CN116257335A - Unmanned plane auxiliary MEC system joint task scheduling and motion trail optimization method - Google Patents
Unmanned plane auxiliary MEC system joint task scheduling and motion trail optimization method Download PDFInfo
- Publication number
- CN116257335A CN116257335A CN202211613821.3A CN202211613821A CN116257335A CN 116257335 A CN116257335 A CN 116257335A CN 202211613821 A CN202211613821 A CN 202211613821A CN 116257335 A CN116257335 A CN 116257335A
- Authority
- CN
- China
- Prior art keywords
- aerial vehicle
- unmanned aerial
- task
- computing
- flight
- 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.)
- Pending
Links
- 238000005457 optimization Methods 0.000 title claims abstract description 27
- 230000033001 locomotion Effects 0.000 title claims abstract description 25
- 238000000034 method Methods 0.000 title claims abstract description 23
- 238000012545 processing Methods 0.000 claims abstract description 25
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 19
- 238000004364 calculation method Methods 0.000 claims description 36
- 230000005540 biological transmission Effects 0.000 claims description 25
- 238000005265 energy consumption Methods 0.000 claims description 8
- 230000008569 process Effects 0.000 claims description 7
- 230000009471 action Effects 0.000 claims description 6
- 238000013528 artificial neural network Methods 0.000 claims description 6
- 238000004891 communication Methods 0.000 abstract description 15
- 230000006872 improvement Effects 0.000 abstract description 3
- 238000004088 simulation Methods 0.000 abstract description 3
- 230000006870 function Effects 0.000 description 11
- 238000010586 diagram Methods 0.000 description 3
- 230000002787 reinforcement Effects 0.000 description 3
- 230000004044 response Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000012141 concentrate Substances 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000013468 resource allocation Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 238000000844 transformation Methods 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/48—Program initiating; Program switching, e.g. by interrupt
- G06F9/4806—Task transfer initiation or dispatching
- G06F9/4843—Task transfer initiation or dispatching by program, e.g. task dispatcher, supervisor, operating system
- G06F9/4881—Scheduling strategies for dispatcher, e.g. round robin, multi-level priority queues
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/44—Arrangements for executing specific programs
- G06F9/445—Program loading or initiating
- G06F9/44594—Unloading
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
- G06F9/5027—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals
- G06F9/5038—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals considering the execution order of a plurality of tasks, e.g. taking priority or time dependency constraints into consideration
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2209/00—Indexing scheme relating to G06F9/00
- G06F2209/50—Indexing scheme relating to G06F9/50
- G06F2209/509—Offload
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Software Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Mobile Radio Communication Systems (AREA)
Abstract
The invention discloses a combined task scheduling and motion trail optimization method of an unmanned aerial vehicle auxiliary MEC system, which fully considers the situation that the position of a mobile device changes and a computing task part is unloaded in a dynamic environment (under the condition that a ground terminal device and an unmanned aerial vehicle are both in dynamic movement), and the unmanned aerial vehicle motion parameters (including a flight angle and a flight speed) and power distribution are selected by combining optimization user scheduling and computing task unloading modes, so that communication, computing and motion are combined and optimized, the computing task processing time delay of the terminal device is effectively reduced, and the unloading efficiency of an unmanned aerial vehicle auxiliary edge computing system is improved. And compared with a base line algorithm such as DQN in a simulation experiment, the DDPG algorithm is found to have obvious improvement on the processing time delay.
Description
Technical Field
The invention relates to the technical field of internet of vehicles service privacy protection, in particular to an unmanned aerial vehicle auxiliary MEC system joint task scheduling and motion trail optimization method.
Background
With the development of the fifth generation mobile communication technology, novel applications such as automatic navigation, face recognition, online games and the like, which are computationally intensive and delay sensitive, are rapidly emerging. However, the computing power of the terminal device is often low and it is difficult to handle a large number of computing tasks; traditional cloud computing mode concentrates computing resources on the cloud, and the conditions that the cloud is far away from terminal equipment and the intensive computing task access will lead to larger transmission delay. The mobile edge computing (Mobile Edge Computing, MEC) can conveniently provide computing services to process terminal intensive computing tasks by deploying computing servers at the network edge close to the user side, so that computing processing time delay is effectively reduced, and service experience of the user is improved. Therefore, the occurrence of the mobile edge computing mode greatly relieves the pressure of network bandwidth and cloud computing centers, and optimizes the response capability of computing and storage services.
