is unloaded by combining the sending time;
according to the number of sub-channels required by the user and the vehicle and output by the communication constraint module, the energy consumption required by the processing task and output by the calculation constraint module and the constructed user experience evaluation model, and based on a bilateral auction theory, performing preliminary matching of the user and the vehicle-mounted edge calculation node; measuring user experience of a user through three aspects of node availability, node reputation and communication overhead, and pairing the user and the vehicle-mounted edge computing node based on a bilateral auction theory;
and according to the user-vehicle pairing scheme output by the primary matching module, performing sub-channel resource allocation based on linear programming, and outputting a final matching and spectrum resource allocation result.
According to the technical scheme, the task request sent by the user comprises the task size, the QoE value to be achieved, time constraint, sending power, static power, maximum CPU frequency and user coordinates.
According to the technical scheme, the state information of the vehicle comprises the coordinates, the speed, the movement direction, the maximum CPU frequency, the credit and the availability of the vehicle.
According to the technical scheme, the node availability refers to the probability that the node can provide services after receiving the tasks, the node reputation refers to the probability that the node can provide services according to quality after receiving the tasks, and the communication overhead represents the energy consumed by a user for sending and receiving data.
According to the technical scheme, the user and the vehicle respectively send the task request and the state information to the wireless access point through the WiFi network, and the user and the vehicle-mounted edge computing node are in short-distance communication based on the orthogonal frequency division multiple access technology.
In one embodiment of the present invention, it is assumed that the communication radius of the wireless access point is 100m, the number of users is 10, and the number of vehicles is 10. The users and the vehicles are uniformly distributed in a circle with the AP as the center and the radius of 100 m. The user transmitting power is 100mW, and the static power is 10 mW. The maximum CPU frequency of the user is 0.5GHz, and the CPU frequency of the vehicle is 1.2 GHz. The total bandwidth is 5Mb/s and the number of subchannels is 500. The user's coordinates are (-21.67, 47.54), (72.77, 67.66), (20.32, -8.07), (-6.45, -8.38), (-10.87, 1.41), (-4.07, -4.88), (-38.87, -11.21), (-26.96, -35.82), (-35.19, -9.95), (-13.81, -75.09) (unit: m). The QoE thresholds of the users are 0.6255, 0.6543, 0.6865, 0.6945, 0.5384, 0.5277, 0.6392, 0.5187, 0.6050, 0.6060, respectively. The time constraints of the users are 145.55, 74.50, 125.72, 105.75, 81.22, 114.07, 76.36, 110.52, 81.65, 79.59 (units: ms). The coordinates of the vehicle are (16.89, 64.00), (-10.53, -40.26), (7.27, 83.87), (60.93, 56.78), (-20.03, -16.00), (-58.35, 18.90), (-56.27, 14.93), (-28.41, -46.00), (11.05, -86.28), (-15.59, 21.39) (unit: m). The reputation values of the vehicles are 0.9345, 0.8452, 0.9440, 0.9312, 0.8405, 0.8018, 0.9396, 0.9778, 0.8286, 0.9111, respectively. The availability of the vehicles is 0.8561, 0.7546, 0.8294, 0.8518, 0.81486, 0.7972, 0.9472, 0.8531, 0.9630, 0.8438, respectively.
The final output user-vehicle mapping relationship is σ {8,3,0,0,0,0,0, 2,4 }. σ (1) ═ 8 indicates that the user with the sequence number 1 has matched the in-vehicle edge calculation node with the sequence number 8, and σ (3) ═ 0 indicates that the user with the sequence number 3 has not successfully matched the in-vehicle edge calculation node, as follows. The final total user experience is calculated to be 3.0453. When the total bandwidth is 5Mb/s, the allocated bandwidths are as follows: user 1-1.46 Mb/s, user 2-0.6 Mb/s, user 9-1.3 Mb/s, user 10-1.52 Mb/s. The total occupied bandwidth is 4.88Mb/s and less than 5 Mb/s.
It will be understood that modifications and variations can be made by persons skilled in the art in light of the above teachings and all such modifications and variations are intended to be included within the scope of the invention as defined in the appended claims.

