CN115665802A - Calculation unloading and resource allocation method based on Lyapunov optimization - Google Patents
Calculation unloading and resource allocation method based on Lyapunov optimization Download PDFInfo
- Publication number
- CN115665802A CN115665802A CN202211347119.7A CN202211347119A CN115665802A CN 115665802 A CN115665802 A CN 115665802A CN 202211347119 A CN202211347119 A CN 202211347119A CN 115665802 A CN115665802 A CN 115665802A
- Authority
- CN
- China
- Prior art keywords
- vehicle
- mobile terminal
- task
- mounted mobile
- total
- 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
- 238000004364 calculation method Methods 0.000 title claims abstract description 128
- 238000005457 optimization Methods 0.000 title claims abstract description 38
- 238000000034 method Methods 0.000 title claims abstract description 26
- 238000013468 resource allocation Methods 0.000 title claims abstract description 18
- 238000005265 energy consumption Methods 0.000 claims abstract description 58
- 238000013508 migration Methods 0.000 claims abstract description 18
- 230000005012 migration Effects 0.000 claims abstract description 18
- 230000005540 biological transmission Effects 0.000 claims description 16
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 9
- 238000004422 calculation algorithm Methods 0.000 claims description 5
- 238000004458 analytical method Methods 0.000 claims description 4
- 238000009795 derivation Methods 0.000 claims description 3
- 238000005562 fading Methods 0.000 claims description 3
- 239000002184 metal Substances 0.000 claims description 3
- 238000012887 quadratic function Methods 0.000 claims description 3
- 238000012545 processing Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 3
- 238000011160 research Methods 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
Images
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a calculation unloading and resource allocation method based on Lyapunov optimization, which is applied to a mobile edge calculation task unloading system. The method determines the weighted sum of the total time delay, the total energy consumption and the total task migration overhead of the mobile edge computing task unloading system as the system overhead, and models the system overhead. Then, the problem of minimizing the total system overhead is solved by using a Lyapunov optimization theory, an unloading decision, the calculation rate of the vehicle-mounted mobile terminal to the task, the transmitting power of the vehicle-mounted mobile terminal and the calculation rate of the mobile edge calculation server to the task are calculated by using an iterative method, the optimal system overhead is obtained, and further the total system overhead is reduced.
Description
Technical Field
The invention belongs to the technical field of mobile edge computing, and particularly relates to a computing unloading and resource allocation method based on Lyapunov optimization.
Background
The 5G provides a highly flexible and extensible network technology, so that everything can be connected with each other, namely everything interconnection is supported. With the continuous development and perfection of the internet of things and wireless communication technology, various novel mobile applications are popular, so that the number of mobile terminals is increased explosively, and a large amount of mobile resources are consumed. As a key technology of 5G, the mobile edge computing server can be regarded as a new architecture, and compared with mobile cloud computing, the mobile edge computing server provides lower time delay and better computing flexibility for task offloading. The mobile terminal task can be sunk to be deployed on the mobile edge computing server for processing, so that the computing time delay and energy consumption are reduced, and the service quality and the user experience quality can be effectively improved.
In the prior art, task migration overhead generated by a vehicle-mounted mobile terminal is difficult to take into account. In a complex mobility scene, the problem analysis is performed based on a one-dimensional random uniform motion or a two-dimensional random uniform motion scene of a user, and a real scene user is generally a moving scene of a two-dimensional non-uniform random motion. In the current research of unloading the edge calculation task, aiming at the condition of a multi-mobile-terminal multi-mobile-edge calculation server, the research of minimizing average processing time delay and processing energy consumption is more perfect, but the additional overhead brought by the dynamic cross-area of the mobile terminal is not considered, so that the calculation deviation is larger.
Disclosure of Invention
The invention aims to solve the problems in the background art and provides a calculation unloading and resource allocation method based on Lyapunov optimization.
In order to achieve the purpose, the technical scheme adopted by the invention is as follows:
the invention provides a calculation unloading and resource allocation method based on Lyapunov optimization, which is applied to a mobile edge calculation task unloading system, wherein the mobile edge calculation task unloading system comprises M vehicle-mounted mobile terminals and N mobile edge calculation servers, and the calculation unloading and resource allocation method based on the Lyapunov optimization comprises the following steps:
calculating the weighted sum of the total time delay, the total energy consumption and the total task migration cost of the task unloading system by using the mobile edge as the system cost, and establishing an optimization model as follows:
wherein,
wherein E is m (t) represents total energy consumption, q m (T) represents the total task migration overhead, T m (t) represents total delay, C (t) represents overhead,
an overhead weighting factor indicating the occurrence of region migration,
a weight factor that represents the total time delay,
weight factor representing the total consumed energy, E total Indicating the total amount of power of the vehicle-mounted mobile terminal, E cur (t) represents a current remaining capacity of the in-vehicle mobile terminal,
the percentage of the current residual capacity of the vehicle-mounted mobile terminal is represented, C1 represents that the calculation rate of the vehicle-mounted mobile terminal to the task cannot be larger than the maximum calculation rate per se
C2 represents that the calculation rate of the mobile edge calculation server to the task cannot be larger than the maximum calculation rate of the mobile edge calculation server
C3 represents that the calculation rate of all tasks in the area of the mobile edge calculation server cannot be larger than the maximum calculation rate of the mobile edge calculation server
C4 represents that the total consumed energy consumption cannot be larger than the current residual electric quantity of the vehicle-mounted mobile terminal, and C5 represents that the total time delay cannot be larger than the maximum tolerant time delay T of the task max And C6 represents that the transmitting power of the vehicle-mounted mobile terminal cannot be larger than the maximum transmitting power of the vehicle-mounted mobile terminal
C7 represents a value range of the unloading decision, C8 represents a value range of a weight factor of the total delay, and T ∈ T = {1,2.
Introducing a delay penalty function phi (t) into the optimization model, and simplifying the optimization model into:
П 2 :
s.t.C2,C3,C7,C8
according to time segments 2 Splitting to obtain the subproblems of the optimization model:
П 3 :
s.t.C2,C3,C7,C8
establishing a virtual queue and a penalty function for energy consumption according to the Lyapunov theory, calculating the drift upper boundary of the Lyapunov, and converting the subproblems of the optimization model into:
Π 4 :
s.t.C2,C3,C7,C8
wherein Q is m (t) represents the energy consumption backlog at time t, and V is a constant control parameter representing the tradeoff between system overhead and virtual queues.