Existing mobile edge computing servers/computing centers often provide services to users in a fixed deployment. However, in the case of sparse field communication facilities and sudden disasters, it is often difficult for the fixed infrastructure to provide effective services. The unmanned aerial vehicle (Unmanned Aerial Vehicles, UAV) has the characteristics of flexible maneuvering and easy deployment, and can be used for supporting ground communication relay, edge computing service and the like by establishing Line of Sight (LOS) connection with a ground terminal.
In the scheme that the unmanned aerial vehicle is used for supporting ground communication, the unmanned aerial vehicle is kept still and serves as an air base station, and can be applied to scenes such as infrastructure damage or large-scale active communication traffic burst and the like. In addition, by utilizing the computing capability of the unmanned aerial vehicle, the unmanned aerial vehicle can provide computing resources for the ground terminal equipment according to the needs, and the service experience of a user is effectively improved.
Currently, academia and industry are developing research around UAV-assisted MEC systems. In document [1] (t.ren et al., "Enabling Efficient Scheduling in Large-Scale UAV-Assisted Mobile-Edge Computing via Hierarchical Reinforcement Learning," in IEEE Internet of Things Journal, vol.9, no.10, pp.7095-7109,15may15,2022, doi: 10.1109/jiot.2021.3071531.) the scheduling problem is calculated for unmanned aerial vehicle motion planning and ground terminal equipment, the scheduling problem is decomposed into two layers of sub-problems by the authors, and alternate optimization is performed using a hierarchical reinforcement learning algorithm to obtain a real-time scheduling strategy in a dynamic environment.
However, in the prior art, the unmanned aerial vehicle is mainly considered to serve as a communication relay node to provide communication service, or the unmanned aerial vehicle is used as an edge computing node to provide computing service for terminal equipment, and a scheme of cooperation of two service modes is rarely considered. Because the computing capacity of the unmanned aerial vehicle is limited, the situation that the computing task unloading requirement of the terminal equipment cannot be met often occurs.
Disclosure of Invention
Aiming at the problem that the calculation capability of the existing unmanned aerial vehicle is limited and the requirement of terminal equipment on calculation task unloading cannot be met, the invention provides the unmanned aerial vehicle auxiliary MEC system joint task scheduling and motion trail optimization method, which not only considers the situation that the unmanned aerial vehicle is used as an edge calculation node to provide calculation service for users, but also considers the situation that the unmanned aerial vehicle is used as a communication relay node to forward the calculation task to a base station side edge calculation server, thereby carrying out joint optimization on user scheduling, calculation task unloading selection, unmanned aerial vehicle flight parameters and communication resource allocation, effectively reducing the calculation task processing time delay of ground terminal equipment and improving the unloading efficiency of the unmanned aerial vehicle auxiliary edge calculation system.
In order to achieve the above object, the present invention provides the following technical solutions:
the unmanned aerial vehicle auxiliary MEC system joint task scheduling and motion trail optimizing method aims at minimizing and completing all terminal calculation task processing to be an optimizing target, takes unmanned aerial vehicle transmission power, task unloading variables, unmanned aerial vehicle flight speed, unmanned aerial vehicle flight angle, unmanned aerial vehicle self energy, a moving area of the unmanned aerial vehicle and a moving area of terminal equipment as constraints, and adopts a depth certainty strategy gradient algorithm to solve the optimizing target.
Further, the optimization goals and constraints are expressed as follows:
C1:
C2:
C3:
C4:0≤β(n)≤2π
C5:
C6:{x u (n)∈[0,L],y u (n)∈[0,W]}
C7:{x m (n)∈[0,L],y m (n)∈[0,W]}
wherein ,tsum (n) is the total delay to complete the processing of the computing task in time slot T, expressed as:
wherein ,
the transmission delay for completely unloading the calculation tasks to the unmanned aerial vehicle for the ground terminal device is +.>
Transmission delay for unloading calculation task part to base station for unmanned aerial vehicle, t mu (n) calculating task time delay for unmanned aerial vehicle, t mb (n) time required for the base station to deploy MEC server side to complete the calculation task;
c1 is the transmission power P to the unmanned aerial vehicle u (n) aboutBundles, C2 is a variable offloaded to a task
C3 is the constraint on the unmanned aerial vehicle flight speed v u (n) constraint, C4 is constraint on unmanned plane flight angle β (n), C5 is constraint on unmanned plane self energy, and C6 and C7 are limits on unmanned plane and terminal equipment movement area, respectively; e (E) fly ＝φ||v u (n)|| 2 Is the flight energy consumption of the unmanned aerial vehicle, wherein phi=0.5M UAV t fly M is unmanned aerial vehicle mass, t fly For time of flight, E mu (n)＝γ u (f u ) 3 t mu (n)，E mu (n) locally calculating energy consumption for unmanned aerial vehicle, f u Is the calculation capability of unmanned aerial vehicle, gamma u For influencing the CPU processing by the chip structure, E U Is self-charged quantity of the unmanned aerial vehicle.
