CN111884829B - Method for maximizing profit of multi-unmanned aerial vehicle architecture - Google Patents

Method for maximizing profit of multi-unmanned aerial vehicle architecture Download PDF

Info

Publication number
CN111884829B
CN111884829B CN202010566129.4A CN202010566129A CN111884829B CN 111884829 B CN111884829 B CN 111884829B CN 202010566129 A CN202010566129 A CN 202010566129A CN 111884829 B CN111884829 B CN 111884829B
Authority
CN
China
Prior art keywords
things
mobile edge
internet
nodes
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
CN202010566129.4A
Other languages
Chinese (zh)
Other versions
CN111884829A (en
Inventor
任智源
郑书亚
程文驰
王晨
陈晨
张海林
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.)
Xidian University
Original Assignee
Xidian 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 Xidian University filed Critical Xidian University
Priority to CN202010566129.4A priority Critical patent/CN111884829B/en
Publication of CN111884829A publication Critical patent/CN111884829A/en
Application granted granted Critical
Publication of CN111884829B publication Critical patent/CN111884829B/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
    • 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
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/004—Artificial life, i.e. computing arrangements simulating life
    • G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04B—TRANSMISSION
    • H04B7/00—Radio transmission systems, i.e. using radiation field
    • H04B7/14—Relay systems
    • H04B7/15—Active relay systems
    • H04B7/185—Space-based or airborne stations; Stations for satellite systems
    • H04B7/18502—Airborne stations
    • H04B7/18506—Communications with or from aircraft, i.e. aeronautical mobile service
    • 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
    • H04L67/10—Protocols in which an application is distributed across nodes in the network
    • 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
    • H04L67/50—Network services
    • H04L67/56—Provisioning of proxy services
    • H04L67/568—Storing data temporarily at an intermediate stage, e.g. caching
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W28/00—Network traffic management; Network resource management
    • H04W28/02—Traffic management, e.g. flow control or congestion control
    • H04W28/10—Flow control between communication endpoints
    • H04W28/14—Flow control between communication endpoints using intermediate storage
    • 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
    • Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
    • Y02T10/00—Road transport of goods or passengers
    • Y02T10/10—Internal combustion engine [ICE] based vehicles
    • Y02T10/40—Engine management systems

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • General Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Theoretical Computer Science (AREA)
  • Mathematical Physics (AREA)
  • Mathematical Analysis (AREA)
  • General Engineering & Computer Science (AREA)
  • Pure & Applied Mathematics (AREA)
  • Mathematical Optimization (AREA)
  • Computational Mathematics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Software Systems (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Health & Medical Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Computing Systems (AREA)
  • Molecular Biology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Computational Linguistics (AREA)
  • General Health & Medical Sciences (AREA)
  • Evolutionary Biology (AREA)
  • Biomedical Technology (AREA)
  • Evolutionary Computation (AREA)
  • Operations Research (AREA)
  • Probability & Statistics with Applications (AREA)
  • Biophysics (AREA)
  • Algebra (AREA)
  • Databases & Information Systems (AREA)
  • Astronomy & Astrophysics (AREA)
  • Aviation & Aerospace Engineering (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)
  • Mobile Radio Communication Systems (AREA)

Abstract

The invention relates to the technical field of Internet of things, in particular to a method for maximizing multi-unmanned aerial vehicle architecture profit.