According to II 4 And constructing an unloading decision, a calculation rate of the vehicle-mounted mobile terminal to the task and a mathematical expression of the transmitting power of the vehicle-mounted mobile terminal.
And randomly generating an unloading decision, the calculation rate of the vehicle-mounted mobile terminal to the task, the transmitting power of the vehicle-mounted mobile terminal and an initial solution of the calculation rate of the mobile edge calculation server to the task, then performing iterative calculation by using an iterative algorithm, updating the system overhead after each iteration until the value of the system overhead is converged, and outputting the final system overhead as the optimal system overhead.
Preferably, the total delay is expressed as follows:
T m (t)≤T max ≤τ
wherein,
H m.n (t)＝h m,n (t)g 0 (d 0 /d m,n ) θ
wherein,
represents the execution time delay of the in-vehicle mobile terminal,
representing the execution latency of the mobile edge compute server,
represents the transmission delay, T, of the vehicle-mounted mobile terminal for unloading the task to the mobile edge computing server max Representing the maximum tolerant time delay of each task, tau represents the working time slot of each mobile edge computing server, and the time slot tau is a time segment of the moment t, s m (t) denotes the unload decision at time t, and s m (t) =0 denotes that all tasks are performed at the in-vehicle mobile terminal, s m (t) =1 denotes that the task is completely unloaded to the mobile edge computing server for execution, 0 < s m (t) < 1 indicates that a part of tasks are executed in the vehicle-mounted mobile terminal and another part of tasks are executed in the mobile edge computing server, lambda m (t) represents the calculation task amount generated by downloading the mobile terminal at the moment t, and the unit is bit, c m The CPU period required by the vehicle-mounted mobile terminal to calculate each bit of data is represented, and the unit is cycle/bit, f m (t) represents the calculated rate of the task at time t of the in-vehicle mobile terminal,
representing the computation rate of the mobile edge computation server on the task, r m,n (t) represents the task transmission rate at time t,
representing the transmission power of the mobile terminal loaded at time t, I representing the average interference in each zone, σ 2 Representing the channel background noise, ω the channel bandwidth, H m,n (t) denotes channel gain, g 0 Represents a constant of path loss, and theta represents a road stiffness loss exponent，d 0 Denotes a reference distance, d m,n Indicates the distance from the vehicle-mounted mobile terminal to the mobile edge calculation server in the area, h m,n (t) represents a small-scale Rayleigh fading factor between the vehicle-mounted mobile terminal and the mobile edge computing server in the area where the vehicle-mounted mobile terminal is located,
indicating the connection state of the vehicle-mounted mobile terminal and the mobile edge computing server in the area, and when the connection state is
When the coverage of the mobile edge calculation server in the area of the vehicle-mounted mobile terminal is represented, the vehicle-mounted mobile terminal is in
Indicating that the vehicle-mounted mobile terminal is not under the coverage of the mobile edge computing server in the area.
Preferably, the total overhead is expressed as follows:
wherein,
when the area of the vehicle-mounted mobile terminal is migrated, the following requirements are met:
q m (t)＝ε
when the vehicle-mounted mobile terminal does not have the region migration, the following requirements are met:
q m (t)＝0
wherein epsilon represents the overhead generated by the region migration of the vehicle-mounted mobile terminal,
showing the connection state of the vehicle-mounted mobile terminal and the mobile edge computing server in the area at the moment t,
vehicle-mounted mobile terminal for representing t-1 momentA connection status with a mobile edge computing server within the area.
Preferably, a virtual queue and a penalty function are established for energy consumption according to the Lyapunov theory, the drifting upper boundary of the Lyapunov is calculated, and the sub-problem of the optimization model is converted into an II 4 The method comprises the following steps:
because the energy consumption backlog quantity of the vehicle-mounted mobile terminal is influenced by the current residual electric quantity which depends on the total energy consumption at the last moment, the virtual queue Q is established according to the Lyapunov theory m (t) represents the backlog of energy consumption at time t:
Q m (t+1)＝max{Q m (t)+E cur (t)-E m (t),0}
wherein Q m (t + 1) represents the energy consumption backlog at time t + 1;
according to the lyapunov theory, the quadratic lyapunov function at time t is expressed as:
wherein L (t) represents a virtual queue Q m (t) a scalar of the total backlog;
the difference between the scalar quantity of the total backlog of the virtual queue at time t +1 and the scalar quantity of the total backlog of the virtual queue at time t is called Lyapunov drift DeltaL (t), and is expressed as:
ΔL(t)＝L(t+1)-L(t)
substituting lyapunov drift into drift theorem, and expressing as:
wherein V is a constant control parameter representing a tradeoff between system overhead and virtual queues;
substituting the drift theorem into the derivation of the energy consumption backlog:
therefore, the temperature of the molten metal is controlled,
and,
wherein E is max The maximum energy consumption required by the vehicle-mounted mobile terminal to execute the task under the single time slot is represented;
B(t)≤B
and,
the obtained drift theorem satisfies:
since B is constant, the formula (1) can be converted into pi 4 。
Preferably according to II 4 And constructing a mathematical expression of an unloading decision, a calculation rate of the vehicle-mounted mobile terminal to the task and the transmitting power of the vehicle-mounted mobile terminal, wherein the mathematical expression comprises the following steps:
∏ 4 the solution of (ii) can be regarded as two parts, namely, the combined solution of the current residual capacity of the vehicle-mounted mobile terminal, the total consumption energy consumption and the system overhead is established, and a current residual capacity model pi of the vehicle-mounted mobile terminal is established 4.1 ：
∏ 4.1 :
Current residual capacity E of vehicle-mounted mobile terminal cur (t) satisfies:
wherein,
represents the average of the required maximum energy consumption of all tasks, and
optimal solution of current remaining capacity of the vehicle-mounted mobile terminal
Comprises the following steps:
establishing a total consumption energy consumption and system overhead model pi 4.2 ：
∏ 4.2 :
s.t.C2,C3,C7,C8
Calculating an unloading decision: defining the calculation rate of the vehicle-mounted mobile terminal to the task, the calculation rate of the mobile edge calculation server to the task and the emission power of the vehicle-mounted mobile terminal to be known, solving an unloading decision under a single time slot tau, and connecting II 4.2 Rewritten as n 4.2.1 ：
∏ 4.2.1 :
s.t.