Further, the process of solving the optimization target by adopting the depth deterministic strategy gradient algorithm is as follows:
the state space is expressed as
Wherein q (n) represents positional information of the unmanned aerial vehicle, p m (n) represents the position information of the ground terminal equipment, D m (n) represents the size of the remaining calculation task data amount,/->
Representing the residual electric quantity of the unmanned aerial vehicle;
the action space is expressed as
wherein vu (n) is the speed of the unmanned aerial vehicle, beta (n) is the flight angle of the unmanned aerial vehicle, and P u (n) is the transmission power of the unmanned aerial vehicle, < >>
Unloading variables for the task;
the bonus function is denoted as r n ＝-t sum (n) wherein the reward function is a negative total delay;
the state-action values, i.e., Q (s, a), are recorded and updated using the Q-table, with updates using the critic and actor networks, as follows:
θ Q' ←ρθ Q +(1-ρ)θ Q'
θ μ' ←ρθ μ +(1-ρ)θ μ'
wherein ,θQ Is a parameter of the critic neural network, ρ is a constant, θ μ Is a parameter of the actor neural network.
Further, the critic network updates as:
L(θ Q )＝E μ' [(y n -Q(s n ,a n |θ Q )) 2 ]
wherein ,Eμ' To calculate the mean value function, y n For Q target value, y n ＝r n +γQ(s n+1 ,μ(s n+1 )|θ Q ) Gamma is the discount factor, r n Is a bonus function.
Further, the Actor network updates as:
wherein ,
for the policy gradient function, μ is the policy network and N is the size randomly extracted from the experience pool.
Compared with the prior art, the invention has the beneficial effects that:
according to the unmanned aerial vehicle auxiliary MEC system joint task scheduling and motion trail optimization method, the situation that the position of mobile equipment changes and the calculation task is partially unloaded in a dynamic environment (under the condition that ground terminal equipment and an unmanned aerial vehicle are in dynamic movement) is fully considered, the unmanned aerial vehicle motion parameters (including flight angles and flight speeds) and power distribution are selected through joint optimization of user scheduling and calculation task unloading modes, communication, calculation and motion are jointly optimized, the calculation task processing time delay of the terminal equipment is effectively reduced, and the unloading efficiency of an unmanned aerial vehicle auxiliary edge computing system is improved. And compared with a base line algorithm such as DQN in a simulation experiment, the DDPG algorithm is found to have obvious improvement on the processing time delay.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings that are needed in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments described in the present invention, and other drawings may be obtained according to these drawings for a person having ordinary skill in the art.
Fig. 1 is a model diagram of an unmanned aerial vehicle auxiliary edge system.
Fig. 2 is a system architecture diagram of an unmanned aerial vehicle auxiliary MEC system joint task scheduling and motion trail optimization method provided by an embodiment of the present invention.
Fig. 3 shows the comparison of DDPG and DQN in terms of processing delay.
Fig. 4 is a comparison of DDPG with Local baseline algorithm and Offload baseline algorithm over processing time delay.
Detailed Description
For a better understanding of the present technical solution, the method of the present invention is described in detail below with reference to the accompanying drawings.
1) System model
Fig. 1 is a diagram of an unmanned aerial vehicle assisted MEC system, comprising an unmanned aerial vehicle, a plurality of ground terminal devices (MDs) and a base station. Consider that a slot can be divided into t= {1,2,..n }. Within time slot n, terminal device m has a position coordinate p m (n)＝[x m (n),y m (n),0] T And at a speed v m Randomly moving; the position coordinates of the unmanned aerial vehicle are
The coordinates of base station b are l b (n)＝[x b (n),y b (n),h] T 。/>
Considering that the drone is deployed in the air, the line-of-sight transmission between the drone and the MDs is dominant, and the channel gain between the drone and the MDs can be expressed as:
wherein ,g0 Representing the channel gain at a reference distance of 1m,
is the distance between the drone and the MDs.