Description

Method for maximizing multi-unmanned aerial vehicle architecture income
Technical Field
The invention relates to the technical field of Internet of things, in particular to a method for maximizing the income of a multi-unmanned aerial vehicle architecture.
Background
With the diversity and complexity of the development of the internet of things, the computing requirements of tasks on the internet of things reach an unprecedented level. However, due to limited computing and battery capabilities of the internet of things, it is difficult to process locally computation-intensive tasks, and due to low cost, high flexibility and air-to-ground visible communication channel connection of the unmanned aerial vehicle, the unmanned aerial vehicle has been widely used to provide enhanced information coverage and relay services for the internet of things, the unmanned aerial vehicle flies over an area to be served, each area has multiple internet of things devices, and the unmanned aerial vehicle forwards locally intractable tasks from the internet of things devices to a mobile edge computing server, and performs computing processing through the mobile edge computing server, wherein the mobile edge computing is a computing processing technology that deploys computing resources at the edge of a wireless network to provide nearby services for the internet of things.
However, there is now a lack of research on the revenue generated by the drone-assisted mobile edge computing system, and therefore, it is urgently needed to maximize the net revenue computing model of the drone-assisted mobile edge computing system under the comprehensive consideration of the user experience quality and the operator operation cost.
Disclosure of Invention
In order to solve the technical problems, the invention provides a method for solving the joint optimization problem of maximizing the multi-unmanned aerial vehicle architecture profit by balancing the experience quality of a user and the operation cost of an operator, designing a parameterized net profit model, jointly optimizing communication, calculation and cache resource allocation strategies, maximizing the net profit of unmanned aerial vehicle assisted mobile edge calculation while meeting the user demand, and on the basis, providing a multidimensional hybrid adaptive particle swarm algorithm to solve the joint optimization problem.
The method for maximizing the multi-unmanned aerial vehicle architecture profit comprises the following steps:
step 1, acquiring a set of areas to be served in a UAMEC system area and a set of nodes of the Internet of things in the areas to be served by mobile edge calculation assisted by deployment of an unmanned aerial vehicle, acquiring coordinates of the areas to be served and the nodes of the Internet of things, acquiring the flight height of the unmanned aerial vehicle in the areas to be served and acquiring tasks generated by the nodes of the Internet of things;
Step 2, establishing a communication model according to the data acquired in the step 1;
step 3, establishing a node task total processing time calculation model;
step 4, establishing a user energy consumption calculation model and an operator energy consumption calculation model;
step 5, establishing a net income calculation model of the UAMEC system assisted by the unmanned aerial vehicle;
step 6, jointly optimizing an unloading strategy, a caching strategy, bandwidth allocation and computing resource allocation to maximize net income of an operator, and constructing an objective function of a method for maximizing income of a multi-unmanned aerial vehicle architecture;
in the step 2, the transmission time delay of the calculation result transmitted back to the node of the Internet of things by the mobile edge calculation MEC server is negligible; therefore, if the task is
Figure GDA0003551010780000021
By the drone being offloaded to the mobile edge computing MEC server, it should go through two transmission processes:
1) and (3) first transmission: node of internet of things → unmanned aerial vehicle when unmanned aerial vehicle is convoluted in the center of the area, unmanned aerial vehicle will have the same horizontal coordinate with the center of the area; hence the node
Figure GDA0003551010780000022
With unmanned plane U k Is a distance of
Figure GDA0003551010780000023
Considering the shielding of obstacles, the communication link between the unmanned aerial vehicle and the node of the Internet of things is the probability superposition of sight line LOS and non-sight line NLOS channels; node point
Figure GDA0003551010780000024
With unmanned plane U k Has a path loss of
Figure GDA0003551010780000025
Wherein eta L And η NL Respectively representing attenuation factors corresponding to LOS and NLOS links; c is the speed of light, g is the carrier frequency; node point
Figure GDA0003551010780000031
With unmanned plane U k Has LOS link probability of
Figure GDA0003551010780000032
Wherein δ and ε represent coefficients relating to the environment, and
Figure GDA0003551010780000033
is composed of
Figure GDA0003551010780000034
Thus, a node
Figure GDA0003551010780000035
With unmanned plane U k Has an average path loss of
Figure GDA0003551010780000036
Frequency Division Multiple Access (FDMA) technology is adopted between the nodes of the Internet of things and the unmanned aerial vehicle, and the nodes of the Internet of things in the same area share the bandwidth; the bandwidth between the unmanned aerial vehicle and the node of the Internet of things is assumed to be B U And is and
Figure GDA0003551010780000037
to be allocated to a node
Figure GDA0003551010780000038
The bandwidth ratio of (a); therefore, the bandwidth allocation strategy can be set by
Figure GDA0003551010780000039
Represents; according to Shannon's theorem, nodes
Figure GDA00035510107800000310
With unmanned plane U k Has an average transmission rate of
Figure GDA00035510107800000311
Wherein
Figure GDA00035510107800000312
Represents the transmission power of the nodes of the Internet of things, and sigma 2 Is white gaussian noise;
the bandwidth allocation policy must be satisfied
Figure GDA00035510107800000313
2) And (3) second transmission: UAV → Mobile edge computing MEC Server unmanned aerial vehicle U k Distance from the Mobile edge computing MEC Server is
Figure GDA00035510107800000314
The communication between the drone and the mobile edge computing MEC server is so good that the case of NLOS link can be neglected; similar to (2), unmanned plane U k The path loss with the mobile edge computing MEC server can be expressed as
Figure GDA0003551010780000041
A link between the unmanned aerial vehicle and the MEC adopts a time division multiple access TDMA mode; b is the available spectrum bandwidth between the drone and the mobile edge computing MEC server; unmanned plane U k The average transfer rate with the mobile edge computing MEC server can be calculated as
Figure GDA0003551010780000042
Wherein
Figure GDA0003551010780000043
A transmit power density representative of the drone;
wherein step 3 builds a model based on the processing times of the local and MEC calculation methods;
1) local calculation: defining nodes
Figure GDA0003551010780000044
Has a computing power of
Figure GDA0003551010780000045
Wherein
Figure GDA0003551010780000046
The number of cycles per second; different nodes of the internet of things may have different computing capabilities; local computing task
Figure GDA0003551010780000047
Is calculated as
Figure GDA0003551010780000048
2) And calculating the MEC: for MEC computation, node
Figure GDA0003551010780000049
Should first perform his computational tasks
Figure GDA00035510107800000410
Through unmanned plane U k Off-loading onto a mobile edge computing MEC server, which can then process the task
Figure GDA00035510107800000411
But if the task is
Figure GDA00035510107800000412
The required data are cached on the mobile edge computing MEC server, and the mobile edge computing MEC server can directly process the tasks
Figure GDA00035510107800000413
Namely, the data transmission process is not needed any more; after the cache is deployed, the task
Figure GDA00035510107800000414
An offload delay of
Figure GDA00035510107800000415
The computing resources of the mobile edge computing MEC server can be represented by F (cycles per second), and