Claims (10)

1. A vehicle-mounted edge computing node selection system is characterized by comprising a plurality of users, a wireless access node and a plurality of vehicles;
the plurality of users send task requests to the wireless access point and complete the task unloading process according to the returned matched vehicle information;
each vehicle is used as a vehicle-mounted edge computing node and sends own state information to the wireless access point at regular intervals; waiting for successful matching with the user, executing the corresponding task, and assisting the user to finish task unloading;
the wireless access point receives and processes data sent by the user and the vehicle edge computing node, executes a node selection method, and pairs the user and the vehicle edge computing node;
the wireless access node specifically comprises a communication constraint module, a calculation constraint module, a preliminary matching module and a resource allocation module;
the communication constraint module calculates a time constraint condition according to the mobility of the vehicle and determines the number of sub-channels required by pairing between the user and the vehicle; dividing a channel into a plurality of sub-channels with equal bandwidth through orthogonal frequency division multiple access, wherein the average data transmission rate of each sub-channel of the same link is equal; calculating the maximum communication time length of the vehicle according to the vehicle state information, and taking the smaller of the time constraint of the task and the communication time length of the vehicle as the actual total time length of the task, wherein the total time length comprises the time of the task sending and processing process; calculating task processing time and task sending time according to the size of each task, calculating required data transmission rate, and then calculating actual data transmission rate according to a Shannon formula, noise power and a channel gain matrix between a user and a vehicle; the required data transmission rate is equal to the actual data transmission rate, and the number of sub-channels distributed between each pair of users and the vehicle is obtained;
the calculation constraint module is used for modeling the energy consumed by the processing task and is divided into two cases for modeling according to whether unloading is carried out or not: 1) if not, the task is processed locally; under the condition that the task execution time is just the time constraint of the task, the CPU frequency of the user equipment takes a smaller value between the maximum frequency and the task required frequency according to the dynamic voltage scaling technology; then calculating local consumed energy; 2) if unloading is carried out, the task is processed at the vehicle-mounted edge computing node; calculating energy consumed by the user equipment for transmitting and receiving data and static power; if the data volume of the task result is very small, the consumed energy and time are ignored; calculating the energy consumed by the user equipment when the task is unloaded by combining the sending time;
the preliminary matching module is used for carrying out preliminary matching on the user-vehicle edge computing nodes based on a bilateral auction theory according to the number of the user-vehicle required sub-channels output by the communication constraint module, the energy consumption required by the processing tasks output by the computing constraint module and a user experience evaluation model; measuring user experience of a user through three aspects of node availability, node reputation and communication overhead, and pairing the user and the vehicle-mounted edge computing node based on a bilateral auction theory;
and the resource allocation module is used for allocating sub-channel resources based on linear programming according to the user-vehicle pairing scheme output by the primary matching module and outputting a final matching and spectrum resource allocation result.
2. The system of claim 1, wherein the user-sent task request comprises a task size, a QoE value that should be achieved, a time constraint, a transmit power, a quiescent power, a maximum CPU frequency, and user coordinates.
3. The system of claim 1, wherein the status information of the vehicle includes coordinates of the vehicle, vehicle speed, direction of movement, maximum CPU frequency, reputation and availability.
4. The system of claim 1, wherein the node availability refers to a probability that a node can provide service after receiving a task, the node reputation refers to a probability that a node can provide service according to quality after receiving a task, and the communication overhead represents energy consumed by a user for sending and receiving data.
5. The system of claim 1, wherein the task request and the status information are sent to the wireless access point by the WiFi network between the user and the vehicle, and the user and the vehicle-mounted edge computing node communicate with each other via short-range communication based on orthogonal frequency division multiple access technology.
6. A vehicle-mounted edge computing node selection method is characterized by comprising the following steps:
a plurality of users send task requests to the wireless access point, and the task unloading process of the users is completed according to the returned matched vehicle information;
each vehicle is used as a vehicle-mounted edge computing node, and the plurality of vehicles send own state information to the wireless access point at regular intervals; waiting for successful matching with the user, executing the corresponding task, and assisting the user to finish task unloading;
the wireless access point receives and processes data sent by a user and a vehicle edge computing node, executes a node selection method, pairs the user and the vehicle edge computing node, and specifically comprises the following steps:
calculating a time constraint condition according to the mobility of the vehicle, and determining the number of sub-channels required by pairing between the user and the vehicle; dividing a channel into a plurality of sub-channels with equal bandwidth through orthogonal frequency division multiple access, wherein the average data transmission rate of each sub-channel of the same link is equal; calculating the maximum communication time length of the vehicle according to the vehicle state information, and taking the smaller of the time constraint of the task and the communication time length of the vehicle as the actual total time length of the task, wherein the total time length comprises the time of the task sending and processing process; calculating task processing time and task sending time according to the size of each task, calculating required data transmission rate, and then calculating actual data transmission rate according to a Shannon formula, noise power and a channel gain matrix between a user and a vehicle; the required data transmission rate is equal to the actual data transmission rate, and the number of sub-channels distributed between each pair of users and the vehicle is obtained;
modeling energy consumed by processing tasks, and modeling under two conditions according to whether unloading is carried out: 1) if not, the task is processed locally; under the condition that the task execution time is just the time constraint of the task, the CPU frequency of the user equipment takes a smaller value between the maximum frequency and the task required frequency according to the dynamic voltage scaling technology; then calculating local consumed energy; 2) if unloading is carried out, the task is processed at the vehicle-mounted edge computing node; calculating energy consumed by the user equipment for transmitting and receiving data and static power; if the data volume of the task result is very small, the consumed energy and time are ignored; calculating the energy consumed by the user equipment when the task is unloaded by combining the sending time;
performing preliminary matching of user-vehicle edge computing nodes based on a bilateral auction theory according to the number of sub-channels distributed between each pair of users and vehicles, energy consumption required by processing tasks and a constructed user experience evaluation model; measuring user experience of a user through three aspects of node availability, node reputation and communication overhead, and pairing the user and the vehicle-mounted edge computing node based on a bilateral auction theory;
and according to the result of pairing the user and the vehicle-mounted edge computing node, performing sub-channel resource allocation based on linear programming, and outputting a final matching and spectrum resource allocation result.
7. The method of claim 6, wherein the user-sent task request comprises a task size, a QoE value that should be achieved, a time constraint, a transmit power, a static power, a maximum CPU frequency, and user coordinates.
8. The method of claim 6, wherein the status information of the vehicle includes coordinates of the vehicle, vehicle speed, direction of movement, maximum CPU frequency, reputation and availability.
9. The method of claim 6, wherein the node availability refers to a probability that the node can provide service after receiving the task, the node reputation refers to a probability that the node can provide service according to quality after receiving the task, and the communication overhead represents energy consumed by a user for sending and receiving data.
10. The method of claim 6, wherein the task request and the status information are sent to the wireless access point by the WiFi network between the user and the vehicle, and the user and the vehicle-mounted edge computing node communicate via short-range communication based on orthogonal frequency division multiple access technology.
CN201710434272.6A 2017-06-09 2017-06-09 Vehicle-mounted edge computing node selection system and method Active CN107317700B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201710434272.6A CN107317700B (en) 2017-06-09 2017-06-09 Vehicle-mounted edge computing node selection system and method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201710434272.6A CN107317700B (en) 2017-06-09 2017-06-09 Vehicle-mounted edge computing node selection system and method