II 4.2.1 For functional representation:
wherein λ is m (τ) represents the amount of computational tasks that would be generated by loading a mobile terminal in time slot τ, f m (τ) represents the calculated rate of tasks at the time slot τ for the loaded mobile terminal, r m,n (τ) represents the task transmission rate at slot τ,
representing the computation rate of the mobile edge computing server for the task at time slot tau,
representing the transmit power of the downloaded mobile terminal in time slot tau,
representing a monotonic function whose monotonicity depends on V-Q m (t) and
positive and negative, unload decision s m The value range of (t) can be determined by pi 4.2.1 The constraint of (2) obtains:
from this, an offloading strategy S can be derived m (t)：
Calculating the calculation rate of the vehicle-mounted mobile terminal to the task: defining an offloading decision,The calculation rate of the mobile edge calculation server to the task and the transmission power of the vehicle-mounted mobile terminal are known, the calculation rate of the vehicle-mounted mobile terminal to the task is solved in a single time slot, and pi is obtained 4.2 Rewritten as n 4.2.2 ：
Π 4.2.2 ：
s.t.(1-s m (τ))λ m (τ)c m /f m (t)≤T max
Wherein, E min The minimum energy consumption required by the vehicle-mounted mobile terminal to execute the task in a single time slot is represented.
II 4.2.2 Expressed in functional form:
wherein s is m (τ) indicates the offload decision at time slot τ, function
Is a quadratic function of the opening direction of the function and V-Q m The positive and negative of (t) are related when V-Q m (t) is positive with opening upward when V-Q m (t) is negative, open downwards, calculating f from the constraint m (t) the following:
calculating the task by the vehicle-mounted mobile terminalRate f m (t)：
Calculating the transmitting power of the vehicle-mounted mobile terminal: defining unloading decision, calculating speed of the vehicle-mounted mobile terminal to the task and calculating speed of the mobile edge calculating server to the task to be known, solving transmitting power of the vehicle-mounted mobile terminal in a single time slot, and calculating pi 4.2 Rewritten as pi 4.2.3 ：
II 4.2.3 Expressed in functional form:
constructing helper functions
Analysis of
Monotonicity of the function:
and,
the first derivative is taken of the helper function:
order:
to pair
And (5) obtaining a derivative:
when in use
When the temperature of the water is higher than the set temperature,
when in use
When the temperature of the water is higher than the set temperature,
at this time
Is a monotonic function whose monotonicity is V-Q m (t) determining. For the
Is solved for, due to the constraints contained therein
Is difficult to solve directly, and therefore a taxonomic discussion is required, as follows:
according to the total time delay T in the formula (2) max To obtain a constraint condition of
The lower bound of (1):
then, the solution is carried out according to the second constraint condition in the formula (2) when
When the temperature of the water is higher than the set temperature,
then the second constraint in equation (2) is also obtained as:
transmitting power of the in-vehicle mobile terminal
Wherein,
two values of the transmitting power of the vehicle-mounted mobile terminal are respectively:
calculating the calculation rate of the mobile edge calculation server to the task: defining an optimal unloading decision, a calculation rate of the vehicle-mounted mobile terminal to the task and vehicle-mounted movementThe transmitting power of the terminal is known, the calculation rate of the mobile edge calculation server to the task is solved under the condition of single time slot, and pi is used 4.2 Rewritten as pi 4.2.4 ：
Π 4.2.4 :
s.t.
II 4.2.4 Expressed in functional form:
function(s)
Is a monotonic function, and is monotonic with V-Q m (t) is related.
According to II 4.2.4 Solving the constraint conditions of
The value range of (A):
computing the computing rate of the server to the task
Compared with the prior art, the invention has the beneficial effects that:
the method determines the weighted sum of the total time delay, the total energy consumption and the total task migration overhead of the mobile edge computing task unloading system as the system overhead, and models the system overhead. Then, the problem of minimizing the total system overhead is solved by using a Lyapunov optimization theory, an unloading decision, the calculation rate of the vehicle-mounted mobile terminal to the task, the transmitting power of the vehicle-mounted mobile terminal and the calculation rate of the mobile edge calculation server to the task are calculated by using an iterative method, the optimal system overhead is obtained, and further the total system overhead is reduced.
Drawings
FIG. 1 is a block diagram of a Lyapunov optimization-based computing offload and resource allocation method of the present invention;
FIG. 2 is a flow chart of a calculation offloading and resource allocation method based on Lyapunov optimization according to the present invention.
Detailed Description
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
It will be understood that when an element is referred to as being "connected" to another element, it can be directly connected to the other element or intervening elements may also be present. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used in the description of the present application herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the application.
As shown in fig. 1-2, a calculation offloading and resource allocation method based on lyapunov optimization is applied to a mobile edge calculation task offloading system, where the mobile edge calculation task offloading system includes M vehicle-mounted mobile terminals and N mobile edge calculation servers, and the calculation offloading and resource allocation method based on lyapunov optimization includes:
it should be noted that, as shown in fig. 1, there is one moving edge computing server in each area, and there are multiple vehicle-mounted mobile terminals, and because of limited computing capability of the vehicle-mounted mobile terminals, each vehicle-mounted mobile terminal can offload all or part of its task to the moving edge computing server in its area for remote execution. Besides, the positions of the vehicle-mounted mobile terminals are randomly distributed and are in a continuously moving state, the edge computing server is in a static state, and the position of the vehicle-mounted mobile terminal is static at each time slot.
Step S1, calculating the weighted sum of the total time delay, the total energy consumption and the total task migration overhead of a task unloading system by using a mobile edge as the system overhead, and establishing an optimization model as follows:
wherein,
wherein E is m (t) represents total energy consumption, q m (T) represents the total task migration overhead, T m (t) represents total delay, C (t) represents overhead,
an overhead weighting factor indicating the occurrence of region migration,
a weight factor representing the total time delay,
weight factor representing the total consumed energy, E total Indicating the total amount of power of the vehicle-mounted mobile terminal, E cur (t) represents a current remaining capacity of the in-vehicle mobile terminal,
the percentage of the current residual capacity of the vehicle-mounted mobile terminal is represented, C1 represents that the calculation rate of the vehicle-mounted mobile terminal to the task cannot be larger than the maximum calculation rate per se
C2 represents that the calculation rate of the mobile edge calculation server to the task cannot be larger than the maximum calculation rate of the mobile edge calculation server
C3 represents that the calculation rate of all tasks in the area of the mobile edge calculation server cannot be larger than the maximum calculation rate of the mobile edge calculation server
C4 represents that the total consumed energy consumption cannot be larger than the current residual electric quantity of the vehicle-mounted mobile terminal, and C5 represents that the total time delay cannot be larger than the maximum tolerance time delay T of the task max And C6 represents that the transmitting power of the vehicle-mounted mobile terminal cannot be larger than the maximum transmitting power of the vehicle-mounted mobile terminal
C7 represents a value range of the unloading decision, C8 represents a value range of a weight factor of the total delay, and T ∈ T = {1,2.