Figure GDA00035510107800000416
Representing allocation of Mobile edge computing MEC servers to nodes
Figure GDA00035510107800000417
The computing resource proportion, so the allocation strategy of the computing resource can be controlled by
Figure GDA00035510107800000418
Represents; similar to equation (11), the Mobile edge computing MEC Server compute task
Figure GDA00035510107800000419
Is calculated as
Figure GDA00035510107800000420
Computing resource allocation policy needs to be satisfied
Figure GDA0003551010780000051
In general, tasks
Figure GDA0003551010780000052
General description of the inventionThe treatment time is
Figure GDA0003551010780000053
In order to complete a task within a specified time, the processing latency of the task must be less than the deadline:
Figure GDA0003551010780000054
step 4, respectively establishing an energy consumption model from the perspective of a user (namely, an internet of things node) and the perspective of an operator (namely, a mobile edge computing MEC server and an unmanned aerial vehicle);
1) energy consumption of the user: for the nodes of the Internet of things, the energy consumption is calculated when a task is calculated locally, or the task is unloaded to a mobile edge computing MEC server for transmission;
when task
Figure GDA0003551010780000055
In the local calculation, the energy consumption is
Figure GDA0003551010780000056
Wherein
Figure GDA0003551010780000057
For power calculation, μ is a constant that depends on the average switch capacitance and the average activity factor, and β (β ≧ 2) is a constant (typically close to 3), when the task is
Figure GDA0003551010780000058
When being unloaded to the mobile edge computing MEC server, the energy consumption is
Figure GDA0003551010780000059
2) Energy consumption by the operator: the mobile edge computing MEC server mainly consumes energy when receiving and computing tasks, and the energy consumption of the unmanned aerial vehicle is mainly consumed on forwarding tasks;
Similar to equation (19), when the task is computed by the mobile edge computing MEC server, the energy consumption is
Figure GDA00035510107800000510
Similar to equation (20), the task
Figure GDA00035510107800000511
Slave node
Figure GDA00035510107800000512
When forwarding to the mobile edge computing MEC server, the energy consumption of the MEC and the unmanned aerial vehicle is
Figure GDA00035510107800000513
Figure GDA0003551010780000061
Wherein
Figure GDA0003551010780000062
And
Figure GDA0003551010780000063
representing the received power of the drone and the MEC, respectively;
step 5, researching the income of UAMEC (unmanned aerial vehicle-assisted mobile edge computing) according to the user QoE (quality of experience) of a user and the operation cost OPEX of an operator; the cost for deploying, configuring and maintaining the UAMEC system assisted by the unmanned aerial vehicle is out of the research range of the invention, and the income brought by the system after the system is built is concerned; more specifically, operators provide communication, computing and caching resources to internet of things nodes to obtain profit, while energy consumed by mobile edge computing MEC servers and drones is cost; particularly, the improved QoS is used as a charging standard, and the energy cost consumed by the MEC server and the unmanned aerial vehicle is calculated by the mobile edge to be the operation cost OPEX of an operator;
with the aid of the mobile edge compute MEC server, tasks can be scheduled by the mobile edge compute MEC server
Figure GDA0003551010780000064
Is saved in computing time and energy consumption
Figure GDA0003551010780000065
Figure GDA0003551010780000066
Charging for time and energy that can be saved, and for nodes
Figure GDA0003551010780000067
The unit price charged is respectively
Figure GDA0003551010780000068
And
Figure GDA0003551010780000069
the operator can obtain a revenue of
Figure GDA00035510107800000610
When task
Figure GDA00035510107800000611
When processed by the mobile edge computing MEC server, the mobile edge computing MEC server and the unmanned plane U k Is composed of
Figure GDA00035510107800000612
The unit price of energy is gamma, so the operating cost OPEX of the operator is
Figure GDA00035510107800000613
In summary, the net benefit of the drone-assisted mobile edge computing UAMEC system is
Figure GDA0003551010780000071
Wherein step 6 maximizes the net profit of the operator by jointly optimizing the offload policy O, the cache policy H, the bandwidth allocation R and the computational resource allocation F, the objective function of the present invention is
Figure GDA0003551010780000072
The method for maximizing the multi-unmanned aerial vehicle architecture profit comprises the following steps of 1: defining the kth region to be served as DR k Then all the areas to be served can be defined by the set DR ═ DR 1 ,DR 2 ,...,DR K Represents;
is arranged in the region DR k Having N therein k Individual nodes of the internet of things and DR in regions k The ith node in
Figure GDA0003551010780000073
Indicates, therefore, region DR k All nodes in the cluster can be collected
Figure GDA0003551010780000074
Represents;
when the unmanned aerial vehicle is coiled at the center of the area, the unmanned aerial vehicle can provide forwarding service for nodes in the area, the unmanned aerial vehicle flies according to a preset track, the flying height is H, andand will hover over the region DR k Unmanned aerial vehicle record as U k ；
K is set as an index for the area, {1, 2., K }, and N is set as an index for the area k ＝{1,2,...,N k Is region DR k Index subscripts of the inner internet of things nodes;
using a 3D Euclidean coordinate system and its origin as the coordinates of the base station, the region DR k And node
Figure GDA0003551010780000075
Respectively is (X) k ,Y k 0) and
Figure GDA0003551010780000076
assuming that each node of the internet of things has a task to be executed, the node of the internet of things can execute the task locally or unload the task to a mobile edge computing MEC server for execution; node point
Figure GDA0003551010780000077
The generated task is composed of tuples
Figure GDA0003551010780000078
Is shown in which
Figure GDA0003551010780000079
Representing the size of the input data (in bit),
Figure GDA00035510107800000710
is the computational complexity (in cycles/bit) of the task, and
Figure GDA00035510107800000711
is the task's deadline (in units of s); definition of
Figure GDA00035510107800000712
Is the policy of the offloading of the task,
Figure GDA00035510107800000713
may be assigned a 0 or 1 to indicate a task
Figure GDA00035510107800000714
Whether or not to be offloaded to a mobile edge computing, MEC, server;
the first time certain data is transmitted, the mobile edge computing MEC server may choose whether to store the data; if the data is stored, it can be used in the future without transmission; therefore, the data caching strategy can be controlled by
Figure GDA0003551010780000081
Indicating that the MEC server caches data if the mobile edge calculates
Figure GDA0003551010780000082
Otherwise
Figure GDA0003551010780000083
The method for maximizing the multi-unmanned aerial vehicle architecture profit further comprises the following steps: establishing a cache model if the data is
Figure GDA0003551010780000084
Has been cached by the mobile edge computing MEC server, task
Figure GDA0003551010780000085
Should first be offloaded to the mobile edge computing MEC server, therefore, the decision variables
Figure GDA0003551010780000086
Should satisfy
Figure GDA0003551010780000087
Moreover, since the cache resources of the mobile edge computing MEC server are limited, assuming that the cache resources of the MEC are Ce, in order to ensure that the data cache does not exceed the maximum cache capacity, the cache policy should be satisfied
Figure GDA0003551010780000088
According to the method for maximizing the multi-unmanned aerial vehicle architecture profit, the problem P is a high-dimensional hybrid optimization problem containing discrete and continuous variables, so that a deterministic algorithm is difficult to solve; in order to obtain the optimal solution of the problem P under the limitation, the invention constructs a high-dimensional hybrid self-adaptive particle swarm MHAPSO algorithm; defining MaxI as maximum iteration number, MaxP as total particle number, each particle in the group representing a feasible solution of the problem P, and the position and flight speed of the u-th particle in the t generation
