CN113676917B - Game theory-based energy consumption optimization method for unmanned aerial vehicle hierarchical mobile edge computing network - Google Patents

Game theory-based energy consumption optimization method for unmanned aerial vehicle hierarchical mobile edge computing network Download PDF

Info

Publication number
CN113676917B
CN113676917B CN202110978441.9A CN202110978441A CN113676917B CN 113676917 B CN113676917 B CN 113676917B CN 202110978441 A CN202110978441 A CN 202110978441A CN 113676917 B CN113676917 B CN 113676917B
Authority
CN
China
Prior art keywords
alliance
unmanned aerial
federation
head
aerial vehicle
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
CN202110978441.9A
Other languages
Chinese (zh)
Other versions
CN113676917A (en
Inventor
吴启晖
陈佳馨
苏哲
冯斯梦
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.)
Nanjing University of Aeronautics and Astronautics
Original Assignee
Nanjing University of Aeronautics and Astronautics
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 Nanjing University of Aeronautics and Astronautics filed Critical Nanjing University of Aeronautics and Astronautics
Priority to CN202110978441.9A priority Critical patent/CN113676917B/en
Publication of CN113676917A publication Critical patent/CN113676917A/en
Application granted granted Critical
Publication of CN113676917B publication Critical patent/CN113676917B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • 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
    • H04W16/22—Traffic simulation tools or models
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10—Complex mathematical operations
    • G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W52/00—Power management, e.g. TPC [Transmission Power Control], power saving or power classes
    • H04W52/02—Power saving arrangements
    • H04W52/0209—Power saving arrangements in terminal devices
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D30/00—Reducing energy consumption in communication networks
    • Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • General Physics & Mathematics (AREA)
  • Pure & Applied Mathematics (AREA)
  • Mathematical Optimization (AREA)
  • Theoretical Computer Science (AREA)
  • Computational Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Mathematical Analysis (AREA)
  • Evolutionary Biology (AREA)
  • Operations Research (AREA)
  • Probability & Statistics with Applications (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Algebra (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Databases & Information Systems (AREA)
  • Software Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Mobile Radio Communication Systems (AREA)

Abstract

The invention discloses an energy consumption optimization method of an unmanned aerial vehicle hierarchical mobile edge computing network based on game theory, which comprises the following steps: step 1, establishing an unmanned aerial vehicle hierarchical mobile edge calculation scene, analyzing and deducing a cost function when a coalition member carries out local calculation and task unloading, and a cost function when a coalition head is used as a relay and a service provider; step 2, modeling the energy consumption problem of the unmanned aerial vehicle hierarchical mobile edge computing network into a Steinberg game model; and 3, solving the optimal strategy of the upper and lower unmanned aerial vehicles by using a hierarchical iterative learning algorithm based on log-linear-optimal response, so that the lower unmanned aerial vehicle alliance members obtain optimal channel selection, and the upper unmanned aerial vehicle alliance head obtains optimal position selection and role selection. The energy consumption optimization method of the unmanned aerial vehicle hierarchical mobile edge computing network based on the game theory can effectively reduce network consumption and improve the cruising ability of the unmanned aerial vehicle hierarchical mobile edge computing network.