Specifically, a set of vehicle-mounted Mobile terminals M = {1,2,. M., M }, a set of Mobile Edge Computing (MEC) servers N = {1,2,. N., N }, different times T ∈ T = {1,2,..., T }, the working time slot of each Mobile Edge Computing server is τ, and the time slot τ is a time slice of the time T.
Each task of the in-vehicle mobile terminal is represented by a quintuple { lambda [ ] m (t),c m ,T max ,A m (t),B m (t) }, in which, A m (t) the position of the vehicle-mounted mobile terminal at time t is represented by a coordinate (alpha) m (t),β m (t)) represents; b is m (t) represents the moving direction a of the in-vehicle mobile terminal m (t) and moving speed v m (t) of (d). And the vehicle-mounted mobile terminal t is at the time A m The abscissa and ordinate of (t) can be expressed as:
tasks of the vehicle-mounted mobile terminal are independent.
Because the downlink data volume is very small compared with the uplink data volume (namely, the data volume fed back after the mobile edge computing server finishes processing the task is smaller than the data volume transmitted to the mobile edge computing server), the time delay of the computing result returned by the mobile edge computing server to the vehicle-mounted mobile terminal can be ignored.
The total delay is expressed as follows:
T m (t)≤T max ≤τ
wherein,
H m.n (t)＝h m,n (t)g 0 (d 0 /d m,n ) θ
wherein,
represents the execution time delay of the in-vehicle mobile terminal,
representing the execution latency of the mobile edge compute server,
represents the transmission delay, T, of the vehicle-mounted mobile terminal for unloading the task to the mobile edge computing server max Representing the maximum tolerant time delay of each task, tau represents the working time slot of each mobile edge computing server, and the time slot tau is a time segment of the moment t, s m (t) denotes the unload decision at time t, and s m (t) =0 denotes that all tasks are performed at the in-vehicle mobile terminal, s m (t) =1 denotes that the task is completely unloaded to the mobile edge computing server for execution, 0 < s m (t) < 1 indicates that a part of tasks are executed in the vehicle-mounted mobile terminal and another part of tasks are executed in the mobile edge computing server, lambda m (t) represents the calculation task amount generated by downloading the mobile terminal at the moment t, and the unit is bit, c m The CPU period required by the vehicle-mounted mobile terminal to calculate each bit of data is represented, and the unit is cycle/bit, f m (t) represents the calculated rate of the tasks at the time t of the vehicle-mounted mobile terminal,
representing the computation rate of the mobile edge computation server on the task, r m,n (t) represents the task transmission rate at time t,
denotes the transmission power of the mobile terminal loaded at time t, I denotes the average interference in the respective region, σ 2 Representing the channel background noise, ω the channel bandwidth, H m,n (t) denotes channel gain, g 0 Represents the path loss constant, theta represents the road power loss exponent, d 0 Denotes a reference distance, d m,n Indicating the distance h from the vehicle-mounted mobile terminal to the mobile edge calculation server in the area m,n (t) represents a small-scale Rayleigh fading factor between the vehicle-mounted mobile terminal and the mobile edge calculation server in the area where the vehicle-mounted mobile terminal is located,
indicating the connection state of the vehicle-mounted mobile terminal and the mobile edge computing server in the area, and when the connection state is
Indicating that the vehicle-mounted mobile terminal is under the coverage of the mobile edge computing server in the area where the vehicle-mounted mobile terminal is located
Indicating that the vehicle-mounted mobile terminal is not under the coverage of the mobile edge computing server in the area.
The total consumed energy is expressed as follows:
wherein,
wherein,
represents the execution energy consumption of the in-vehicle mobile terminal,
represents the transmission energy consumption of the vehicle-mounted mobile terminal for unloading the task to the mobile edge computing server,
the power of the vehicle-mounted mobile terminal for executing the task is represented, and k represents a power traction coefficient.
The total overhead is expressed as follows:
wherein,
when the area of the vehicle-mounted mobile terminal is migrated, the following requirements are met:
q m (t)＝ε
when the vehicle-mounted mobile terminal does not have the region migration, the following requirements are met:
q m (t)＝0
wherein epsilon represents the overhead generated by the region migration of the vehicle-mounted mobile terminal,
showing the connection state of the vehicle-mounted mobile terminal and the mobile edge computing server in the area at the moment t,
representing the calculation of the moving edge of the vehicle-mounted mobile terminal and the area at the t-1 momentThe connection status of the server.
S2, introducing a delay penalty function phi (t) into the optimization model, and simplifying the optimization model into:
П 2 :
s.t.C2,C3,C7,C8
and phi (t) = + (plus) varies when the execution energy consumption of the vehicle-mounted mobile terminal exceeds the current residual electric quantity of the vehicle-mounted mobile terminal at the moment t or the execution delay of the vehicle-mounted mobile terminal exceeds the maximum tolerance delay of the task, and phi (t) =0 when the execution energy consumption of the vehicle-mounted mobile terminal does not exceed the current residual electric quantity of the vehicle-mounted mobile terminal at the moment t or the execution delay of the vehicle-mounted mobile terminal does not exceed the maximum tolerance delay of the task.
Step S3, according to the time segments 2 Splitting to obtain the subproblems of the optimization model:
П 3 :
s.t.C2,C3,C7,C8
s4, establishing a virtual queue and a penalty function for energy consumption according to the Lyapunov theory, calculating the drifting upper boundary of the Lyapunov, and converting the subproblems of the optimization model into the following steps:
Π 4 :
s.t.C2,C3,C7,C8
wherein Q is m (t) represents the energy consumption backlog at time t, and V is a constant control parameter representing the tradeoff between system overhead and virtual queues.