Figure GDA0003551010780000089
Wherein
Figure GDA00035510107800000810
The distribution proportion value for the task unloading of the nodes of the Internet of things,
Figure GDA00035510107800000811
the allocation proportion value of the resources is cached for the nodes of the Internet of things,
Figure GDA00035510107800000812
the allocation proportion value of the node bandwidth resources of the Internet of things,
Figure GDA00035510107800000813
Calculating the allocation proportion value of resources for the nodes of the Internet of things;
Figure GDA00035510107800000814
the iterative update speed of the allocation strategy for the node task offloading of the internet of things,
Figure GDA00035510107800000815
the iterative update speed of the allocation strategy for the node cache resources of the internet of things,
Figure GDA00035510107800000816
the iterative update speed of the allocation strategy of the node bandwidth resources of the Internet of things,
Figure GDA00035510107800000817
calculating the iterative update speed of the allocation strategy of the resources for the nodes of the Internet of things;
estimating a fitness function of a particle location as
Figure GDA0003551010780000091
Wherein
Figure GDA0003551010780000092
Is a feasible region, rho and
Figure GDA0003551010780000093
respectively a penalty factor and a penalty function,
Figure GDA0003551010780000094
is composed of
Figure GDA0003551010780000095
Wherein the content of the first and second substances,
Figure GDA0003551010780000096
represents: ensuring that the total data cache amount does not exceed the maximum cache capacity;
Figure GDA0003551010780000097
represents: ensuring that the proportion sum of the total bandwidth allocated to the nodes of the Internet of things cannot exceed 1;
Figure GDA0003551010780000098
represents: ensuring that the proportion sum of computing resources allocated to the nodes of the Internet of things cannot exceed 1;
Figure GDA0003551010780000099
represents: each task can be ensured to be completed within a specified time;
the velocity update formula of the u-th particle is
Figure GDA00035510107800000910
Wherein
Figure GDA00035510107800000911
Is the inertial weight of the u-th particle in the (t +1) th generation;
Figure GDA00035510107800000912
and
Figure GDA00035510107800000913
is a learning factor; xi 1 And xi 2 Is a random number between (0, 1);
Figure GDA00035510107800000914
and G best (t) represents historical optimality of the u-th particle and global optimality of the t generation;
the weight is updated in a self-adaptive way
Figure GDA0003551010780000101
Wherein, the first and the second end of the pipe are connected with each other,
Figure GDA0003551010780000102
and
Figure GDA0003551010780000103
represents the minimum and average fitness, w, at the t generation max And w min Maximum and minimum inertial weights;
renewal of the particles to
Figure GDA0003551010780000104
Figure GDA0003551010780000105
Figure GDA0003551010780000106
Figure GDA0003551010780000107
Wherein A (: represents each element in the matrix A), ξ 3 And xi 4 Is a random number between (0, 1).
Compared with the prior art, the invention has the beneficial effects that: by balancing the experience quality of a user and the operation cost of an operator, a parameterized net gain model is designed, communication, calculation and cache resource allocation strategies are optimized together, the net gain of unmanned aerial vehicle assisted mobile edge calculation is maximized while the user requirements are met, and on the basis, a multidimensional mixed self-adaptive particle swarm algorithm is provided to solve the joint optimization problem.
Drawings
FIG. 1 is a system architecture diagram of the present invention;
FIG. 2 is a summary table of the necessary parameters;
FIG. 3 is a graph of revenue comparisons between different users;
FIG. 4 is a graph of revenue versus different user densities;
FIG. 5 is a graph of revenue comparisons between different types of tasks;
fig. 6 is a detailed algorithm of the present invention.
Detailed Description
The following detailed description of embodiments of the present invention is provided in connection with the accompanying drawings and examples. The following examples are intended to illustrate the invention but are not intended to limit the scope of the invention.
In order to verify the validity of the mechanism proposed by the present invention, the present invention has conducted a number of experiments. The simulation of the present invention is based on the MATLAB platform and the necessary parameters are summarized in fig. 2. Particularly, the energy unit price is set according to the real-time electricity price of China. The invention considers that there are 9 regions to be served, which are randomly distributed at 500 x 500m 2 Each area of the area is 30 x 30m 2 And 3-6 nodes of the Internet of things are arranged in each area.
Different users have different requirements, for example, users with delay sensitive tasks are concerned about the delay that a drone assisted mobile edge computing UAMEC system can reduce for it, whereas energy limited users are more concerned about the energy consumption pressure that a drone assisted mobile edge computing UAMEC can relieve for it. In order to respond to different requirements, users are divided into 4 types, including all insensitivity, delay insensitivity & energy consumption sensitivity, delay sensitivity & energy consumption insensitivity and all sensitivity. Correspondingly, four charging prices are established, as shown in fig. 3, when the user has higher requirements, the operator can earn more income, which is reasonable from the practical point of view.
As shown in fig. 4, fig. 4 is a graph of net profit at different user densities (i.e., number of users per region), where the average of the input data is 10Mb, and the profit increases dramatically as the user density increases. But when the user density reaches about 20, the rate of increase in revenue is significantly slowed. The main reason is that the unmanned aerial vehicle assisted mobile edge computing UAMEC system has limited resources, and the algorithm automatically rejects some unloading requests generated by the nodes of the Internet of things.
Different computational tasks have different computational complexity, as shown in fig. 5, where fig. 5 is the net gain of different computational tasks, where the mean of the input data is 2 Mb. As computational complexity increases, revenue increases even though users vary. The internet of things nodes tend to offload tasks of high computational complexity to the mobile edge computing MEC server, limited by their computational resources, however the mobile edge computing MEC server also consumes a lot of energy when handling the tasks of high computational complexity, and therefore it is reasonable to charge a higher fee when the user has the tasks of high computational complexity.
In the invention, the net profit of the UAMEC system is calculated by the unmanned aerial vehicle assisted mobile edge based on the user QoE and the operation cost OPEX of the operator. Furthermore, the method is simple. And jointly allocating computing resources, communication resources and cache resources to improve the net benefit of the UAMEC system assisted by the unmanned aerial vehicle. In order to solve the joint optimization problem, an MHAPSO algorithm is designed. The final embodiment shows that MHAPSO can maximize the net benefit of the unmanned aerial vehicle assisted mobile edge computing UAMEC system on the basis of meeting the user requirements.
The above description is only a preferred embodiment of the present invention, and it should be noted that, for those skilled in the art, several modifications and variations can be made without departing from the technical principle of the present invention, and these modifications and variations should also be regarded as the protection scope of the present invention.