Publications (2)

Publication Number Publication Date
CN107317700A CN107317700A (en) 2017-11-03
CN107317700B true CN107317700B (en) 2020-06-30

Family

ID=60183978

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201710434272.6A Active CN107317700B (en) 2017-06-09 2017-06-09 Vehicle-mounted edge computing node selection system and method

Country Status (1)

Country Link
CN (1) CN107317700B (en)

Families Citing this family (25)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108415763B (en) * 2018-02-11 2020-12-11 中南大学 Distribution method of edge computing system
CN110308716B (en) * 2018-03-27 2022-02-25 上海汽车集团股份有限公司 Cluster-based method and device for automatically driving vehicle and vehicle
CN108809695B (en) * 2018-04-28 2021-07-06 国网浙江省电力有限公司电力科学研究院 Distributed uplink unloading strategy facing mobile edge calculation
CN108777852B (en) * 2018-05-16 2021-07-09 国网吉林省电力有限公司信息通信公司 Internet of vehicles content edge unloading method and mobile resource distribution system
CN108921437B (en) * 2018-07-10 2020-10-09 电子科技大学 Multi-vehicle multi-calculation task scheduling method based on fog calculation
CN109343946B (en) * 2018-09-19 2021-08-13 长安大学 Software-defined Internet of vehicles computing task migration and scheduling method
CN109257432B (en) * 2018-10-12 2021-01-12 桂林电子科技大学 Task switching method, computer device and readable storage medium
CN109511134B (en) * 2018-10-23 2022-03-04 郑州航空工业管理学院 Unmanned aerial vehicle auxiliary wireless communication system load distribution method based on optimal energy efficiency
US11252533B2 (en) * 2018-12-07 2022-02-15 T-Mobile Usa, Inc. UAV supported vehicle-to-vehicle communication
DE102019200924A1 (en) * 2018-12-21 2020-06-25 Volkswagen Aktiengesellschaft Method for operating a decentralized computing network, in particular an edge cloud computer of the decentralized computing network
CN109905859B (en) * 2019-01-14 2021-09-28 南京信息工程大学 Efficient edge computing migration method for Internet of vehicles application
US10527706B1 (en) * 2019-02-19 2020-01-07 Great Wall Motor Company Limited Localization methods and systems for autonomous systems
CN109951873B (en) * 2019-02-28 2020-09-29 华北电力大学 Task unloading mechanism under asymmetric and uncertain information in fog computing of Internet of things
CN110113140B (en) * 2019-03-08 2020-08-11 北京邮电大学 Calculation unloading method in fog calculation wireless network
CN110012507B (en) * 2019-04-02 2021-01-26 华南理工大学 Internet of vehicles resource allocation method and system with priority of user experience
CN110519725B (en) * 2019-08-29 2022-04-22 中国电子科技集团公司第五十四研究所 Communication computing combination-based low-delay transmission optimization method for information of unmanned aerial vehicle group
CN110868455B (en) * 2019-10-29 2021-01-01 北京邮电大学 Computing unloading method and system based on air-space-ground remote Internet of things
CN111355779B (en) * 2020-02-18 2021-05-28 湖南大学 Service-based Internet of vehicles task unloading method and unloading device thereof
CN111405467B (en) * 2020-03-20 2023-10-13 腾讯科技（深圳）有限公司 Vehicle communication method, vehicle-road cooperation system, related device, and readable medium
CN111462487B (en) * 2020-03-31 2021-01-22 长安大学 Optimized edge computing node selection method and system in Internet of vehicles environment
CN113611103B (en) * 2020-06-12 2022-09-16 南京理工大学 Vehicle networking bilateral auction type edge calculation migration method based on vehicle flow prediction
CN111882904B (en) * 2020-07-27 2022-04-22 扬州大学 Port area environment unmanned heavy truck safety early warning method based on edge calculation
CN114095527A (en) * 2020-08-06 2022-02-25 展讯通信（上海）有限公司 Method for enhancing calculation task in Internet of vehicles discovery process and related product
CN113709886B (en) * 2021-08-23 2023-08-18 西安电子科技大学 Dynamic resource allocation method in mobile edge calculation based on matching theory
CN113949706B (en) * 2021-09-17 2022-08-19 中标慧安信息技术股份有限公司 Method and system for selecting mobile edge computing node

