CN110941667B - Method and system for calculating and unloading in mobile edge calculation network - Google Patents
Method and system for calculating and unloading in mobile edge calculation network Download PDFInfo
- Publication number
- CN110941667B CN110941667B CN201911082719.3A CN201911082719A CN110941667B CN 110941667 B CN110941667 B CN 110941667B CN 201911082719 A CN201911082719 A CN 201911082719A CN 110941667 B CN110941667 B CN 110941667B
- Authority
- CN
- China
- Prior art keywords
- unloading
- optimal
- user
- task
- calculation
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/27—Replication, distribution or synchronisation of data between databases or within a distributed database system; Distributed database system architectures therefor
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
- H04L67/1001—Protocols in which an application is distributed across nodes in the network for accessing one among a plurality of replicated servers
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D30/00—Reducing energy consumption in communication networks
- Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Computing Systems (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mobile Radio Communication Systems (AREA)
Abstract
The invention provides a computing unloading method and a computing unloading system in a mobile edge computing network, which can reduce the energy consumption and time delay of user terminal equipment. The method comprises the following steps: decomposing a calculation unloading problem in a mobile edge calculation network into two sub-problems of user unloading decision solving and calculation resource allocation solving; initializing calculation resource allocation; screening unloading decisions according to different time delay requirements of users to obtain a feasible unloading decision set of each user, and determining an optimal unloading decision according to an initialization result; calculating to obtain an optimal calculation resource allocation scheme under the current optimal unloading decision; and substituting the obtained optimal computing resource allocation scheme into an objective function to solve the unloading strategy, judging whether the obtained unloading strategy is the same as the optimal unloading decision, if so, updating the optimal unloading strategy until the optimal unloading strategy is not changed any more or the maximum iteration number is reached, and stopping iteration. The invention relates to the field of Internet of things and artificial intelligence.
Description
Technical Field
The invention relates to the field of Internet of things and artificial intelligence, in particular to a computing unloading method and a computing unloading system in a mobile edge computing network.
Background
With the rapid development of wireless communication technology, the communication means and method of people are changing greatly. On one hand, due to the continuous maturity of the 5G technology and the Internet of things technology, a series of brand new communication concepts such as interconnection of everything and mass machine communication are proposed. With the advent of these emerging communication concepts, there is a massive influx of various network devices. Various terminal devices such as smart phones, smart sensors and wearable smart devices need to be connected into the network, which undoubtedly causes huge influence and impact on the existing network. On the other hand, in order to meet different communication demands from machine to machine, and from person to machine, various new communication services are beginning to enter people's lives, such as unmanned technology, virtual reality technology, augmented reality technology, natural language processing technology, and the like. Unlike the conventional communication services, the above technologies all require real-time and fast processing of a large amount of data generated by an application program, and have relatively strict requirements on time delay.
However, the pursuit of mobility for portable communication devices has resulted in a lack of physical size, battery power, and computing power for these mobile communication devices. Therefore, when the computation tasks with the computation intensity and the time delay sensitivity are executed, the phenomena of long computation period, fast battery power consumption and the like may occur, and the experience at the user side is that the running of the application program is unsmooth, the endurance performance of the communication device is poor, that is, the user experience is greatly influenced.
Disclosure of Invention
The technical problem to be solved by the present invention is to provide a method and a system for offloading computation in a mobile edge computing network, so as to solve the problems of small battery capacity and limited computation capability of a user terminal in the prior art.
To solve the foregoing technical problem, an embodiment of the present invention provides a method for offloading computation in a mobile edge computing network, including:
s101, decomposing a calculation unloading problem in a mobile edge calculation network into two sub-problems of user unloading decision solving and calculation resource allocation solving;
s102, initializing calculation resource allocation;
s103, screening unloading decisions according to different time delay requirements of users to obtain a feasible unloading decision set of each user, and determining an optimal unloading decision according to an initialization result;
s104, calculating to obtain an optimal calculation resource allocation scheme under the current optimal unloading decision by taking the sum of the minimized time delay and the energy consumption as a target function;
and S105, substituting the obtained optimal computing resource allocation scheme into an objective function to solve the unloading strategy, judging whether the obtained unloading strategy is the same as the optimal unloading decision, if so, updating the optimal unloading strategy, and returning to execute S104 until the optimal unloading strategy is not changed any more or the maximum iteration number is reached, and stopping iteration.