Description

Game theory-based energy consumption optimization method for unmanned aerial vehicle hierarchical mobile edge computing network
Technical Field
The invention relates to an energy consumption optimization method of an unmanned aerial vehicle hierarchical mobile edge computing network based on a game theory, and belongs to the field of resource allocation.
Background
Due to the characteristics of dynamic deployment, high autonomy and the like, the unmanned aerial vehicle plays an extremely important role in the military and civil fields. The unmanned aerial vehicle alliance can complete a series of tasks more flexibly and efficiently, and the application prospect is very wide.
Mobile Edge Computing (MEC), one of the most promising technologies in fifth generation mobile communication networks (5G), can significantly improve latency performance and reduce power consumption of mobile devices by offloading data to be processed to a peripheral resource-rich, higher-performance MEC server. Therefore, the technology is very suitable for the unmanned aerial vehicle alliance with limited energy.
Different from the traditional ground MEC network, the unmanned aerial vehicle alliance network has the following characteristics: the unmanned aerial vehicle is miniaturized, so that the performance of the alliance member is weak, the data processing cannot be independently completed, the high performance of the alliance head can help the alliance member to quickly and efficiently complete the data processing, and therefore the layered MEC network of the unmanned aerial vehicle has the characteristic of multiple servers and multiple terminals. In addition, in a conventional terrestrial MEC network, the MEC server is located in a base station or an access point and thus cannot be moved. The flight dynamics of the drone can be used to optimize the channel quality between the MEC server and the terminal. Therefore, the energy consumption optimization method in the unmanned aerial vehicle hierarchical mobile edge computing network becomes an urgent problem to be solved in the resource allocation of the unmanned aerial vehicle cluster.
Disclosure of Invention
Aiming at the defects in the prior art, the invention provides an energy consumption optimization method of an unmanned aerial vehicle hierarchical mobile edge computing network based on a game theory.
In order to achieve the purpose, the invention adopts the following technical scheme:
an energy consumption optimization method of an unmanned aerial vehicle hierarchical mobile edge computing network based on game theory comprises the following steps:
step 1, establishing an unmanned aerial vehicle layered mobile edge calculation scene, wherein the scene comprises two layers, the upper layer is an unmanned aerial vehicle alliance head, the lower layer is an unmanned aerial vehicle alliance member, meanwhile, a cost function when the alliance member carries out local calculation and task unloading is analyzed and deduced, and the alliance head is used as a cost function when a relay and a service provider.
Step 2, modeling the energy consumption problem of the unmanned aerial vehicle hierarchical mobile edge computing network into a Steinberg game model, wherein a game leader is an upper-layer unmanned aerial vehicle alliance head, and a game follower is a lower-layer unmanned aerial vehicle alliance member;
and 3, solving the optimal strategy of the upper and lower unmanned aerial vehicles by using a hierarchical iterative learning algorithm based on logarithmic linearity-optimal response, so that the members of the lower unmanned aerial vehicle alliance obtain optimal channel selection, and the members of the upper unmanned aerial vehicle alliance obtain optimal position selection and role selection.
Further, in step 1, consider a hierarchical mobile edge computing network of drone based on federations, comprising N drone federation headers and Q drone federation members,
for coalition member mn,kWhose coordinates are expressed as
Figure BDA0003228192360000021
Wherein
Figure BDA0003228192360000022
Respectively represent the coalition members mn,kThe abscissa, the ordinate and the vertical distance from the horizontal ground in three-dimensional space, define
Figure BDA0003228192360000023
Is a member m of the federationn,kChannel selection, wherein
Figure BDA0003228192360000024
Figure BDA0003228192360000025
Is a set of available channels in the network, if
Figure BDA0003228192360000026
Then represents member mn,kPerform local calculation if
Figure BDA0003228192360000027
Then represents member mn,kThrough the channel
Figure BDA0003228192360000028
Carrying out data unloading; for federation header hnIts policy is defined as combining role and location selection
Figure BDA0003228192360000029
Wherein
Figure BDA00032281923600000210
For the role selection at the federation head, relay represents the relay role, server represents the facilitator role,
Figure BDA00032281923600000211
for location selection of federation header, wherein
Figure BDA00032281923600000212
Respectively represent federation headers hnThe abscissa, the ordinate and the vertical distance from the horizontal ground in three-dimensional space; the energy consumption of local computation performed by the coalition members is defined as follows:
Figure BDA00032281923600000213
wherein the content of the first and second substances,
Figure BDA00032281923600000214
is a member m of the federationn,kThe effective switched capacitance parameter associated with the chip structure,
Figure BDA00032281923600000215
indicating the number of required CPU operations revolution,
Figure BDA00032281923600000216
is a member m of the federationn,kThe computing power of (a);
if federate member mn,kSelecting on a channel
Figure BDA00032281923600000217
To its alliance head hnData offload, assume mn,kAt constant transmission power
Figure BDA00032281923600000218
The data is unloaded, then its transmission rate
Figure BDA00032281923600000219
Comprises the following steps:
Figure BDA00032281923600000220
wherein, BlowerIndicating the channel bandwidth that the member uses for task offloading,
Figure BDA00032281923600000221
and piRespectively represent a member mn,kAnd the transmission power of the member i, aiIndicating the channel selection of member i, N0Is the background noise that is the noise of the background,
Figure BDA00032281923600000222