The channel gain between the drone and the base station may be expressed as:
wherein ,
is the distance between the drone and the base station.
Within time slot n, the computing task carried by MD m can be expressed as:
W m (n)＝{D m (n),C m (n),T m (n)}
wherein Dm (n) for calculating the task data size, C m (n) CPU cycles, T, required to process data per bit m (n) is the time that is maximally allowed to process the computational task.
2) Communication model
(1) Communication model between MDs and unmanned aerial vehicle
The MDs completely offloads the calculation tasks to the unmanned aerial vehicle, and the transmission time delay is that
wherein Dm (n) is the size of the data volume of the computing task carried by the MD, < >>
Is the transmission rate between the MD and the drone.
wherein ,Bu Is the available bandwidth for communication between the drone and the MDs, P m Is the transmission power of the MDs,
is white Gaussian noise, P nlos Is the non-line-of-sight transmission loss, f (n) is a binary function, and f (n) ∈ {0,1}. (f (n) =0 indicates no shielding between the drone and MDs, and f (n) =1 indicates shielding between the drone and MDs
Thus, the transmission energy consumption can be expressed as:
(2) Communication model between unmanned aerial vehicle and base station
The transmission delay of the unmanned aerial vehicle for unloading the calculation task part to the base station is as follows:
wherein 
Offloading variables for tasks->
Is the transmission rate between the drone and the base station.
wherein ,Bk Is the available bandwidth for communication between the drone and the base station,
is the transmission power of the unmanned aerial vehicle, +.>
Is the maximum transmission power of the unmanned aerial vehicle, +.>
Is white gaussian noise.
Thus, after a transmission delay between the drone and the base station is obtained, the transmission energy consumption can be expressed as:
(3) Unmanned aerial vehicle mobile model
Within time slot n, the flight of the drone from q (n) to the new position coordinates can be expressed as:
q(n+1)＝[x u (n)+l b (n)cosβ(n),y u (n)+l b (n)sinβ(n)]
wherein ,lb (n)＝v u (n)t fly Distance t for unmanned aerial vehicle to fly fly For time of flight, β (n) is the unmanned plane's flight angle.
Thus, the flight energy consumption of the drone may be expressed as:
E fly ＝φ||v u (n)|| 2
wherein phi=0.5m UAV t fly M is the drone mass.
3) Calculation model
First, unmanned aerial vehicle calculates task time delay as:
wherein fu The calculation power of the unmanned aerial vehicle is calculated in units of turns of the CPU per second.
The energy consumption generated by the UAV calculation task at this time is:
E mu (n)＝γ u (f u ) 3 t mu (n)
wherein ,γu Is an influencing factor of the chip structure to CPU processing.
In addition, the time required for the base station to deploy the MEC server side to complete the calculation task is as follows:
wherein ,fb The unit is the number of turns of the CPU per second, which is the computing power of the base station.
As shown in fig. 2, in the unmanned aerial vehicle auxiliary edge computing system, after a computing task is randomly generated by a terminal and transmitted to the unmanned aerial vehicle, two processing modes are divided: the unmanned aerial vehicle is used as an edge computing node to finish the unloaded computing task processing, or the unmanned aerial vehicle is regarded as relay forwarding to finish the computing task processing together with the edge node (base station).
The invention provides an unmanned aerial vehicle auxiliary MEC system joint task scheduling and motion trail optimization method, which is specifically as follows.
Optimization target:
considering completion of the computational task processing within time slot T, the total latency can be expressed as:
wherein ,
the invention aims at minimizing the completion of all terminal calculation task processing, considers ground terminal equipment and calculation task unloading selection adjustment, UAV motion parameters (including flight angle and flight speed) and transmission power joint optimization, and models the joint optimization problem as follows:
C1:
C2:
C3:
C4:0≤β(n)≤2π
C5:
C6:{x u (n)∈[0,L],y u (n)∈[0,W]}
C7:{x m (n)∈[0,L],y m (n)∈[0,W]}
wherein, C1 is the constraint to unmanned aerial vehicle transmission power, C2 is the constraint to task uninstallation variable, C3 is the constraint to unmanned aerial vehicle flight speed, C4 is the constraint to unmanned aerial vehicle flight angle, C5 is the constraint to unmanned aerial vehicle self energy, C6 and C7 are the restriction to unmanned aerial vehicle and terminal equipment's removal region.