Claims (3)

1. A method of maximizing multi-drone architecture revenue, comprising:
step 1, acquiring a set of areas to be served in a UAMEC system area and a set of nodes of the Internet of things in the areas to be served by mobile edge calculation assisted by deployment of an unmanned aerial vehicle, acquiring coordinates of the areas to be served and the nodes of the Internet of things, acquiring the flight height of the unmanned aerial vehicle in the areas to be served and acquiring tasks generated by the nodes of the Internet of things,
setting a total of K regions to be served, and defining the kth region to be served as DR k Then all the areas to be served can be defined by the set DR ═ DR 1 ,DR 2 ,...,DR K Represents;
is arranged in the region DR k Having N therein k Individual nodes of the internet of things and DR in regions k The ith node in
Figure FDA0003595554810000011
Indicates, therefore, region DR k All nodes in the cluster can be collected
Figure FDA0003595554810000012
Represents;
when the unmanned aerial vehicle is coiled at the center of the area, the unmanned aerial vehicle can provide forwarding service for nodes in the area, the unmanned aerial vehicle flies according to a preset track, has the flying height H, and is coiled in the area DR k Unmanned aerial vehicle record as U k ；
K is set as an index for the area, {1, 2., K }, and N is set as an index for the area k ＝{1,2,...,N k Is region DR k Index subscripts of the inner internet of things nodes;
using a 3D Euclidean coordinate system and its origin as the coordinates of the base station, the region DR k And node
Figure FDA0003595554810000013
Respectively is (X) k ,Y k 0) and
Figure FDA0003595554810000014
step 2, establishing a communication model according to the data acquired in the step 1;
step 3, establishing a node task total processing time calculation model;
step 4, establishing a user energy consumption calculation model and an operator energy consumption calculation model;
step 5, establishing a net income calculation model of the UAMEC system assisted by the unmanned aerial vehicle;
step 6, jointly optimizing an unloading strategy, a caching strategy, bandwidth allocation and computing resource allocation to maximize net income of an operator, and constructing an objective function of a method for maximizing income of a multi-unmanned aerial vehicle architecture;
in the step 2, the transmission time delay of the calculation result transmitted back to the node of the Internet of things by the mobile edge calculation MEC server is negligible; from nodes of the internet of things
Figure FDA0003595554810000021
Can be defined as
Figure FDA0003595554810000022
If task
Figure FDA0003595554810000023
By the drone being offloaded to the mobile edge computing MEC server, it should go through two transmission processes:
1) And (3) first transmission: the node of the Internet of things → the unmanned aerial vehicle, when the unmanned aerial vehicle is coiled at the center of the area, the unmanned aerial vehicle has the same horizontal coordinate with the center of the area; region DR k And node
Figure FDA0003595554810000024
Respectively is (X) k ,Y k 0) and
Figure FDA0003595554810000025
hence the node
Figure FDA0003595554810000026
With unmanned plane U k Is a distance of
Figure FDA0003595554810000027
Considering the shielding of obstacles, the communication link between the unmanned aerial vehicle and the node of the Internet of things is the probability superposition of sight line LOS and non-sight line NLOS channels; node point
Figure FDA0003595554810000028
With unmanned plane U k Has a path loss of
Figure FDA0003595554810000029
Wherein eta L And η NL Respectively representing attenuation factors corresponding to LOS and NLOS links; c is the speed of light and g isA carrier frequency; node point
Figure FDA00035955548100000210
With unmanned plane U k Has LOS link probability of
Figure FDA00035955548100000211
Wherein δ and ε represent coefficients relating to the environment, and
Figure FDA00035955548100000212
is composed of
Figure FDA00035955548100000213
Thus, a node
Figure FDA00035955548100000214
With unmanned plane U k Has an average path loss of
Figure FDA00035955548100000215
Frequency Division Multiple Access (FDMA) technology is adopted between the nodes of the Internet of things and the unmanned aerial vehicle, and the nodes of the Internet of things in the same area share the bandwidth; the bandwidth between the unmanned aerial vehicle and the node of the Internet of things is assumed to be B U And is and
Figure FDA00035955548100000216
to be allocated to a node
Figure FDA00035955548100000217
The bandwidth ratio of (a); therefore, the bandwidth allocation strategy can be set by
Figure FDA00035955548100000218
Represents; according to Shannon's theorem, nodes
Figure FDA00035955548100000219
With unmanned plane U k Has an average transmission rate of
Figure FDA0003595554810000031
Wherein
Figure FDA0003595554810000032
Transmit power, and σ, representing nodes of the Internet of things 2 Is Gaussian white noise;
the bandwidth allocation policy must be satisfied
Figure FDA0003595554810000033
Wherein the content of the first and second substances,
Figure FDA0003595554810000034
representing tasks
Figure FDA0003595554810000035
Whether or not to be offloaded to the MEC server,
Figure FDA0003595554810000036
on behalf of the unloading
Figure FDA0003595554810000037
Is not unloaded, but
Figure FDA0003595554810000038
Representing tasks
Figure FDA0003595554810000039
Whether it is cached by the MEC server or not,
Figure FDA00035955548100000310
on behalf of the cache
Figure FDA00035955548100000311
Representing no caching;
2) and (3) second transmission: UAV → Mobile edge computing MEC Server, unmanned aerial vehicle U k Distance from the Mobile edge computing MEC Server is
Figure FDA00035955548100000312
The communication between the drone and the mobile edge computing MEC server is so good that the case of NLOS link can be neglected; unmanned plane U k The path loss with the mobile edge computing MEC server can be expressed as
Figure FDA00035955548100000313
A link between the unmanned aerial vehicle and the MEC adopts a time division multiple access TDMA mode; b is the available spectrum bandwidth between the drone and the mobile edge computing MEC server; unmanned plane U k The average transfer rate with the mobile edge computing MEC server can be calculated as
Figure FDA00035955548100000314
Wherein
Figure FDA00035955548100000315
A transmit power density representative of the drone;
wherein step 3 builds a model based on the processing times of the local and MEC calculation methods;
1) local calculation: defining nodes