represents a member mn,kTo its federation head hnTaking into account a free space propagation model, i.e.
Figure BDA00032281923600000223
Wherein the content of the first and second substances,
Figure BDA00032281923600000224
is member mn,kTo its federation head hnα is a path loss factor; in the same way, the method for preparing the composite material,
Figure BDA00032281923600000225
representing member i to federation head hnChannel gain of, i.e.
Figure BDA00032281923600000226
Wherein xi，yi，ziRespectively representing the abscissa, the ordinate and the vertical distance from the horizontal ground, SET, of the member i in three-dimensional spacememberIs the set of all unmanned aerial vehicle alliance members at the lower layer, therefore alliance member mn,kThe data offload energy consumption is expressed as:
Figure BDA0003228192360000031
wherein the content of the first and second substances,
Figure BDA0003228192360000032
is a member m of the federationn,kThe amount of data that needs to be processed;
if the alliance head hnBeing a service provider, the calculated energy consumption is expressed as:
Figure BDA0003228192360000033
wherein
Figure BDA0003228192360000034
Is a federation header hnThe effective switched capacitance parameter associated with the chip structure,
Figure BDA0003228192360000035
for alliance head hnThe set of all drone federation members managed,
Figure BDA0003228192360000036
for alliance head hnIf the alliance header h is anFor relaying, it offloads the data to the location (x) againcenter,ycenter,hcenter) Central drone of (1), wherein xcenter，ycenter，hcenterRespectively representing the abscissa and the ordinate of the central unmanned aerial vehicle in a three-dimensional space and the vertical distance from the horizontal ground; assuming that the channel resources from the alliance head to the central drone are pre-allocated, there is no mutual interference between the alliance heads, given a channel bandwidth BupperAlliance head hnThe rate of relaying data to the central drone is:
Figure BDA0003228192360000037
wherein
Figure BDA0003228192360000038
Representing federation header hnTransmission power, N0Is background noise.
Figure BDA0003228192360000039
Representing federation header hnChannel gain to central drone, wherein
Figure BDA00032281923600000310
For alliance head hnDistance to a central drone; therefore, the transmission energy consumption of the alliance head as a relay is as follows:
Figure BDA00032281923600000311
thus, define federation member mn,kAnd federation header hnThe cost functions of (a) are:
Figure BDA00032281923600000312
Figure BDA0003228192360000041
the optimization objective is to minimize the energy consumption of the federation head and federation members.
Further, step 2 models the energy consumption problem of the unmanned aerial vehicle hierarchical mobile edge computing network as a steinberg game model, which is defined as:
Figure BDA0003228192360000042
therein, SETheadAnd SETmemberRespectively represents a leader upper-layer unmanned aerial vehicle alliance head set and a follower lower-layer unmanned aerial vehicle alliance member set,
Figure BDA0003228192360000043
and
Figure BDA0003228192360000044
respectively represent federation headers hnAnd a federation member mn,kThe set of policies of (a) is,
Figure BDA0003228192360000045
for alliance head hnThe utility function of (a) is determined,
Figure BDA0003228192360000046
is a member m of the federationn,kThe cost function of (2).
Further, the step 3 provides a hierarchical iterative learning algorithm based on log-linear-optimal response, which includes a log-linear probability update rule and an optimal response strategy update rule, and performs hierarchical iterative learning on the alliance head and the alliance members to minimize the energy consumption of the alliance head and the alliance members, and the specific algorithm is as follows:
step 3.1, initialization: each federation leader randomly selects a position in the 1 st iteration of the algorithm
Figure BDA0003228192360000047
And role
Figure BDA0003228192360000048
Figure BDA0003228192360000049
Step 3.2, in each subsequent iteration, randomly selecting one alliance head hnUpdating, and keeping the strategy of other alliance heads unchanged;
step 3.3, the lower layer alliance members select according to the actions of all the alliance heads at present, and update rules by adopting an optimal response strategy to obtain optimal channel selection; the policy update rule is based on the optimal response as follows:
step 3.3.1, all coalition members choose local computation, i.e.
Figure BDA00032281923600000410
Wherein the content of the first and second substances,
Figure BDA00032281923600000411
representing sub-slot coalition member m at 1 st timen,kSelecting a channel of (1);
step 3.3.2, randomly select a coalition member mn,kAnd updating the strategy according to the following rules:
Figure BDA00032281923600000412
wherein the content of the first and second substances,
Figure BDA00032281923600000413
representing a coalition member mn,kChannel selection at sub-slot t +1,
Figure BDA00032281923600000414
indicating the channel selection of the remaining coalition members in the tth sub-slot,
Figure BDA00032281923600000415
is a member m of the federationn,kThe available channel set of (a);
step 3.3.3, if the lower layer network of the stage converges or reaches the maximum sub-time slot times, the optimal channel resource is distributed to each alliance member; otherwise, repeating the step 3.3.2 until the lower layer network converges;
step 3.4, alliance head hnCalculating the cost function in the current situation, and recording as
Figure BDA0003228192360000051
Step 3.5, alliance head hnBy probability
Figure BDA0003228192360000052
Randomly selecting a policy b in its set of available policiesexploreWherein
Figure BDA0003228192360000053
Finger union head hnThe other alliance heads keep the strategy unchanged, the lower layer alliance members obtain the optimal response strategy according to the step 3.3, and the alliance head hnCalculate its cost function, as
Figure BDA0003228192360000054
Step 3.6, alliance head hnUpdating the joint position and role selection according to a log-linear probability updating rule:
Figure BDA0003228192360000055
Figure BDA0003228192360000056
where Pr represents the probability of selection, γ is a learning parameter,