In response to the optimization problem described above, the present invention converts the non-convex optimization problem into a Markov Decision Process (MDP). The channel condition and the equipment state of the system are dynamic and time-varying, and further, the problem is solved by adopting a depth deterministic strategy gradient (DDPG) algorithm, and the algorithm can effectively solve the optimization problem with continuous action space.
The process for solving the optimization target is as follows:
the state space is expressed as
Wherein q (n) represents positional information of the unmanned aerial vehicle, p m (n) represents a ground terminalLocation information of device D m (n) represents the size of the remaining calculation task data amount,/->
Representing the residual electric quantity of the unmanned aerial vehicle;
the action space is expressed as
wherein vu (n) is the speed of the unmanned aerial vehicle, beta (n) is the flight angle of the unmanned aerial vehicle, and P u (n) is the transmission power of the unmanned aerial vehicle, < >>
Unloading variables for the task;
the bonus function is expressed as
Wherein the reward function is a negative total delay.
In the conventional reinforcement learning, Q-learning is widely used, which uses Q-table to record and update state-action values, i.e., Q (s, a), however, Q-learning has difficulty in extracting and generalizing features from previous experiences, and has low effect of dealing with large-dimension problems. DDPG can not only predict Q value by constant training from previous experience, but also update with critic and actor networks. In the DDPG algorithm, the critic network can be updated as:
L(θ Q )＝E u' [(y n -Q(s n ,a n |θ Q )) 2 ]
wherein ,Eμ' To calculate the mean value function, y n For Q target value, y n ＝r n +γQ(s n+1 ,μ(s n+1 )|θ Q ) Gamma is the discount factor, r n Is a bonus function.
The Actor network updates as:
wherein ,
for the purpose of policy gradient function, μ is the policy network.
The DDPG soft update critic and actor target network procedures are expressed as follows:
θ Q' ←ρθ Q +(1-ρ)θ Q'
θ μ' ←ρθ μ +(1-ρ)θ μ'
wherein ,θQ Is a parameter of the neural network, ρ is a constant, θ μ Is a parameter of the actor neural network.
The DDPG algorithm is as follows:
randomly initializing critic and actor networks;
constructing critic and actor target networks;
initializing an experience playback pool;
For episode＝1，M do；
initializing a random noise for exploration;
initial state S 1 ；
For t＝1，T do；
Randomly selecting an action a according to the current policy t ；
After executing the action, the environment feedback rewards r in time t+1 And new state s t+1 ；
Adding transitions(s) t ，a t ，r t+1 ，s t+1 ) To an experience pool;
randomly extracting a small batch of transformations(s) from an experience pool t ，a t ，r t+1 ，s t+1 )；
Calculating Q value based on mean square loss
Updating critic network parameters;
updating an actor network;
and finally, soft updating the critic and actor target networks.
According to the unmanned aerial vehicle auxiliary edge computing system, the situation that the position of the mobile equipment changes and the computing task is partially unloaded in a dynamic environment is fully considered, and the unmanned aerial vehicle motion parameters (including the flight angle and the flight speed) and the power distribution are selected by jointly optimizing the user scheduling and the computing task unloading mode, so that the computing task processing time delay of the terminal equipment is effectively reduced, and the unloading efficiency of the unmanned aerial vehicle auxiliary edge computing system is effectively improved. And compared with a base line algorithm such as DQN in a simulation experiment, the DDPG algorithm is found to have obvious improvement on the processing time delay. The comparison of DDPG and DQN in terms of processing delay (as shown in FIG. 3) shows that DDPG is improved by nearly 20% over DQN. As shown in fig. 4, DDPG is improved by nearly 60% in processing delay compared to the Local baseline algorithm (calculation tasks are only calculated locally), and DDPG is improved by nearly 34% in processing delay compared to the Offload baseline algorithm (calculation tasks are all offloaded to the base station for edge calculation).