Figure FDA0003595554810000041
Has a computing power of
Figure FDA0003595554810000042
Wherein
Figure FDA0003595554810000043
The number of cycles per second; different nodes of the internet of things may have different computing capabilities; local computing task
Figure FDA0003595554810000044
Is calculated as
Figure FDA0003595554810000045
Wherein the content of the first and second substances,
Figure FDA0003595554810000046
representing tasks
Figure FDA0003595554810000047
The amount of data required to be processed, and
Figure FDA0003595554810000048
the computational complexity of the representative data;
2) and calculating the MEC: for MEC computation, node
Figure FDA0003595554810000049
Should first perform his computational tasks
Figure FDA00035955548100000410
Through unmanned plane U k Off-loading onto a mobile edge computing MEC server, which can then process the task
Figure FDA00035955548100000411
But if the task is
Figure FDA00035955548100000412
The required data are cached on the mobile edge computing MEC server, and the mobile edge computing MEC server can directly process the tasks
Figure FDA00035955548100000413
Namely, the data transmission process is not needed any more; after the cache is deployed, the task
Figure FDA00035955548100000414
An offload delay of
Figure FDA00035955548100000415
The computing resource of the mobile edge computing MEC server can be changed by the period number F of each second m Is shown, and
Figure FDA00035955548100000416
representing allocation of Mobile edge computing MEC servers to nodes
Figure FDA00035955548100000417
The computing resource proportion, so the allocation strategy of the computing resource can be controlled by
Figure FDA00035955548100000418
Represents; mobile edge computing MEC server computing tasks
Figure FDA00035955548100000419
Is calculated as
Figure FDA00035955548100000420
Computing resource allocation policy needs to be satisfied
Figure FDA00035955548100000421
In general, tasks
Figure FDA00035955548100000422
The total processing time is
Figure FDA00035955548100000423
In order to complete a task within a specified time, the processing latency of the task must be less than the deadline:
Figure FDA0003595554810000051
here, the number of the first and second electrodes,
Figure FDA0003595554810000052
representing tasks
Figure FDA0003595554810000053
Acceptable maximum processing time;
Step 4, respectively establishing an energy consumption model from the perspective of a user and the perspective of an operator, wherein the user is an internet of things node, and the operator is a mobile edge computing MEC server and an unmanned aerial vehicle;
1) energy consumption of the user: for the nodes of the Internet of things, the energy consumption is calculated when a task is calculated locally, or the task is unloaded to a mobile edge computing MEC server for transmission;
when task
Figure FDA0003595554810000054
In the local calculation, the energy consumption is
Figure FDA0003595554810000055
Wherein
Figure FDA0003595554810000056
For calculating the power, μ is a constant dependent on the average switch capacitance and the average activity factor, β is a constant, and β ≧ 2
Figure FDA0003595554810000057
When being unloaded to the mobile edge computing MEC server, the energy consumption is
Figure FDA0003595554810000058
2) Energy consumption by the operator: the mobile edge computing MEC server mainly consumes energy when receiving and computing tasks, and the energy consumption of the unmanned aerial vehicle is mainly consumed on forwarding tasks;
when the task is calculated by the mobile edge computing MEC server, the energy consumption is
Figure FDA0003595554810000059
Will task
Figure FDA00035955548100000510
Slave node
Figure FDA00035955548100000511
When forwarding to the mobile edge computing MEC server, the energy consumption of the MEC and the unmanned aerial vehicle is
Figure FDA00035955548100000512
Figure FDA00035955548100000513
Wherein
Figure FDA00035955548100000514
And
Figure FDA00035955548100000515
representing the received power of the drone and the MEC, respectively;
wherein step 5 is based on the user quality of experience of the user
QoE and operator's operating cost OPEX have studied the revenue of unmanned aerial vehicle assisted mobile edge computing UAMEC; paying attention to the income brought by the system after the system is built; more specifically, operators provide communication, computing and caching resources to internet of things nodes to obtain profit, while energy consumed by mobile edge computing MEC servers and drones is cost; the improved QoS is used as a charging standard, and the energy cost consumed by the MEC server and the unmanned aerial vehicle is calculated by the mobile edge to be the operation cost OPEX of an operator;
With the aid of a mobile edge compute MEC server, tasks can be transferred by means of the mobile edge compute MEC server
Figure FDA0003595554810000061
Is saved in computing time and energy consumption
Figure FDA0003595554810000062
Figure FDA0003595554810000063
Charging for time and energy that can be saved, and for nodes
Figure FDA0003595554810000064
The unit price charged is respectively
Figure FDA0003595554810000065
And
Figure FDA0003595554810000066
the operator can obtain a revenue of
Figure FDA0003595554810000067
When task
Figure FDA0003595554810000068
When processed by the mobile edge computing MEC server, the mobile edge computing MEC server and the unmanned plane U k Is composed of
Figure FDA0003595554810000069
The unit price of energy is gamma, so the operating cost OPEX of the operator is
Figure FDA00035955548100000610
In summary, the net benefit of the drone-assisted mobile edge computing UAMEC system is
Figure FDA00035955548100000611
Establishing a cache model if the data is
Figure FDA00035955548100000612
Has been cached by the mobile edge computing MEC server, task
Figure FDA00035955548100000613
Should first be offloaded to the mobile edge computing MEC server, therefore, the decision variables
Figure FDA00035955548100000614
Should satisfy
Figure FDA00035955548100000615
Moreover, since the cache resources of the mobile edge computing MEC server are limited, assuming that the cache resources of the MEC are Ce, in order to ensure that the data cache does not exceed the maximum cache capacity, the cache policy should be satisfied
Figure FDA0003595554810000071
Wherein step 6 maximizes the net profit of the operator by jointly optimizing the offload policy O, the cache policy H, the bandwidth allocation R and the computational resource allocation F, with an objective function of
Figure FDA0003595554810000072