Figure BDA0003228192360000057
representing federation header hnThe policy selection in the k-th iteration,
Figure BDA0003228192360000058
representing federation header hnIn the strategy selection in the (k +1) th iteration, the function exp {. cndot.) is an exponential function.
Step 3.7, when the network converges, the algorithm is ended; otherwise, step 3.2 is repeated until the network converges.
The invention has the beneficial effects that: the invention adopts the idea of game theory, understands the upper layer alliance head and the lower layer alliance member in the unmanned aerial vehicle hierarchical mobile edge computing network as the leader and follower of the game, solves the energy consumption problem of the unmanned aerial vehicle hierarchical mobile edge computing network by the Stanberg game, and aims to minimize the energy consumption of the alliance head and the alliance member. A hierarchical iterative learning algorithm based on log-linear-optimal response is developed to solve the game. The invention can effectively reduce network consumption and improve the cruising ability of the unmanned aerial vehicle mobile edge computing network.
Drawings
FIG. 1 is a schematic view of a hierarchical mobile edge computing network scenario for an unmanned aerial vehicle according to the present invention;
FIG. 2 is a flowchart of a layered iterative learning algorithm based on log-linear-optimal response proposed by the present invention;
FIG. 3(a) is a diagram of experimental simulation results of the relationship between the total energy consumption of the upper-level alliance head and the number of alliance members; FIG. 3(b) is a diagram of the experimental simulation result of the relationship between the total energy consumption of the lower-layer coalition members and the number of coalition members.
Detailed Description
The present invention will now be described in further detail with reference to the accompanying drawings.
In order to make the objects, technical solutions and advantages of the present invention more apparent, embodiments of the present invention are described in further detail below with reference to the accompanying drawings. 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, consider a hierarchical mobile edge computing network of drone based on federations, comprising N drone federation heads and Q drone federation members.
For coalition member mn,kWhose coordinates are expressed as
Figure BDA0003228192360000061
Definition of
Figure BDA0003228192360000062
For its channel selection, wherein
Figure BDA0003228192360000063
If it is
Figure BDA0003228192360000064
Then represents member mn,kPerform local calculation if
Figure BDA0003228192360000065
Then represents member mn,kThrough the channel
Figure BDA0003228192360000066
And carrying out data unloading. For federation header hnIts policy is defined as combining role and location selection
Figure BDA0003228192360000067
Wherein
Figure BDA0003228192360000068
The role selection for the federation head is performed,
Figure BDA0003228192360000069
for the location selection of the alliance head, the energy consumption of local computation performed by the alliance members is defined as follows:
Figure BDA00032281923600000610
wherein the content of the first and second substances,
Figure BDA00032281923600000611
is a member m of the federationn,kThe effective switched capacitance parameter associated with the chip structure,
Figure BDA00032281923600000612
indicating the number of required CPU operations revolution,
Figure BDA00032281923600000613
is a member m of the federationn,kThe computing power of (a).
If federate member mn,kSelecting on a channel
Figure BDA00032281923600000614
To its alliance head hnData offload, assume mn,kAt constant transmission power
Figure BDA00032281923600000615
The data is unloaded, then its transmission rate
Figure BDA00032281923600000616
Comprises the following steps:
Figure BDA00032281923600000617
wherein, BlowerIndicating the channel bandwidth that the member uses for task offloading,
Figure BDA00032281923600000618
and piRespectively represent a member mn,kAnd the transmit power of member i. Furthermore, N0Is the background noise that is the noise of the background,
Figure BDA00032281923600000619
represents a member mn,kTo its federation head hnThe channel gain of (1). We consider free space propagation models, i.e.
Figure BDA00032281923600000620
Wherein the content of the first and second substances,
Figure BDA00032281923600000621
is member mn,kTo its federation head hnα is a path loss factor. In the same way, the method for preparing the composite material,
Figure BDA00032281923600000622
representing member i to federation head hnChannel gain of, i.e.
Figure BDA00032281923600000623
SETmemberIs the set of all the unmanned aerial vehicle union members of the lower layer. Thus, federation member mn,kThe data offload energy consumption is expressed as:
Figure BDA00032281923600000624
wherein the content of the first and second substances,
Figure BDA0003228192360000071
is a member m of the federationn,kThe amount of data that needs to be processed.
If the alliance head hnBeing a service provider, the calculated energy consumption is expressed as:
Figure BDA0003228192360000072
wherein
Figure BDA0003228192360000073
Is a federation header hnThe effective switched capacitance parameter associated with the chip structure,
Figure BDA0003228192360000074
as a federation header hnA set of all drone federation members managed. If the alliance head hnFor relaying, it offloads the data to the location (x) againcenter,ycenter,hcenter) The central drone of (1) assumes that channel resources from the alliance head to the central drone are pre-allocated, and there is no mutual interference between the alliance heads. Given channel bandwidth BupperAlliance head hnThe rate of relaying data to the central drone is:
Figure BDA0003228192360000075
wherein
Figure BDA0003228192360000076
Representing federation header hnTransmission power, N0Is background noise.
Figure BDA0003228192360000077
Representing federation header hnChannel gain to central drone, wherein
Figure BDA0003228192360000078
For alliance head hnDistance to a central drone; hence, the federation head acts asThe transmission energy consumption of the relay is as follows:
Figure BDA0003228192360000079
thus, define federation member mn,kAnd federation header hnThe cost functions of (a) are:
Figure BDA00032281923600000710
Figure BDA00032281923600000711
the optimization objective is to minimize the energy consumption of the federation head and federation members.
In the step 2, the energy consumption problem of the unmanned aerial vehicle hierarchical mobile edge computing network is modeled into a Steinberg game model, and the game model is defined as:
Figure BDA00032281923600000712
therein, SETheadAnd SETmemberRespectively representing a leader upper-layer unmanned aerial vehicle alliance head set and a follower lower-layer unmanned aerial vehicle alliance member set.
Figure BDA0003228192360000081
And
Figure BDA0003228192360000082
respectively represent federation headers hnAnd a federation member mn,kThe policy set of (1).
Figure BDA0003228192360000083
For alliance head hnThe utility function of (a) is determined,
Figure BDA0003228192360000084
is a connectionAlly member mn,kThe cost function of (2).
In the step 3, a hierarchical iterative learning algorithm based on a log-linear-optimal response is provided, so that energy consumption of the alliance head and the alliance members is minimized, as shown in fig. 2, the specific algorithm is as follows:
step 3.1, initialization: each federation leader randomly selects a position in the 1 st iteration of the algorithm
Figure BDA0003228192360000085
And role
Figure BDA0003228192360000086
Figure BDA0003228192360000087
Step 3.2, in each subsequent iteration, randomly selecting one alliance head hnUpdating, and keeping the strategy of other alliance heads unchanged;
step 3.3, the lower layer alliance members select according to the actions of all the alliance heads at present, and update rules by adopting an optimal response strategy to obtain optimal channel selection; the policy update rule is based on the optimal response as follows:
step 3.3.1, all coalition members choose local computation, i.e.
Figure BDA0003228192360000088
Wherein the content of the first and second substances,
Figure BDA0003228192360000089
representing sub-slot coalition member m at 1 st timen,kSelecting a channel of (1);
step 3.3.2, randomly select a coalition member mn,kAnd updating the strategy according to the following rules:
Figure BDA00032281923600000810
wherein the content of the first and second substances,
Figure BDA00032281923600000811
representing a coalition member mn,kChannel selection at sub-slot t +1,
Figure BDA00032281923600000812
indicating the channel selection of the remaining coalition members in the tth sub-slot,
Figure BDA00032281923600000813
as a member m of the federationn,kThe available channel set of (a);
step 3.3.3, if the lower layer network of the stage converges or reaches the maximum sub-time slot times, the optimal channel resource is distributed to each alliance member; otherwise, repeating the step 3.3.2 until the lower layer network converges;
step 3.4, alliance head hnCalculating the cost function in the current situation, and recording as
Figure BDA00032281923600000814
Step 3.5, alliance head hnBy probability
Figure BDA00032281923600000815
Randomly selecting a policy b in its set of available policiesexploreWherein
Figure BDA00032281923600000816
Finger union head hnThe other alliance heads keep the strategy unchanged, the lower layer alliance members obtain the optimal response strategy according to the step 3.3, and the alliance head hnCalculate its cost function, as
Figure BDA00032281923600000817
Step 3.6, alliance head hnUpdating the joint position and role selection according to a log-linear probability updating rule:
Figure BDA0003228192360000091
Figure BDA0003228192360000092
where Pr represents the probability of selection, γ is a learning parameter,
Figure BDA0003228192360000093
representing federation header hnThe policy selection in the k-th iteration,
Figure BDA0003228192360000094
representing federation header hnIn the strategy selection in the (k +1) th iteration, the function exp {. cndot.) is an exponential function.
Step 3.7, when the network is converged, the algorithm is ended; otherwise, step 3.2 is repeated until the network converges.
The energy consumption optimization method of the unmanned aerial vehicle hierarchical mobile edge computing network based on the game theory is applied to a specific example, can effectively reduce the energy consumption of the unmanned aerial vehicle hierarchical mobile edge computing network, and is specifically applied as follows:
consider a network containing 5 drone alliances. The number of available channels in the network is 3, and the transmission power of the unmanned aerial vehicle alliance member and the unmanned aerial vehicle alliance head are respectively
Figure BDA0003228192360000095
And
Figure BDA0003228192360000096
in general, the path loss exponent factor α e (2,4), since the communication between drones considers line-of-sight propagation links, for the sake of calculation, the embodiment preferably chooses α to 2, and background noise N0The bandwidth of the lower network and the upper network is B respectivelylower1MHz and Bupper2 MHz. Fig. 3(a) and 3(b) show the results of the experimental runs.
As shown in fig. 3(a) and fig. 3(b), the simulation result diagrams of the upper layer total energy consumption and the lower layer total energy consumption under the application of the optimal steinburg equalization, the worst steinburg equalization, the proposed algorithm and the random selection are shown, and it can be seen that the energy consumption corresponding to the application of the method designed by the present invention is very close to the optimal value and is much smaller than that of the random selection method.
It should be noted that the terms "upper", "lower", "left", "right", "front", "back", etc. used in the present invention are for clarity of description only, and are not intended to limit the scope of the present invention, and the relative relationship between the terms and the terms is not limited by the technical contents of the essential changes.
The above is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above-mentioned embodiments, and all technical solutions belonging to the idea of the present invention belong to the protection scope of the present invention. It should be noted that modifications and embellishments within the scope of the invention may be made by those skilled in the art without departing from the principle of the invention.