The above embodiments are only for illustrating the technical solution of the present invention, and are not limiting; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may be modified or some technical features may be replaced with others, which may not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims (5)
1. The unmanned aerial vehicle auxiliary MEC system joint task scheduling and motion trail optimizing method is characterized in that all terminal calculation task processing is minimized to be completed to be an optimizing target, unmanned aerial vehicle transmission power, task unloading variables, unmanned aerial vehicle flight speed, unmanned aerial vehicle flight angle, unmanned aerial vehicle self energy, a moving area of an unmanned aerial vehicle and a moving area of terminal equipment are taken as constraints, and a depth deterministic strategy gradient algorithm is adopted to solve the optimizing target.
2. The unmanned aerial vehicle assisted MEC system joint task scheduling and motion trajectory optimization method of claim 1, wherein the optimization objectives and constraints are expressed as follows:
C4:0≤β(n)≤2π
C6:{x u (n)∈[0,L],y u (n)∈[0,W]}
C7:{x m (n)∈[0,L],y m (n)∈[0,W]}
wherein ,tsum (n) is the total delay to complete the processing of the computing task in time slot T, expressed as:
wherein ,
offloading all computing tasks to a drone for ground terminal equipmentTransmission delay->
Transmission delay for unloading calculation task part to base station for unmanned aerial vehicle, t mu (n) calculating task time delay for unmanned aerial vehicle, t mb (n) time required for the base station to deploy MEC server side to complete the calculation task;
c1 is the transmission power P to the unmanned aerial vehicle u (n) constraint, C2 is a variable for task offloading
C3 is the constraint on the unmanned aerial vehicle flight speed v u (n) constraint, C4 is constraint on unmanned plane flight angle β (n), C5 is constraint on unmanned plane self energy, and C6 and C7 are limits on unmanned plane and terminal equipment movement area, respectively; e (E) fly ＝φ||v u (n)|| 2 Is the flight energy consumption of the unmanned aerial vehicle, wherein phi=0.5M UAV t fly M is unmanned aerial vehicle mass, t fly For time of flight, E mu (n)＝γ u (f u ) 3 t mu (n)，E mu (n) locally calculating energy consumption for unmanned aerial vehicle, f u Is the calculation capability of unmanned aerial vehicle, gamma u For influencing the CPU processing by the chip structure, E U Is self-charged quantity of the unmanned aerial vehicle.
3. The unmanned aerial vehicle assisted MEC system joint task scheduling and motion trail optimization method according to claim 1, wherein the process of solving the optimization target by adopting the depth deterministic strategy gradient algorithm is as follows:
the state space is expressed as
Wherein q (n) represents positional information of the unmanned aerial vehicle, p m (n) represents the position information of the ground terminal equipment, D m (n) represents the size of the remaining calculation task data amount,
representing the residual electric quantity of the unmanned aerial vehicle;
the action space is expressed as
wherein vu (n) is the speed of the unmanned aerial vehicle, beta (n) is the flight angle of the unmanned aerial vehicle, and P u (n) is the transmission power of the unmanned aerial vehicle, < >>
Unloading variables for the task;
the bonus function is denoted as r n ＝-t sum (n) wherein the reward function is a negative total delay;
the state-action values, i.e., Q (s, a), are recorded and updated using the Q-table, with updates using the critic and actor networks, as follows:
θ Q' ←ρθ Q +(1-ρ)θ Q'
θ μ' ←ρθ μ +(1-ρ)θ μ'
wherein ,θQ Is a parameter of the critic neural network, ρ is a constant, θ μ Is a parameter of the actor neural network.
4. The unmanned aerial vehicle assisted MEC system joint task scheduling and motion trajectory optimization method of claim 3, wherein the critic network updates as:
L(θ Q )＝E μ' [(y n -Q(s n ,a n |θ Q )) 2 ]
wherein ,Eμ' To calculate the mean value function, y n For Q target value, y n ＝r n +γQ(s n+1 ,μ(s n+1 )|θ Q ) Gamma is the discount factor, r n Is a bonus function.