2. The method of maximizing multi-drone architecture revenue according to claim 1, characterized by step 1: defining the kth region to be served as DR k Then all the areas to be served can be defined by the set DR ═ DR 1 ,DR 2 ,...,DR K Represents;
is arranged in the region DR k Having N therein k Individual nodes of the internet of things and DR in regions k The ith node in
Figure FDA0003595554810000073
Indicates, therefore, region DR k All nodes in the cluster can be collected
Figure FDA0003595554810000074
Represents;
when the unmanned aerial vehicle is coiled at the center of the area, the unmanned aerial vehicle can provide forwarding service for nodes in the area, the unmanned aerial vehicle flies according to a preset track, has the flying height H, and is coiled in the area DR k Unmanned aerial vehicle record as U k ；
K is set as an index for the area, {1, 2., K }, and N is set as an index for the area k ＝{1,2,...,N k Is region DR k Index subscripts of the inner internet of things nodes;
using a 3D Euclidean coordinate system and its origin as the coordinates of the base station, the region DR k And node
Figure FDA0003595554810000075
Respectively is (X) k ,Y k 0) and
Figure FDA0003595554810000076
assuming that each node of the internet of things has a task to be executed, the node of the internet of things can execute the task locally or unload the task to a mobile edge computing MEC server for execution; node point
Figure FDA0003595554810000077
The generated task is composed of tuples
Figure FDA0003595554810000078
Is shown in which
Figure FDA0003595554810000079
Representing the size of input data, with the unit being bit;
Figure FDA00035955548100000710
Is the computational complexity of the task in units ofcycles/bit; and is provided with
Figure FDA00035955548100000711
The unit is s, which is the deadline of the task; definition of
Figure FDA00035955548100000712
Is the policy of the offloading of the task,
Figure FDA00035955548100000713
may be assigned a 0 or 1 to indicate a task
Figure FDA00035955548100000714
Whether or not to be offloaded to a mobile edge computing, MEC, server;
the first time certain data is transmitted, the mobile edge computing MEC server may choose whether to store the data; if the data is stored, it can be used in the future without transmission; therefore, the data caching strategy can be controlled by
Figure FDA0003595554810000081
Indicating that the MEC server caches data if the mobile edge calculates
Figure FDA0003595554810000082
Otherwise
Figure FDA0003595554810000083
3. The method of maximizing multi-drone architecture revenue of claim 2, wherein problem P is a high-dimensional hybrid optimization problem involving discrete and continuous variables, so deterministic algorithms are difficult to solve; in order to obtain the optimal solution of the problem P under the limitation, a high-dimensional hybrid self-adaptive particle swarm MHAPSO algorithm is constructed; defining MaxI as maximum iteration number, MaxP as total particle number, each particle in the group representing a feasible solution of the problem P, and the position and flight speed of the u-th particle in the t generation
Figure FDA0003595554810000084
Wherein the content of the first and second substances,
Figure FDA0003595554810000085
the distribution proportion value for the task unloading of the nodes of the Internet of things,
Figure FDA0003595554810000086
The allocation proportion value of the resources is cached for the nodes of the Internet of things,
Figure FDA0003595554810000087
the allocation proportion value of the node bandwidth resources of the Internet of things,
Figure FDA0003595554810000088
calculating the allocation proportion value of resources for the nodes of the Internet of things;
Figure FDA0003595554810000089
the iterative update speed of the distribution strategy for the node task unloading of the Internet of things,
Figure FDA00035955548100000810
the iterative update speed of the allocation strategy for the node cache resources of the internet of things,
Figure FDA00035955548100000811
the iterative update speed of the allocation strategy of the node bandwidth resources of the Internet of things,
Figure FDA00035955548100000812
calculating the iterative update speed of the allocation strategy of the resources for the nodes of the Internet of things;
Figure FDA00035955548100000813
representing the number of all internet of things nodes;
estimating a fitness function of a particle location as
Figure FDA00035955548100000814
Wherein
Figure FDA0003595554810000091
Is a feasible region, rho and
Figure FDA0003595554810000092
respectively a penalty factor and a penalty function,
Figure FDA0003595554810000093
is composed of
Figure FDA0003595554810000094
Wherein the content of the first and second substances,
Figure FDA0003595554810000095
represents: ensuring that the total data cache amount does not exceed the maximum cache capacity;
Figure FDA0003595554810000096
represents: ensuring that the proportion sum of the total bandwidth allocated to the nodes of the Internet of things cannot exceed 1;
Figure FDA0003595554810000097
represents: ensuring that the proportion sum of computing resources allocated to the nodes of the Internet of things cannot exceed 1;
Figure FDA0003595554810000098
represents: each task can be ensured to be completed within a specified time;
the velocity update formula of the u-th particle is
Figure FDA0003595554810000099
Wherein
Figure FDA00035955548100000910
Is the inertial weight of the u-th particle in the (t +1) th generation;
Figure FDA00035955548100000911
and
Figure FDA00035955548100000912
is a learning factor; xi shape 1 And xi 2 Is a random number between (0, 1);
Figure FDA00035955548100000913
And G best (t) represents historical optimality of the u-th particle and global optimality of the t generation;
the weight is updated in a self-adaptive way
Figure FDA00035955548100000914
Wherein the content of the first and second substances,
Figure FDA00035955548100000915
and
Figure FDA00035955548100000916
represents the minimum and average fitness, w, at the t generation max And w min Maximum and minimum inertial weights;
renewal of the particles to
Figure FDA0003595554810000101
Figure FDA0003595554810000102
Figure FDA0003595554810000103
Figure FDA0003595554810000104
Wherein A (: represents each element in the matrix A), ξ 3 And xi 4 Is a random number between (0, 1).
CN202010566129.4A 2020-06-19 2020-06-19 Method for maximizing profit of multi-unmanned aerial vehicle architecture Active CN111884829B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010566129.4A CN111884829B (en) 2020-06-19 2020-06-19 Method for maximizing profit of multi-unmanned aerial vehicle architecture