Specifically, because the energy consumption backlog quantity of the vehicle-mounted mobile terminal is influenced by the current residual energy quantity, which depends on the total energy consumption at the last moment, the virtual queue Q is established according to the lyapunov theory m (t) represents the backlog of energy consumption at time t:
Q m (t+1)＝max{Q m (t)+E cur (t)-E m (t),0}
wherein Q is m (t + 1) represents the energy consumption backlog at time t + 1;
according to the lyapunov theory, the quadratic lyapunov function at time t is expressed as:
wherein L (t) represents a virtual queue Q m (t) a scalar of the total backlog;
the difference between the scalar quantity of the total backlog of the virtual queue at time t +1 and the scalar quantity of the total backlog of the virtual queue at time t is called Lyapunov drift DeltaL (t), and is expressed as:
ΔL(t)＝L(t+1)-L(t)
substituting lyapunov drift into the drift theorem, expressed as:
wherein V is a constant control parameter representing a tradeoff between system overhead and virtual queues;
substituting the drift theorem into the derivation of the energy consumption backlog:
therefore, the temperature of the molten metal is controlled,
and,
wherein E is max The maximum energy consumption required by the vehicle-mounted mobile terminal to execute the task under the single time slot is represented;
B(t)≤B
and the number of the first and second groups is,
the obtained drift theorem satisfies:
since B is constant, equation (1) can be converted into pi 4 。
Step S5, according to pi 4 And constructing a mathematical expression of the unloading decision, the calculation rate of the vehicle-mounted mobile terminal to the task and the transmitting power of the vehicle-mounted mobile terminal.
In particular, pi 4 The solution can be regarded as two parts, namely the combined solution of the current residual capacity, the total consumption energy consumption and the system overhead of the vehicle-mounted mobile terminal, and the current residual capacity model pi of the vehicle-mounted mobile terminal is established 4.1 ：
П 4.1 :
Current residual capacity E of vehicle-mounted mobile terminal cur (t) satisfies:
wherein,
represents the average of the required maximum energy consumption of all tasks, and
optimal solution of current remaining capacity of the vehicle-mounted mobile terminal
Comprises the following steps:
establishing a total consumption energy consumption and system overhead model pi 4.2 ：
∏ 4.2 :
s.t.C2,C3,C7,C8
Calculating an unloading decision: defining the calculation rate of the vehicle-mounted mobile terminal to the task, the calculation rate of the mobile edge calculation server to the task and the emission power of the vehicle-mounted mobile terminal to be known, solving an unloading decision under a single time slot tau, and connecting II 4.2 Rewritten as pi 4.2.1 ：
∏ 4.2.1 :
s.t.
II 4.2.1 For functional representation:
wherein λ is m (τ) represents the amount of computational tasks that would be generated by loading a mobile terminal in time slot τ, f m (τ) represents the calculated speed of the task by the loaded mobile terminal at time slot τRate, r m,n (τ) represents the task transmission rate at slot τ,
representing the computation rate of the mobile edge computing server for the task at time slot tau,
representing the transmit power of the downloaded mobile terminal in time slot tau,
representing a monotonic function whose monotonicity depends on V-Q m (t) and
positive and negative, unload decision s m The value range of (t) can be determined by pi 4.2.1 The constraint of (2) obtains:
from this, an offloading strategy S can be derived m (t)：
Calculating the calculation rate of the vehicle-mounted mobile terminal to the task: defining an unloading decision, calculating the task rate of the mobile edge calculation server and the transmitting power of the vehicle-mounted mobile terminal to be known, solving the task rate of the vehicle-mounted mobile terminal in a single time slot, and combining pi 4.2 Rewritten as n 4.2.2 ：
Π 4.2.2 ：
s.t.(1-s m (τ))λ m (τ)c m /f m (t)≤T max
Wherein E is min The minimum energy consumption required by the vehicle-mounted mobile terminal to execute the task under the single time slot is represented;
II 4.2.2 Expressed in functional form:
wherein s is m (τ) indicates the offload decision at time slot τ, function
Is a quadratic function of the relationship, and the opening direction of the function and V-Q m The positive and negative of (t) are related when V-Q m (t) is positive with opening upward when V-Q m (t) is negative, open downwards, calculating f from the constraint m (t) the following:
the vehicle-mounted mobile terminal calculates the task rate f m (t)：
Calculating the transmitting power of the vehicle-mounted mobile terminal: defining unloading decision, calculating speed of the vehicle-mounted mobile terminal to the task and calculating speed of the mobile edge calculating server to the task to be known, solving transmitting power of the vehicle-mounted mobile terminal in a single time slot, and combining II 4.2 Rewritten as pi 4.2.3 ：
II 4.2.3 Expressed in functional form:
constructing helper functions
Analysis of
Monotonicity of the function:
and,
the first derivative is taken of the helper function:
order:
for is to
And (5) obtaining a derivative:
when in use
When the temperature of the water is higher than the set temperature,
when the temperature is higher than the set temperature
When the temperature of the water is higher than the set temperature,
at this time
Is a monotonic function whose monotonicity is V-Q m (t) determining. For the
Is solved for, due to the constraints contained therein
Is difficult to solve directly, and therefore a classification discussion is required, as follows:
according to the total time delay T in the formula (2) max To obtain a constraint condition of
The lower bound of (1):
then, the solution is carried out according to the second constraint condition in the formula (2) when
When the temperature of the water is higher than the set temperature,
then the second constraint in equation (2) is also obtained as:
transmitting power of the in-vehicle mobile terminal
Wherein,
respectively two values of the transmitting power of the vehicle-mounted mobile terminal:
calculating the calculation rate of the mobile edge calculation server to the task: defining an optimal unloading decision, the calculation rate of the vehicle-mounted mobile terminal to the task and the transmission power of the vehicle-mounted mobile terminal to be known, solving the calculation rate of the mobile edge calculation server to the task in a single time slot, and calculating the II 4.2 Rewritten as II 4.2.4 ：
Π 4.2.4 :
s.t.
II 4.2.4 Expressed in functional form:
function(s)
Is a monotonic function, and is monotonic with V-Q m (t) related;
according to II 4.2.4 Solving the constraint conditions of
The value range of (A):
computing the computing rate of the server to the task
And S6, randomly generating an unloading decision, the calculation rate of the vehicle-mounted mobile terminal to the task, the transmitting power of the vehicle-mounted mobile terminal and the initial solution of the calculation rate of the mobile edge calculation server to the task, then performing iterative calculation by using an iterative algorithm, updating the system overhead after each iteration until the value of the system overhead is converged, and outputting the final system overhead as the optimal system overhead.