5. The unmanned aerial vehicle assisted MEC system joint task scheduling and motion trail optimization method of claim 3, wherein the Actor network updates as:
wherein ,
for the policy gradient function, μ is the policy network and N is the size randomly extracted from the experience pool. />
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211613821.3A CN116257335A (en) | 2022-12-15 | 2022-12-15 | Unmanned plane auxiliary MEC system joint task scheduling and motion trail optimization method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211613821.3A CN116257335A (en) | 2022-12-15 | 2022-12-15 | Unmanned plane auxiliary MEC system joint task scheduling and motion trail optimization method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116257335A true CN116257335A (en) | 2023-06-13 |
Family
ID=86685290
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211613821.3A Pending CN116257335A (en) | 2022-12-15 | 2022-12-15 | Unmanned plane auxiliary MEC system joint task scheduling and motion trail optimization method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116257335A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117354759A (en) * | 2023-12-06 | 2024-01-05 | 吉林大学 | Task unloading and charging scheduling combined optimization method for multi-unmanned aerial vehicle auxiliary MEC |
| CN117376985A (en) * | 2023-12-08 | 2024-01-09 | 吉林大学 | Energy efficiency optimization method for multi-unmanned aerial vehicle auxiliary MEC task unloading under rice channel |
-
2022
- 2022-12-15 CN CN202211613821.3A patent/CN116257335A/en active Pending
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117354759A (en) * | 2023-12-06 | 2024-01-05 | 吉林大学 | Task unloading and charging scheduling combined optimization method for multi-unmanned aerial vehicle auxiliary MEC |
| CN117354759B (en) * | 2023-12-06 | 2024-03-19 | 吉林大学 | Task unloading and charging scheduling combined optimization method for multi-unmanned aerial vehicle auxiliary MEC |
| CN117376985A (en) * | 2023-12-08 | 2024-01-09 | 吉林大学 | Energy efficiency optimization method for multi-unmanned aerial vehicle auxiliary MEC task unloading under rice channel |
| CN117376985B (en) * | 2023-12-08 | 2024-03-19 | 吉林大学 | Energy efficiency optimization method for multi-unmanned aerial vehicle auxiliary MEC task unloading under rice channel |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111787509B (en) | Unmanned aerial vehicle task unloading method and system based on reinforcement learning in edge calculation | |
| CN116257335A (en) | Unmanned plane auxiliary MEC system joint task scheduling and motion trail optimization method | |
| CN114125708B (en) | Unmanned aerial vehicle cluster trajectory optimization and task unloading method based on digital twinning | |
| CN112399375B (en) | Unmanned aerial vehicle auxiliary edge computing unloading method based on terminal energy efficiency optimization | |
| CN115640131A (en) | Unmanned aerial vehicle auxiliary computing migration method based on depth certainty strategy gradient | |
| CN115659803A (en) | Intelligent unloading method for computing tasks under unmanned aerial vehicle twin network mapping error condition | |
| CN112911618B (en) | Unmanned aerial vehicle server task unloading scheduling method based on resource exit scene | |
| WO2022242468A1 (en) | Task offloading method and apparatus, scheduling optimization method and apparatus, electronic device, and storage medium | |
| CN114626298A (en) | State updating method for efficient caching and task unloading in unmanned aerial vehicle-assisted Internet of vehicles | |
| CN113282352A (en) | Energy-saving unloading method based on multi-unmanned aerial vehicle cooperative auxiliary edge calculation | |
| CN114521002A (en) | Edge computing method for cloud edge and end cooperation | |
| CN117858015A (en) | Air edge calculation data safe transmission and resource allocation method based on deep reinforcement learning | |
| CN117580105A (en) | Unmanned aerial vehicle task unloading optimization method for power grid inspection | |
| CN118034809A (en) | Unmanned aerial vehicle task unloading method and system based on edge computing scene | |
| CN114363803A (en) | Energy-saving multi-task allocation method and system for mobile edge computing network | |
| CN116737391A (en) | Edge computing cooperation method based on mixing strategy in federal mode | |
| CN116419325A (en) | Task unloading, resource allocation and track planning method and system for collaborative calculation of multiple unmanned aerial vehicles | |
| CN114928893B (en) | Architecture based on intelligent reflecting surface and task unloading method | |
| CN115967430A (en) | Cost-optimal air-ground network task unloading method based on deep reinforcement learning | |
| CN114531193B (en) | Bridge state monitoring method based on unmanned aerial vehicle cellular topology networking and mobile edge calculation | |
| CN113950059B (en) | Unmanned aerial vehicle relay auxiliary user task unloading method and system | |
| CN116192228A (en) | Task unloading method and system based on game theory under space-sky-ground integrated network | |
| CN116208968A (en) | Track planning method and device based on federal learning | |
| Dhiman et al. | Computation offloading in mobile edge computing for next generation networks: A deep reinforcement learning approach | |
| CN114513814A (en) | Edge network computing resource dynamic optimization method based on unmanned aerial vehicle auxiliary node |
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 |