Claims (3)

1. An energy consumption optimization method of an unmanned aerial vehicle hierarchical mobile edge computing network based on game theory is characterized by comprising the following steps:
step 1, establishing an unmanned aerial vehicle layered mobile edge calculation scene, wherein the scene comprises two layers, the upper layer is an unmanned aerial vehicle alliance head, the lower layer is an unmanned aerial vehicle alliance member, meanwhile, a cost function when the alliance member carries out local calculation and task unloading is analyzed and deduced, and the alliance head is used as a cost function when a relay and a service provider;
step 2, modeling the energy consumption problem of the unmanned aerial vehicle hierarchical mobile edge computing network into a Steinberg game model, wherein a game leader is an upper-layer unmanned aerial vehicle alliance head, and a game follower is a lower-layer unmanned aerial vehicle alliance member;
step 3, solving the optimal strategy of the upper and lower unmanned aerial vehicles by using a hierarchical iterative learning algorithm based on logarithmic linearity-optimal response, so that the members of the lower unmanned aerial vehicle alliance obtain optimal channel selection, and the members of the upper unmanned aerial vehicle alliance obtain optimal position selection and role selection;
in the step 1, consider a hierarchical mobile edge computing network of unmanned aerial vehicles based on alliance, which comprises N unmanned aerial vehicle alliance heads and Q unmanned aerial vehicle alliance members,
for coalition member mn,kWhose coordinates are expressed as
Figure FDA0003531946130000011
Wherein
Figure FDA0003531946130000012
Respectively represent the coalition members mn,kThe abscissa, the ordinate and the vertical distance from the horizontal ground in three-dimensional space, define
Figure FDA0003531946130000013
Is a member m of the federationn,kChannel selection, wherein
Figure FDA0003531946130000014
Figure FDA0003531946130000015
Is a set of available channels in the network, if
Figure FDA0003531946130000016
Then represents member mn,kPerform local calculation if
Figure FDA0003531946130000017
Then represents member mn,kThrough the channel
Figure FDA0003531946130000018
Carrying out data unloading; for federation header hnIts policy is defined as combining role and location selection
Figure FDA0003531946130000019
Wherein
Figure FDA00035319461300000110
For the role selection of the federation header, relay represents the relay role, server represents the facilitator role,
Figure FDA00035319461300000111
for location selection of federation header, wherein
Figure FDA00035319461300000112
Respectively represent federation headers hnThe abscissa, the ordinate and the vertical distance from the horizontal ground in three-dimensional space; the energy consumption of local computation performed by the coalition members is defined as follows:
Figure FDA00035319461300000113
wherein the content of the first and second substances,
Figure FDA00035319461300000114
is a member m of the federationn,kThe effective switched capacitance parameter associated with the chip structure,
Figure FDA00035319461300000115
indicating the number of required CPU operations revolution,
Figure FDA00035319461300000116
is a member m of the federationn,kThe computing power of (a);
if federate member mn,kSelecting on a channel
Figure FDA00035319461300000117
To its alliance head hnData offload, assume mn,kAt constant transmission power
Figure FDA00035319461300000118
The data is unloaded, then its transmission rate
Figure FDA00035319461300000119
Comprises the following steps:
Figure FDA0003531946130000021
wherein, BlowerIndicating the channel bandwidth that the member uses for task offloading,
Figure FDA0003531946130000022
and piRespectively represent a member mn,kAnd the transmission power of the member i, aiIndicating the channel selection of member i, N0Is the background noise that is the noise of the background,
Figure FDA0003531946130000023
represents a member mn,kTo its federation head hnTaking into account a free space propagation model, i.e.
Figure FDA0003531946130000024
Wherein the content of the first and second substances,
Figure FDA0003531946130000025
is member mn,kTo its federation head hnα is a path loss factor; in the same way, the method for preparing the composite material,
Figure FDA0003531946130000026
representing member i to federation head hnChannel gain of, i.e.
Figure FDA0003531946130000027
Wherein xi，yi，ziRespectively representing the abscissa, the ordinate and the vertical distance from the horizontal ground, SET, of the member i in three-dimensional spacememberIs the set of all unmanned aerial vehicle alliance members at the lower layer, therefore alliance member mn,kThe data offload energy consumption is expressed as:
Figure FDA0003531946130000028
wherein the content of the first and second substances,
Figure FDA0003531946130000029
is a member m of the federationn,kThe amount of data that needs to be processed;
if the alliance head hnBeing a service provider, the calculated energy consumption is expressed as:
Figure FDA00035319461300000210
wherein
Figure FDA00035319461300000211
Is a federation header hnThe effective switched capacitance parameter associated with the chip structure,
Figure FDA00035319461300000212
for alliance head hnThe set of all drone federation members managed,
Figure FDA00035319461300000213
for alliance head hnIf the alliance header h is anFor relaying, it offloads the data to the location (x) againcenter,ycenter,hcenter) Central drone of (1), wherein xcenter，ycenter，hcenterRespectively representing the abscissa and the ordinate of the central unmanned aerial vehicle in a three-dimensional space and the vertical distance from the horizontal ground; assuming that the channel resources from the alliance head to the central drone are pre-allocated, there is no mutual interference between the alliance heads, given a channel bandwidth BupperAlliance head hnThe rate of relaying data to the central drone is:
Figure FDA0003531946130000031
wherein
Figure FDA0003531946130000032
Representing federation header hnTransmission power, N0Is the background noise that is the noise of the background,
Figure FDA0003531946130000033
representing federation header hnChannel gain to central drone, wherein
Figure FDA0003531946130000034
For alliance head hnDistance to a central drone; therefore, the transmission energy consumption of the alliance head as a relay is as follows:
Figure FDA0003531946130000035
thus, define federation member mn,kAnd federation header hnThe cost functions of (a) are:
Figure FDA0003531946130000036
Figure FDA0003531946130000037
the optimization objective is to minimize the energy consumption of the federation head and federation members.
2. The method for optimizing the energy consumption of the hierarchical mobile edge computing network of unmanned aerial vehicles based on the game theory as claimed in claim 1, wherein the step 2 models the energy consumption problem of the hierarchical mobile edge computing network of unmanned aerial vehicles as a steinberg game model defined as:
Figure FDA0003531946130000038
therein, SETheadAnd SETmemberRespectively represents a leader upper-layer unmanned aerial vehicle alliance head set and a follower lower-layer unmanned aerial vehicle alliance member set,
Figure FDA0003531946130000039
and
Figure FDA00035319461300000310
respectively represent federation headers hnAnd a federation member mn,kThe set of policies of (a) is,
Figure FDA00035319461300000311
for alliance head hnThe utility function of (a) is determined,
Figure FDA00035319461300000312
is a member m of the federationn,kThe cost function of (2).
3. The method for optimizing the energy consumption of the unmanned aerial vehicle hierarchical mobile edge computing network based on the game theory as claimed in claim 2, wherein a hierarchical iterative learning algorithm based on a log-linear-optimal response is provided in step 3, and includes a log-linear probability updating rule and an optimal response strategy updating rule, and the hierarchical iterative learning is performed on the alliance head and the alliance members, so that the energy consumption of the alliance head and the alliance members is minimized, and the specific algorithm is as follows:
step 3.1, initialization: each federation leader randomly selects a position in the 1 st iteration of the algorithm
Figure FDA00035319461300000313
And role
Figure FDA00035319461300000314
Figure FDA0003531946130000041
Step 3.2, in each subsequent iteration, randomly selecting one alliance head hnUpdating, and keeping the strategy of other alliance heads unchanged;
step 3.3, the lower layer alliance members select according to the actions of all the alliance heads at present, and update rules by adopting an optimal response strategy to obtain optimal channel selection; the policy update rule is based on the optimal response as follows:
step 3.3.1, all coalition members choose local computation, i.e.
Figure FDA0003531946130000042
Wherein the content of the first and second substances,
Figure FDA0003531946130000043
representing sub-slot coalition member m at 1 st timen,kSelecting a channel of (1);
step 3.3.2, randomly select a coalition member mn,kAnd updating the strategy according to the following rules:
Figure FDA0003531946130000044
wherein the content of the first and second substances,
Figure FDA0003531946130000045
representing a coalition member mn,kChannel selection at sub-slot t +1,
Figure FDA0003531946130000046
indicating the channel selection of the remaining coalition members in the tth sub-slot,
Figure FDA0003531946130000047
is a member m of the federationn,kThe available channel set of (a);
step 3.3.3, if the lower network converges or reaches the maximum sub-time slot times after the step 3.3.2, the optimal channel resource is allocated to each alliance member; otherwise, repeating the step 3.3.2 until the lower layer network converges;
step 3.4, alliance head hnCalculating the cost function in the current situation, and recording as
Figure FDA0003531946130000048
Step 3.5, alliance head hnBy probability
Figure FDA0003531946130000049
Randomly selecting a policy b in its set of available policiesexploreWherein
Figure FDA00035319461300000410
Finger union head hnThe other alliance heads keep the strategy unchanged, the lower layer alliance members obtain the optimal response strategy according to the step 3.3, and the alliance head hnCalculate its cost function, as
Figure FDA00035319461300000411
Step 3.6, alliance head hnUpdating the joint position and role selection according to a log-linear probability updating rule:
Figure FDA00035319461300000412
Figure FDA00035319461300000413
where Pr represents the probability of selection, γ is a learning parameter,
Figure FDA00035319461300000414
representing federation header hnThe policy selection in the k-th iteration,
Figure FDA00035319461300000415
representing federation header hnStrategy selection in the (k +1) th iteration, wherein the function exp {. cndot.) is an exponential function;
step 3.7, when the network converges, the algorithm is ended; otherwise, step 3.2 is repeated until the network converges.
CN202110978441.9A 2021-08-25 2021-08-25 Game theory-based energy consumption optimization method for unmanned aerial vehicle hierarchical mobile edge computing network Active CN113676917B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110978441.9A CN113676917B (en) 2021-08-25 2021-08-25 Game theory-based energy consumption optimization method for unmanned aerial vehicle hierarchical mobile edge computing network