Further, the initializing the allocation of computing resources comprises:
each user is assigned an initial value of available computing resources.
Further, the screening the unloading decisions according to different time delay requirements of the users to obtain a feasible unloading decision set of each user, and determining the optimal unloading decision according to the initialization result comprises:
calculating the time delay when the user selects different unloading modes, comparing the calculated time delay with the maximum time delay tolerance value of the user, if the calculated time delay is less than or equal to the maximum time delay tolerance value, the unloading mode is feasible, and adding the unloading mode into a feasible unloading decision set of the user; if the delay time is larger than the maximum delay tolerance value, the unloading mode is not feasible;
and determining the optimal unloading decision by taking the sum of the minimized time delay and the energy consumption as an objective function according to the initialization result and the feasible unloading decision in the feasible unloading decision set.
Further, the objective function is:
C7:α+β＝1
wherein A represents an unloading decision matrix, if element a in matrix A ij =1, then represents the task M of the user i i Selecting an unloading mode j to perform unloading calculation; r represents a calculation resource allocation matrix, and the element R in R ij Representing a task M i Assigned to task M when executed in offload mode j i The computing resources of (a); s represents a set of unloading patterns, S = {0,1,2.., m }; u represents a user set, U = {1,2.., n }; alpha and beta respectively represent the proportion of time delay and energy consumption in the system overhead;
representing a task M i Execution time when executed in offload mode j;
representing a task M i Energy consumption when executed in unloaded mode j;
representing a task M i A transmission time of the data;
representing a transmission task M i Energy consumption of (2);
representing a task M i A maximum delay tolerance value of;
the maximum resource amount which can be allocated by the system when the unloading mode j is selected for carrying out unloading calculation is shown; c1, C2, C3, C4, C5, C6, C7 all represent constraints.
Further, the air conditioner is characterized in that,
expressed as:
wherein, C ij Representing a task M i The computational resources required when executed in offload mode j.
Further, the air conditioner is provided with a fan,
expressed as:
where k represents a capacitance coefficient.
Further, the air conditioner is characterized in that,
expressed as:
wherein P denotes a power allocation matrix, D ij Representing a task M i The data volume needing to be uploaded when the unloading mode j is selected for carrying out unloading calculation。
Further, the air conditioner is provided with a fan,
expressed as:
wherein, the elements in P
Represents the transmit power of user i; RATE ij Task M representing user i i The rate of transmission of task data when not executing locally.
Further, RATE ij Expressed as:
wherein W represents the channel bandwidth; elements of P
Represents the transmit power of user l; w represents the noise power; h ij 、H lj Respectively representing the channel gains when the user i and the user l select the unloading mode j; a is a lj Indicating that the user l selects the unload mode j for the unload calculation.
An embodiment of the present invention further provides a system for offloading computation in a mobile edge computing network, including:
the decomposition module is used for decomposing the calculation unloading problem in the mobile edge calculation network into two sub-problems of user unloading decision solving and calculation resource allocation solving;
an initialization module for initializing allocation of computing resources;
the screening module is used for screening the unloading decisions according to different time delay requirements of users to obtain a feasible unloading decision set of each user and determining an optimal unloading decision according to an initialization result;
the determining module is used for calculating to obtain an optimal calculation resource allocation scheme under the current optimal unloading decision by taking the sum of the minimized time delay and the energy consumption as a target function;
and the optimization module is used for substituting the obtained optimal computing resource allocation scheme into the objective function to solve the unloading strategy, judging whether the obtained unloading strategy is the same as the optimal unloading decision, if so, updating the optimal unloading strategy, and returning to the execution determination module to stop iteration until the optimal unloading strategy is not changed any more or the maximum iteration number is reached.
The technical scheme of the invention has the following beneficial effects:
in the scheme, the calculation unloading problem in the mobile edge calculation network is decomposed into two sub-problems of user unloading decision solving and calculation resource allocation solving; initializing calculation resource allocation; screening unloading decisions according to different time delay requirements of users to obtain a feasible unloading decision set of each user, and determining an optimal unloading decision according to an initialization result; calculating to obtain an optimal calculation resource allocation scheme under the current optimal unloading decision by taking the sum of the minimized time delay and the energy consumption as a target function; and substituting the obtained optimal calculation resource allocation scheme into an objective function to solve the unloading strategy, judging whether the obtained unloading strategy is the same as the optimal unloading decision, if so, updating the optimal unloading strategy until the optimal unloading strategy does not change any more or the maximum iteration number is reached, and stopping iteration. Therefore, by optimizing the calculation unloading strategy in the mobile edge calculation network, the user terminal equipment can unload the calculation task to the edge server closer to the user side for execution, so that the energy consumption (short for energy consumption) and the time delay (short for time delay) of the user terminal equipment can be reduced.