Specifically, as shown in fig. 2, the iterative algorithm in this embodiment is implemented based on python, and the number M of the vehicle-mounted mobile terminals and the number of the mobile edge computing servers are input into the algorithmN, channel bandwidth omega, maximum rate of task calculation by vehicle-mounted mobile terminal
And maximum rate of computation of tasks by the mobile edge compute server
And parameters of a quintuple of the task, then randomly generating an unloading decision, a calculation rate of the vehicle-mounted mobile terminal to the task, a transmitting power of the vehicle-mounted mobile terminal and a calculation rate of the mobile edge calculation server to the task, solving corresponding system overhead, and taking secondary system overhead as initialization system overhead C old (t), the iterative algorithm iterates the unloading decision, the calculation rate of the vehicle-mounted mobile terminal to the task, the transmitting power of the vehicle-mounted mobile terminal and the calculation rate of the mobile edge calculation server to the task, and the updating of the system overhead C is completed each time of iteration new And (t), if the value of the direct system overhead is in convergence, finishing iteration to obtain the optimal system overhead, wherein the unloading decision, the calculation rate of the vehicle-mounted mobile terminal to the task, the transmitting power of the vehicle-mounted mobile terminal and the calculation rate of the mobile edge calculation server to the task are all optimal solutions.
The method determines the weighted sum of the total time delay, the total energy consumption and the total task migration overhead of the mobile edge computing task unloading system as the system overhead, and models the system overhead. Then, the problem of minimizing the total system overhead is solved by using a Lyapunov optimization theory, an unloading decision, the calculation rate of the vehicle-mounted mobile terminal to the task, the transmitting power of the vehicle-mounted mobile terminal and the calculation rate of the mobile edge calculation server to the task are calculated by using an iterative method, the optimal system overhead is obtained, and further the total system overhead is reduced.
The technical features of the embodiments described above may be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the embodiments described above are not described, but should be considered as being within the scope of the present specification as long as there is no contradiction between the combinations of the technical features.
The above-mentioned embodiments only express the more specific and detailed embodiments described in the present application, but not be construed as limiting the claims. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present patent application shall be subject to the appended claims.
Claims (6)
1. A calculation unloading and resource allocation method based on Lyapunov optimization is applied to a mobile edge calculation task unloading system and is characterized in that: the mobile edge computing task unloading system comprises M vehicle-mounted mobile terminals and N mobile edge computing servers, and the computing unloading and resource allocation method based on Lyapunov optimization comprises the following steps:
and (3) calculating the weighted sum of the total time delay, the total energy consumption and the total task migration overhead of the task unloading system by using the mobile edge as the system overhead, and establishing an optimization model as follows:
C4:E m (t)≤E cur (t)
C5:T m (t)≤T max ≤τ
C7:0≤s m (t)≤1
wherein,
wherein E is m (t) represents total energy consumption, q m (T) represents the total task migration overhead, T m (t) represents total delay, C (t) represents overhead,
an overhead weighting factor indicating the occurrence of region migration,
a weight factor representing the total time delay,
weight factor representing the total consumed energy, E total Indicating the total amount of power of the vehicle-mounted mobile terminal, E cur (t) represents a current remaining capacity of the in-vehicle mobile terminal,
the percentage of the current residual capacity of the vehicle-mounted mobile terminal is represented, C1 represents that the calculation rate of the vehicle-mounted mobile terminal to the task cannot be larger than the maximum calculation rate per se
C2 represents that the calculation rate of the mobile edge calculation server to the task cannot be larger than the maximum calculation rate of the mobile edge calculation server
C3 represents that the calculation rate of the mobile edge calculation server for all tasks in the area cannot be larger than the maximum calculation rate of the mobile edge calculation server
C4 represents that the total consumed energy consumption cannot be larger than the current residual electric quantity of the vehicle-mounted mobile terminal, and C5 represents that the total time delay cannot be larger than the maximum tolerant time delay T of the task max And C6 represents that the transmitting power of the vehicle-mounted mobile terminal cannot be larger than the maximum transmitting power of the vehicle-mounted mobile terminal
C7 represents a value range of the unloading decision, C8 represents a value range of a weight factor of the total delay, and T ∈ T = {1,2., T } at different times;
introducing a delay penalty function phi (t) into the optimization model, and simplifying the optimization model into the following steps:
s.t.C2,C3,C7,C8
according to time segments 2 Splitting to obtain the subproblems of the optimization model:
s.t.C2,C3,C7,C8
establishing a virtual queue and a penalty function for energy consumption according to the Lyapunov theory, calculating the drift upper boundary of the Lyapunov, and converting the subproblems of the optimization model into:
s.t.C2,C3,C7,C8
wherein Q m (t) represents the energy consumption backlog at time t, and V is a constant control parameter representing the balance between the system overhead and the virtual queue;
according to II 4 Constructing an unloading decision, a calculation rate of the vehicle-mounted mobile terminal to the task and a mathematical expression of the transmitting power of the vehicle-mounted mobile terminal;
and randomly generating an unloading decision, the calculation rate of the vehicle-mounted mobile terminal to the task, the transmitting power of the vehicle-mounted mobile terminal and an initial solution of the calculation rate of the mobile edge calculation server to the task, then performing iterative calculation by using an iterative algorithm, updating the system overhead after each iteration until the value of the system overhead is converged, and outputting the final system overhead as the optimal system overhead.