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010566129.4A CN111884829B (en) 2020-06-19 2020-06-19 Method for maximizing profit of multi-unmanned aerial vehicle architecture

Publications (2)

Publication Number Publication Date
CN111884829A CN111884829A (en) 2020-11-03
CN111884829B true CN111884829B (en) 2022-07-29

Family

ID=73157012

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010566129.4A Active CN111884829B (en) 2020-06-19 2020-06-19 Method for maximizing profit of multi-unmanned aerial vehicle architecture

Country Status (1)

Country Link
CN (1) CN111884829B (en)

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112541426B (en) * 2020-12-10 2022-09-30 天津(滨海)人工智能军民融合创新中心 Communication bandwidth self-adaptive data processing method based on unmanned aerial vehicle cluster cooperative sensing
CN112230679B (en) * 2020-12-15 2021-03-09 中国人民解放军国防科技大学 Group coupling system cooperative control method and device based on time delay
CN112866368B (en) * 2021-01-12 2022-03-18 北京邮电大学 Air-ground remote Internet of things design method and system
CN113873467A (en) * 2021-09-26 2021-12-31 北京邮电大学 Unmanned aerial vehicle-assisted mobile edge calculation method and device and control equipment
CN113993175B (en) * 2021-10-25 2023-10-17 盛东如东海上风力发电有限责任公司 Unmanned aerial vehicle communication switching method, system, equipment and storage medium
CN117499158B (en) * 2023-12-25 2024-04-16 天地信息网络研究院(安徽)有限公司 Active defense method based on multi-attacker joint or non-joint attack

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
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
CN111294736A (en) * 2018-12-07 2020-06-16 T移动美国公司 Unmanned aerial vehicle supported vehicle-to-vehicle communication

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111182570B (en) * 2020-01-08 2021-06-22 北京邮电大学 User association and edge computing unloading method for improving utility of operator
CN111163521B (en) * 2020-01-16 2022-05-03 重庆邮电大学 Resource allocation method in distributed heterogeneous environment in mobile edge computing

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111294736A (en) * 2018-12-07 2020-06-16 T移动美国公司 Unmanned aerial vehicle supported vehicle-to-vehicle communication
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

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
Energy-Efficient UAV-Assisted Mobile Edge Computing: Resource Allocation and Trajectory Optimization;Mushu Li等;《IEEE Transactions on Vehicular Technology》;20200331;参见全文 *
Minimization of Offloading Delay for Two-Tier UAV with Mobile Edge Computing;Jingfang Liu等;《2019 15th International Wireless Communications & Mobile Computing Conference (IWCMC)》;20190722;参见全文 *

Also Published As

Publication number Publication date
CN111884829A (en) 2020-11-03

Similar Documents

Publication Publication Date Title
CN111884829B (en) Method for maximizing profit of multi-unmanned aerial vehicle architecture
Huda et al. Survey on computation offloading in UAV-Enabled mobile edge computing
Liu et al. Deep reinforcement learning based latency minimization for mobile edge computing with virtualization in maritime UAV communication network
Yao et al. Online task allocation and flying control in fog-aided internet of drones
Chen et al. VFC-based cooperative UAV computation task offloading for post-disaster rescue
CN113939034A (en) Cloud edge-side cooperative resource allocation method for stereo heterogeneous power Internet of things
CN113452956B (en) Intelligent distribution method and system for power transmission line inspection tasks
Singh et al. Multi-objective NSGA-II optimization framework for UAV path planning in an UAV-assisted WSN
CN112929849B (en) Reliable vehicle-mounted edge calculation unloading method based on reinforcement learning
Fahim et al. An optimized LTE-based technique for drone base station dynamic 3D placement and resource allocation in delay-sensitive M2M networks
Zheng et al. Optimal communication-computing-caching for maximizing revenue in UAV-aided mobile edge computing
Faraci et al. Green edge intelligence for smart management of a fanet in disaster-recovery scenarios
Yao et al. Power control in Internet of Drones by deep reinforcement learning
Fu et al. Toward energy-efficient UAV-assisted wireless networks using an artificial intelligence approach
CN113821346B (en) Edge computing unloading and resource management method based on deep reinforcement learning
Hu et al. Reinforcement learning for energy efficiency improvement in UAV-BS access networks: A knowledge transfer scheme
Shah et al. A compendium of radio resource management in UAV-assisted next generation computing paradigms
Srinivas et al. Delay-tolerant charging scheduling by multiple mobile chargers in wireless sensor network using hybrid GSFO
CN117580180A (en) Communication computing storage multi-domain resource allocation method for end-to-end low-delay information delivery
Yao et al. QoS-aware machine learning task offloading and power control in internet of drones
Hadj et al. A cloud of UAVs for the delivery of a sink as a service to terrestrial WSNs
CN115967430A (en) Cost-optimal air-ground network task unloading method based on deep reinforcement learning
Lhazmir et al. UAV for wireless power transfer in IoT networks: A GMDP approach
Lakew et al. Intelligent Self-Optimization for Task Offloading in LEO-MEC-Assisted Energy-Harvesting-UAV Systems
Chowdhury Superactive: a priority, latency, and SLA-aware resource management scheme for software defined space-air-ground integrated networks

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