Drawings
Fig. 1 is a flowchart illustrating a computation offloading method in a mobile edge computing network according to an embodiment of the present invention;
FIG. 2 is a diagram illustrating a communication structure of a mobile edge computing network according to an embodiment of the present invention;
fig. 3 is a schematic structural diagram of a computing offload system in a mobile edge computing network according to an embodiment of the present invention.
Detailed Description
To make the technical problems, technical solutions and advantages of the present invention more apparent, the following detailed description is given with reference to the accompanying drawings and specific embodiments.
The invention provides a calculation unloading method and a system in a mobile edge calculation network, aiming at the problems of smaller battery capacity and limited calculation capacity of the existing user terminal equipment.
To better understand the computation offload method in the mobile edge computing network according to the embodiment of the present invention, a mobile edge computing architecture is briefly described:
in a mobile edge computing architecture, there are many servers with some computing power deployed at the edge of the network. The user terminal equipment can unload the calculation task to the edge server for execution, the user only needs to provide various parameters required by calculation, the specific calculation process is executed by the edge server, and after a final calculation result is obtained, the final calculation result is fed back to the user by the edge server. Through the calculation unloading mechanism, the user terminal equipment does not need to process a large amount of data frequently any more, so that the operation burden of the user terminal equipment is reduced, part of energy consumption is reduced, and the endurance performance is improved to a certain extent. In addition, the physical distance between the edge servers and the user terminal equipment is relatively short, the user does not need to transfer the task to a remote cloud computing center any more, and the edge servers can meet most of computing requirements of the user, so that the communication distance is greatly shortened, and the problem of long transmission delay can be well solved.
Example one
As shown in fig. 1, a method for offloading computation in a mobile edge computing network according to an embodiment of the present invention includes:
s101, decomposing a calculation unloading problem in a mobile edge calculation network into two sub-problems of user unloading decision solving and calculation resource allocation solving;
s102, initializing calculation resource allocation;
s103, screening unloading decisions according to different time delay requirements of users to obtain a feasible unloading decision set of each user, and determining an optimal unloading decision according to an initialization result;
s104, calculating to obtain an optimal calculation resource allocation scheme under the current optimal unloading decision by taking the sum of the minimized time delay and the energy consumption as a target function;
and S105, substituting the obtained optimal calculation resource allocation scheme into an objective function to solve the unloading strategy, judging whether the obtained unloading strategy is the same as the optimal unloading decision, if so, updating the optimal unloading strategy, and returning to execute S104 until the optimal unloading strategy does not change any more or the maximum iteration number is reached to stop iteration.
The calculation unloading method in the mobile edge calculation network according to the embodiment of the invention decomposes the calculation unloading problem in the mobile edge calculation network into two sub-problems of user unloading decision solving and calculation resource allocation solving; initializing computing resource allocation; screening unloading decisions according to different time delay requirements of users to obtain a feasible unloading decision set of each user, and determining an optimal unloading decision according to an initialization result; calculating to obtain an optimal calculation resource allocation scheme under the current optimal unloading decision by taking the sum of the minimized time delay and the energy consumption as a target function; and substituting the obtained optimal calculation resource allocation scheme into an objective function to solve the unloading strategy, judging whether the obtained unloading strategy is the same as the optimal unloading decision, if so, updating the optimal unloading strategy until the optimal unloading strategy does not change any more or the maximum iteration number is reached, and stopping iteration. In this way, by optimizing the computation offload strategy in the mobile edge computing network, the user terminal device can offload the computation task to the edge server closer to the user side for execution, so that the energy consumption and time delay of the user terminal device can be reduced.
The computation offloading method in the mobile edge computing network provided in this embodiment can be used to deal with challenges brought by computation-intensive and delay-sensitive computation tasks to the computation power and battery capacity of the user terminal device, and implement low-energy-consumption and low-delay efficient processing of the computation tasks.