2. The method of lyapunov-optimization-based computational offload and resource allocation according to claim 1, wherein: the total delay is expressed as follows:
T m (t)≤T max ≤τ
wherein,
H m.n (t)＝h m,n (t)g 0 (d 0 /d m,n ) θ
wherein,
represents the execution time delay of the in-vehicle mobile terminal,
representing the execution latency of the mobile edge compute server,
represents the transmission delay, T, of the vehicle-mounted mobile terminal for unloading the task to the mobile edge computing server max Representing the maximum tolerant time delay of each task, representing the working time slot of each mobile edge computing server, wherein the time slot is a time slice of the moment t, s m (t) denotes the offloading decision at time t, and s m (t) =0 denotes that all tasks are performed at the in-vehicle mobile terminal, s m (t) =1 denotes that the task is completely unloaded to the mobile edge computing server for execution, 0 < s m (t) < 1 indicates that a part of tasks are executed in the vehicle-mounted mobile terminal and another part of tasks are executed in the mobile edge computing server, lambda m (t) generated by downloading the mobile terminal at time tCalculating the task amount and the unit is bit, c m The CPU period required by the vehicle-mounted mobile terminal to calculate each bit of data is represented, and the unit is cycle/bit, f m (t) represents the calculated rate of the task at time t of the in-vehicle mobile terminal,
representing the computation rate of the mobile edge computation server on the task, r m,n (t) represents the task transmission rate at time t,
representing the transmission power of the mobile terminal loaded at time t, I representing the average interference in each zone, σ 2 Representing the channel background noise, ω the channel bandwidth, H m,n (t) denotes channel gain, g 0 Represents the path loss constant, theta represents the road power loss exponent, d 0 Denotes a reference distance, d m,n Indicating the distance h from the vehicle-mounted mobile terminal to the mobile edge calculation server in the area m,n (t) represents a small-scale Rayleigh fading factor between the vehicle-mounted mobile terminal and the mobile edge calculation server in the area where the vehicle-mounted mobile terminal is located,
indicating the connection state of the vehicle-mounted mobile terminal and the mobile edge computing server in the area, and when the connection state is
When the coverage of the mobile edge calculation server in the area of the vehicle-mounted mobile terminal is represented, the vehicle-mounted mobile terminal is in
Indicating that the vehicle-mounted mobile terminal is not under the coverage of the mobile edge computing server in the area.
3. The method of lyapunov-optimization-based computational offload and resource allocation according to claim 2, wherein: the total consumed energy is expressed as follows:
wherein,
wherein,
represents the execution energy consumption of the in-vehicle mobile terminal,
represents the transmission energy consumption of the vehicle-mounted mobile terminal for unloading the task to the mobile edge computing server,
the power of the vehicle-mounted mobile terminal for executing the task is represented, and k represents a power traction coefficient.
4. The method of lyapunov-optimization-based computational offload and resource allocation according to claim 3, wherein: the total overhead is expressed as follows:
wherein,
when the area of the vehicle-mounted mobile terminal is migrated, the following requirements are met:
q m (t)＝ε
when the area of the vehicle-mounted mobile terminal is not migrated, the following conditions are met:
q m (t)＝0
wherein epsilon represents the overhead generated by the region migration of the vehicle-mounted mobile terminal,
showing the connection state of the vehicle-mounted mobile terminal and the mobile edge computing server in the area at the moment t,
and the connection state of the vehicle-mounted mobile terminal and the mobile edge computing server in the area at the moment of t-1 is shown.
5. The Lyapunov optimization-based computing offload and resource allocation method of claim 4, wherein: establishing a virtual queue and a penalty function according to the Lyapunov theory pair energy consumption, calculating the drift upper boundary of the Lyapunov, and converting the sub-problem of the optimization model into pi 4 The method comprises the following steps:
because the energy consumption backlog quantity of the vehicle-mounted mobile terminal is influenced by the current residual electric quantity which depends on the total energy consumption at the last moment, the virtual queue Q is established according to the Lyapunov theory m (t) represents the backlog of energy consumption at time t:
Q m (t+1)＝max{Q m (t)+E cur (t)-E m (t),0}
wherein Q is m (t + 1) represents the energy consumption backlog at time t + 1;
according to lyapunov theory, the quadratic lyapunov function at time t is expressed as:
wherein L (t) represents a virtual queue Q m (t) a scalar of the total backlog;
the difference between the scalar quantity of the total backlog of the virtual queue at time t +1 and the scalar quantity of the total backlog of the virtual queue at time t is called Lyapunov drift DeltaL (t), and is expressed as:
ΔL(t)＝L(t+1)-L(t)
substituting lyapunov drift into the drift theorem, expressed as:
wherein V is a constant control parameter representing a tradeoff between system overhead and virtual queues;
substituting the drift theorem into the derivation of the energy consumption backlog:
therefore, the temperature of the molten metal is controlled,
and,
wherein, E max The maximum energy consumption required by the vehicle-mounted mobile terminal to execute the task under the single time slot is represented;
B(t)≤B
and,
the obtained drift theorem satisfies:
since B is constant, the formula (1) can be converted into pi 4 。
6. The Lyapunov optimization-based computing offload and resource allocation method of claim 4, wherein: the base II 4 And constructing a mathematical expression of an unloading decision, a calculation rate of the vehicle-mounted mobile terminal to the task and the transmitting power of the vehicle-mounted mobile terminal, wherein the mathematical expression comprises the following steps:
Π 4 the solution of (ii) can be regarded as two parts, namely, the combined solution of the current residual capacity of the vehicle-mounted mobile terminal, the total consumption energy consumption and the system overhead is established, and a current residual capacity model pi of the vehicle-mounted mobile terminal is established 4.1 ：
Current residual capacity E of vehicle-mounted mobile terminal cur (t) satisfies:
wherein,
represents the average of the required maximum energy consumption of all tasks, and
optimal solution of current remaining capacity of the vehicle-mounted mobile terminal
Comprises the following steps:
establishing a total consumption energy consumption and system overhead model pi 4.2 ：
s.t.C2,C3,C7,C8
Calculating an unloading decision: defining the calculation rate of the vehicle-mounted mobile terminal to the task, the calculation rate of the mobile edge calculation server to the task and the emission power of the vehicle-mounted mobile terminal to be known, solving an unloading decision under a single time slot tau, and connecting II 4.2 Rewritten as n 4.2.1 ：
II 4.2.1 For functional representation:
wherein λ is m (τ) represents the amount of computational tasks that would be generated by loading a mobile terminal in time slot τ, f m (τ) represents the calculated rate of tasks at the time slot τ for the loaded mobile terminal, r m,n (τ) represents the task transmission rate at time slot τ,
representing the computation rate of the mobile edge computing server for the task at time slot tau,