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110978441.9A CN113676917B (en) 2021-08-25 2021-08-25 Game theory-based energy consumption optimization method for unmanned aerial vehicle hierarchical mobile edge computing network

Publications (2)

Publication Number Publication Date
CN113676917A CN113676917A (en) 2021-11-19
CN113676917B true CN113676917B (en) 2022-05-10

Family

ID=78545983

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110978441.9A Active CN113676917B (en) 2021-08-25 2021-08-25 Game theory-based energy consumption optimization method for unmanned aerial vehicle hierarchical mobile edge computing network

Country Status (1)

Country Link
CN (1) CN113676917B (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116319511B (en) * 2022-12-21 2023-11-10 南京航空航天大学 Communication connection method and system based on shortest path tree diagram alliance forming algorithm

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109302463A (en) * 2018-09-17 2019-02-01 上海交通大学 A kind of group cloud framework and optimization method and system certainly towards edge calculations
CN110336861A (en) * 2019-06-18 2019-10-15 西北工业大学 The unloading method for allocating tasks of mobile edge calculations system based on the double-deck unmanned plane
CN111988792A (en) * 2020-08-24 2020-11-24 中国人民解放军陆军工程大学 Unmanned aerial vehicle alliance network unloading model and decision calculation method

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20040030570A1 (en) * 2002-04-22 2004-02-12 Neal Solomon System, methods and apparatus for leader-follower model of mobile robotic system aggregation
CN109495952B (en) * 2018-11-14 2020-04-24 北京航空航天大学 Selection method and device of cellular and unmanned aerial vehicle integrated network
US11703853B2 (en) * 2019-12-03 2023-07-18 University-Industry Cooperation Group Of Kyung Hee University Multiple unmanned aerial vehicles navigation optimization method and multiple unmanned aerial vehicles system using the same
CN111800185A (en) * 2020-07-06 2020-10-20 中国人民解放军陆军工程大学 Distributed air-ground joint deployment method in unmanned aerial vehicle auxiliary communication
CN112118287B (en) * 2020-08-07 2023-01-31 北京工业大学 Network resource optimization scheduling decision method based on alternative direction multiplier algorithm and mobile edge calculation

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109302463A (en) * 2018-09-17 2019-02-01 上海交通大学 A kind of group cloud framework and optimization method and system certainly towards edge calculations
CN110336861A (en) * 2019-06-18 2019-10-15 西北工业大学 The unloading method for allocating tasks of mobile edge calculations system based on the double-deck unmanned plane
CN111988792A (en) * 2020-08-24 2020-11-24 中国人民解放军陆军工程大学 Unmanned aerial vehicle alliance network unloading model and decision calculation method

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
"A Multi-Leader Multi-Follower Stackelberg Game for Coalition-Based UAV MEC Networks";Jiaxin Chen;《IEEE Wireless Communications Letters》;20210727;第10卷(第11期);摘要、第2、3节 *
"移动边缘计算系统中无人机和用户的分层博弈优化方法";崔岩;《通信技术》;20200910;全文 *
"面向异构无人机中继网络的负载均衡:一种分层博弈方法";杨婷婷;《通信技术》;20181110;全文 *

Also Published As

Publication number Publication date
CN113676917A (en) 2021-11-19

Similar Documents

Publication Publication Date Title
CN109814951B (en) Joint optimization method for task unloading and resource allocation in mobile edge computing network
CN112118287B (en) Network resource optimization scheduling decision method based on alternative direction multiplier algorithm and mobile edge calculation
CN112512056B (en) Multi-objective optimization calculation unloading method in mobile edge calculation network
CN111800828B (en) Mobile edge computing resource allocation method for ultra-dense network
CN110798849A (en) Computing resource allocation and task unloading method for ultra-dense network edge computing
CN111586720A (en) Task unloading and resource allocation combined optimization method in multi-cell scene
CN110233755B (en) Computing resource and frequency spectrum resource allocation method for fog computing in Internet of things
CN115022894B (en) Task unloading and computing resource allocation method and system for low-orbit satellite network
CN114885420A (en) User grouping and resource allocation method and device in NOMA-MEC system
CN113613301B (en) Air-ground integrated network intelligent switching method based on DQN
CN113359480A (en) Multi-unmanned aerial vehicle and user cooperative communication optimization method based on MAPPO algorithm
CN113676917B (en) Game theory-based energy consumption optimization method for unmanned aerial vehicle hierarchical mobile edge computing network
CN116112981A (en) Unmanned aerial vehicle task unloading method based on edge calculation
CN111246320B (en) Deep reinforcement learning flow dispersion method in cloud-fog elastic optical network
CN112672371A (en) Air-ground collaborative hierarchical deployment model under heterogeneous demand and access method thereof
CN116321293A (en) Edge computing unloading and resource allocation method based on multi-agent reinforcement learning
CN111796880A (en) Unloading scheduling method for edge cloud computing task
CN113194031A (en) User clustering method and system combining interference suppression in fog wireless access network
CN114640966B (en) Task unloading method based on mobile edge calculation in Internet of vehicles
CN114745386B (en) Neural network segmentation and unloading method in multi-user edge intelligent scene
CN115551013A (en) Unmanned aerial vehicle deployment and task unloading method in multi-unmanned aerial vehicle edge computing network
CN115514405B (en) LEO edge unloading method for joint calculation and communication resource allocation
CN113326112B (en) Multi-unmanned aerial vehicle task unloading and migration method based on block coordinate descent method
He et al. Reinforcement learning based mec architecturewith energy-efficient optimization for arans
Ding et al. UAV-enabled edge computing for virtual reality

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