To better understand the computation offloading method in the mobile edge computing network according to the embodiment of the present invention, the detailed description thereof may specifically include the following steps:
a1, optimizing variable decoupling, and decomposing the original calculation unloading problem into two sub-problems
In this embodiment, a computation offload problem (two optimization variables, i.e., a user offload decision and computation resource allocation with low correlation) in a mobile edge computing network is decomposed into two sub-problems, i.e., a user offload decision solution and a computation resource allocation solution, which are respectively solved.
A2, initializing computing resource allocation
In this embodiment, each user is allocated with a feasible initial value of computing resources, for example, existing computing resources are evenly allocated to each user.
A3, screening the unloading decision of each user to obtain a feasible unloading decision set of each user, and determining the optimal unloading decision
In this embodiment, screening the offloading decisions according to different time delay requirements of the users to obtain a feasible offloading decision set of each user, and determining an optimal offloading decision according to an initialization result, may specifically include the following steps:
a31, calculating time delay when a user selects different unloading modes, comparing the calculated time delay with the maximum time delay tolerance value of the user, if the calculated time delay is less than or equal to the maximum time delay tolerance value, the unloading mode is feasible, and adding the unloading mode into a feasible unloading decision set of the user; if the delay time is larger than the maximum delay tolerance value, the unloading mode is not feasible; in this way, a feasible offloading decision set of all users is obtained;
and A32, determining an optimal unloading decision by taking the sum of the minimized time delay and the energy consumption as an objective function according to the initialization result and the feasible unloading decision in the feasible unloading decision set.
In the embodiment, when the user unloading decision is calculated, the feasible unloading decisions of the user terminal devices are screened in advance according to different time delay requirements of the user terminal devices, so that the complexity of an algorithm can be greatly reduced, and the calculation efficiency is improved.
In this embodiment, a communication system of a single base station shown in fig. 2 is taken as an example for explanation:
an edge Server (MEC Server) is deployed at the Base Station (Base Station) to provide computing service for users. There are n users in the coverage area of the base station, and the set of users is denoted as U = {1,2.., n }; wherein, assuming that each user has a task to perform the offload computation, the task of each user may be denoted as M i If the time delay needed by the calculation unloading exceeds the maximum time delay (namely the maximum time delay tolerance value) T which can be tolerated by the task, i belongs to U MAX Then the unload is considered to have failed.
The inter-device communication can also be used as a means for computation offloading, that is, through the communication means such as inter-device communication among the users, the user with idle computation resources can provide computation service for the user who needs computation offloading, therefore, for each task M that needs to be offloaded i For example, there will be an (m + 1) kind of unload mode, denoted as S = {0,1,2.., m }; for different unloading patterns j, j ∈ S, the explanation is as follows:
when j =0, it indicates the task M i Is unloaded to the edge server for calculation;
when j = i, it indicates the task M i Performing local calculation without unloading;
when j takes other values, the task M representing the user i i Is offloaded to user j for computation.
Based on the unloading mode, the unloading decision of the user is defined as a matrix A of i multiplied by j, i belongs to U and j belongs to S, if an element a in the matrix A ij =1, then represents the task M of the user i i The unload mode j is selected for unload calculations. Taking the system model of fig. 2 as an example for explanation, it can be seen that user 1 (U1) and user 2 (U2) choose to offload tasks to edge server execution, user 3 (U3) choose to offload tasks to user 2 (U2) execution, and user 4 (U4) choose to execute computing tasks locally. Thus, in the offload decision matrix for the user, a 10 ,a 10 ,a 10 ,a 10 ,a 10 ,a 20 ,a 32 ,a 44 Is set to 1. In addition, the computing resource allocation matrix is defined as a matrix R of i x j, i belongs to U and j belongs to S, and the element R in the matrix R ij Representing a task M i Assigned to task M when executed in offload mode j i The computing resources of (1). Defining a power distribution matrix as a matrix P with i multiplied by 1, i epsilon U, and elements in the matrix P
Representing the transmit power of user i.
Task M when user i i When executed in offload mode j, the execution time is expressed as:
wherein, C ij Representing a task M i The computational resources required to be executed in the offload mode j are the number of CPU cycles occupied by the computation task.
Task M i The energy consumption when the offload mode j is executed is:
wherein k >0 is the capacitance coefficient.