representing the transmit power of the downloaded mobile terminal in time slot tau,
representing a monotonic function whose monotonicity depends on V-Q m (t) and
positive and negative, unload decision s m The value range of (t) can be II 4.2.1 The constraint of (2) obtains:
from this, an offloading strategy S can be derived m (t)：
Calculating the calculation rate of the vehicle-mounted mobile terminal to the task: defining an unloading decision, calculating the task rate of the mobile edge calculation server and the transmitting power of the vehicle-mounted mobile terminal to be known, solving the task rate of the vehicle-mounted mobile terminal in a single time slot, and combining pi 4.2 Rewritten as pi 4.2.2 ：
s.t.(1-s m (τ))λ m (τ)c m /f m (t)≤T max
Wherein E is min The minimum energy consumption required by the vehicle-mounted mobile terminal to execute the task under the single time slot is represented;
II 4.2.2 Expressed in functional form:
wherein s is m (τ) indicates the offload decision at time slot τ, function
Is a quadratic function of the opening direction of the function and V-Q m The positive and negative of (t) are related when V-Q m (t) is positive with opening upward when V-Q m (t) is negative, open downwards, calculating f from the constraint m (t) the following:
the calculation rate f of the vehicle-mounted mobile terminal to the task m (t)：
Calculating the transmitting power of the vehicle-mounted mobile terminal: defining unloading decision, calculating speed of the vehicle-mounted mobile terminal to the task and calculating speed of the mobile edge calculating server to the task to be known, and solving vehicle-mounted mobile in single time slotTransmitting power of terminal, will 4.2 Rewritten as pi 4.2.3 ：
II with door 4.2.3 Expressed in functional form:
constructing helper functions
Analysis of
Monotonicity of the function:
and,
the first derivative is taken of the helper function:
order:
to pair
And (5) obtaining a derivative:
when in use
When the utility model is used, the water is discharged,
when in use
When the temperature of the water is higher than the set temperature,
at this time
Is a monotonic function whose monotonicity is V-Q m (t) determining. For the
Is solved for, due to the constraints contained therein
Is difficult to solve directly, and therefore a classification discussion is required, as follows:
according to the total time delay T in the formula (2) max To obtain a constraint condition of
The lower bound of (1):
according to the formula (2)Is solved when the second constraint of (2) is satisfied
When the temperature of the water is higher than the set temperature,
then the second constraint in equation (2) is also obtained as:
transmitting power of the in-vehicle mobile terminal
Wherein,
two values of the transmitting power of the vehicle-mounted mobile terminal are respectively:
calculating the calculation rate of the mobile edge calculation server to the task: defining an optimal unloading decision, calculating the task speed of the vehicle-mounted mobile terminal and the transmitting power of the vehicle-mounted mobile terminal to be known, and solving the mobile edge calculation server pair in a single time slotCalculating rate of task, general 4.2 Rewritten as II 4.2.4 ：
II with door 4.2.4 Expressed in functional form:
function(s)
Is a monotonic function, and is monotonic with V-Q m (t) related;
according to II 4.2.4 Solving the constraint conditions of
The value range of (A):
computing the computing rate of the server to the task
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211347119.7A CN115665802A (en) | 2022-10-31 | 2022-10-31 | Calculation unloading and resource allocation method based on Lyapunov optimization |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211347119.7A CN115665802A (en) | 2022-10-31 | 2022-10-31 | Calculation unloading and resource allocation method based on Lyapunov optimization |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115665802A true CN115665802A (en) | 2023-01-31 |
Family
ID=84993821
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211347119.7A Pending CN115665802A (en) | 2022-10-31 | 2022-10-31 | Calculation unloading and resource allocation method based on Lyapunov optimization |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115665802A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117119444A (en) * | 2023-10-25 | 2023-11-24 | 成都信息工程大学 | Position privacy protection method based on mobile edge calculation |
-
2022
- 2022-10-31 CN CN202211347119.7A patent/CN115665802A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117119444A (en) * | 2023-10-25 | 2023-11-24 | 成都信息工程大学 | Position privacy protection method based on mobile edge calculation |
| CN117119444B (en) * | 2023-10-25 | 2024-01-16 | 成都信息工程大学 | Position privacy protection method based on mobile edge calculation |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111786839B (en) | Calculation unloading method and system for energy efficiency optimization in vehicle-mounted edge calculation network | |
| CN111918311B (en) | Vehicle networking task unloading and resource allocation method based on 5G mobile edge computing | |
| CN111586696B (en) | Resource allocation and unloading decision method based on multi-agent architecture reinforcement learning | |
| CN108809695B (en) | Distributed uplink unloading strategy facing mobile edge calculation | |
| CN111586720B (en) | Task unloading and resource allocation combined optimization method in multi-cell scene | |
| Dai et al. | Joint offloading and resource allocation in vehicular edge computing and networks | |
| CN107766135B (en) | Task allocation method based on particle swarm optimization and simulated annealing optimization in moving cloud | |
| CN112600921B (en) | Heterogeneous mobile edge network-oriented dynamic task unloading method | |
| CN113810233B (en) | Distributed computation unloading method based on computation network cooperation in random network | |
| CN111507601A (en) | Resource optimization allocation decision method based on deep reinforcement learning and block chain consensus | |
| CN113296845A (en) | Multi-cell task unloading algorithm based on deep reinforcement learning in edge computing environment | |
| CN112882815A (en) | Multi-user edge calculation optimization scheduling method based on deep reinforcement learning | |
| CN113286329B (en) | Communication and computing resource joint optimization method based on mobile edge computing | |
| CN111093203A (en) | Service function chain low-cost intelligent deployment method based on environment perception | |
| CN112148380A (en) | Resource optimization method in mobile edge computing task unloading and electronic equipment | |
| CN111913723A (en) | Cloud-edge-end cooperative unloading method and system based on assembly line | |
| CN114051254B (en) | Green cloud edge collaborative computing unloading method based on star-ground fusion network | |
| CN110012039A (en) | Task distribution and power control scheme in a kind of car networking based on ADMM | |
| CN110856259A (en) | Resource allocation and offloading method for adaptive data block size in mobile edge computing environment | |
| CN111130911A (en) | Calculation unloading method based on mobile edge calculation | |
| CN111132074A (en) | Multi-access edge computing unloading and frame time slot resource allocation method in Internet of vehicles environment | |
| CN116541106B (en) | Computing task unloading method, computing device and storage medium | |
| CN114564304A (en) | Task unloading method for edge calculation | |
| CN114980039A (en) | Random task scheduling and resource allocation method in MEC system of D2D cooperative computing | |
| CN115665802A (en) | Calculation unloading and resource allocation method based on Lyapunov optimization |
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 |