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104509090A (en) * 2012-07-27 2015-04-08 歌乐牌株式会社 Vehicle-mounted image recognition device
CN106470449A (en) * 2015-08-14 2017-03-01 电信科学技术研究院 A kind of data transmit-receive, trunking method, device and communication system
WO2017088905A1 (en) * 2015-11-24 2017-06-01 Nec Europe Ltd. A method and network for managing and orchestrating virtual network functions and network applications

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104509090A (en) * 2012-07-27 2015-04-08 歌乐牌株式会社 Vehicle-mounted image recognition device
CN106470449A (en) * 2015-08-14 2017-03-01 电信科学技术研究院 A kind of data transmit-receive, trunking method, device and communication system
WO2017088905A1 (en) * 2015-11-24 2017-06-01 Nec Europe Ltd. A method and network for managing and orchestrating virtual network functions and network applications

Also Published As

Publication number Publication date
CN107317700A (en) 2017-11-03

Similar Documents

Publication Publication Date Title
CN107317700B (en) Vehicle-mounted edge computing node selection system and method
CN111586696B (en) Resource allocation and unloading decision method based on multi-agent architecture reinforcement learning
CN112492626B (en) Method for unloading computing task of mobile user
CN107995660B (en) Joint task scheduling and resource allocation method supporting D2D-edge server unloading
CN109814951B (en) Joint optimization method for task unloading and resource allocation in mobile edge computing network
CN110392079B (en) Node computing task scheduling method and device for fog computing
CN108924796B (en) Resource allocation and unloading proportion joint decision method
CN111819872A (en) Information transmission method, device, communication equipment and storage medium
CN111884696B (en) Relay cooperation mobile edge calculation method based on multiple carriers
CN111132191A (en) Method for unloading, caching and resource allocation of joint tasks of mobile edge computing server
CN113286317B (en) Task scheduling method based on wireless energy supply edge network
CN110545584A (en) Communication processing method of full-duplex mobile edge computing communication system
CN112000481A (en) Task unloading method for maximizing computing capacity of D2D-MEC system
CN111107153A (en) MEC pricing unloading method based on D2D communication in power Internet of things
CN112153728B (en) Optimization method for base station association and module dormancy
CN111200831A (en) Cellular network computing unloading method fusing mobile edge computing
CN111511028B (en) Multi-user resource allocation method, device, system and storage medium
Yao et al. Energy efficient cooperative edge computing with multi-source multi-relay devices
CN109788540A (en) Power control and method for channel allocation in D2D system based on energy acquisition
CN114189521A (en) Method for cooperative computing offload in F-RAN architecture
CN112235387A (en) Multi-node cooperative computing unloading method based on energy consumption minimization
CN113207150B (en) Active and passive hybrid unloading method and device based on backscatter communication
Luo et al. Qoe-driven computation offloading: Performance analysis and adaptive method
CN116367310A (en) Maximum gain oriented channel allocation method for mobile edge calculation
Du et al. Computation energy efficiency maximization for NOMA-based and wireless-powered mobile edge computing with backscatter communication

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