Computing task M when user i i When the task is executed in the unloading mode j, j ≠ i, namely when the task is not executed locally, the transmission of task data needs to be carried out, and the transmission RATE RATE ij Is represented as follows:
wherein W represents the channel bandwidth;
represents the transmit power of user i; elements of P
Represents the transmit power of user l; w represents the noise power; h ij 、H lj Respectively representing the channel gains when the user i and the user l select the unloading mode j; a is lj Representing that the user l selects the unloading mode j to carry out unloading calculation;
representing interference between users using the same channel.
After the task execution is completed, the calculation result needs to be returned to the user initiating the uninstalling request, but because the data amount of the calculation result is negligible compared with the input data, the overhead of the system may not be counted in the downlink data transmission process. The transmission time of the task data is as follows:
wherein P denotes a power allocation matrix, D ij Representing a task M i And selecting the unloading mode j to carry out unloading calculation, wherein the data quantity needs to be uploaded.
The energy consumption of the transmission task is as follows:
combining the above analysis, the objective function can be expressed as:
C7:α+β＝1
wherein A represents an unloading decision matrix, if element a in matrix A ij =1, then represents the task M of the user i i Selecting an unloading mode j to perform unloading calculation; r represents a calculation resource allocation matrix, and an element R in R ij Representing a task M i Assigned to task M when executed in offload mode j i The computing resources of (a); s represents a set of unloading patterns, S = {0,1,2.., m }; u represents a set of users, U = {1,2.., n }; alpha and beta respectively represent the proportion of time delay and energy consumption in the system overhead;
representing a task M i Execution time when executed in offload mode j;
representing a task M i Energy consumption when executed in unloaded mode j;
representing a task M i A transmission time of the data;
representing a transmission task M i Energy consumption of (2);
representing a task M i A maximum delay tolerance value of;
the maximum resource amount which can be allocated by the system when the unloading mode j is selected for carrying out unloading calculation is shown; c1, C2, C3, C4, C5, C6, C7 all represent constraints.
In this embodiment, C1 ensures that each task will be executed; c2 represents that each user can only receive one calculation task at the same time; c3 indicates that the resulting allocation of computing resources cannot exceed
C4 and C5 represent the calculation task M if user i i Executing in the offload mode j, the compute resources must be allocated for the offload mode j; c6 is the premise that the user carries out calculation unloading, and if the calculation unloading is carried out, a time delay condition must be met; in C7, alpha and beta respectively represent the proportion of time delay and energy consumption in the system overhead, and the sum of the time delay and the energy consumption is 1.
A4, calculating to obtain an optimal calculation resource allocation scheme under the current optimal unloading decision by taking the sum of the minimized time delay and the energy consumption as a target function;
in this embodiment, the obtained current optimal offloading decision is substituted into the objective function and the constraint inequality, a convex optimization means is adopted, a non-convex problem which is difficult to solve is converted into a quasi-convex problem or other forms which are easy to solve, and an optimal computing resource allocation scheme under the current optimal offloading decision is obtained through calculation.
And A5, substituting the obtained optimal computing resource allocation scheme into the objective function to solve the unloading strategy, judging whether the obtained unloading strategy is the same as the optimal unloading decision, if so, updating the optimal unloading strategy, and returning to execute A4 until the optimal unloading strategy is not changed any more or the maximum iteration number is reached, and stopping iteration.
In this embodiment, the offloading decision and the calculation resource allocation are iteratively updated through an iterative heuristic algorithm until the optimal offloading policy does not change any more, which indicates that the current solution is the optimal solution of the original problem, i.e., the optimal calculation offloading scheme, and the iteration is ended.
It is to be emphasized that:
the calculation unloading method in the mobile edge calculation network provided by the embodiment of the invention is not only suitable for a communication system consisting of a single base station and a single edge server, but also suitable for a communication system consisting of a plurality of base stations and a plurality of edge servers; and each user may have multiple tasks that require offload computation.
Example two
The present invention further provides a specific embodiment of a computing offloading system in a mobile edge computing network, and since the computing offloading system in the mobile edge computing network provided by the present invention corresponds to the specific embodiment of the computing offloading method in the mobile edge computing network, and the computing offloading system in the mobile edge computing network can achieve the object of the present invention by executing the flow steps in the specific embodiment of the method, the explanation in the specific embodiment of the computing offloading method in the mobile edge computing network is also applicable to the specific embodiment of the computing offloading system in the mobile edge computing network provided by the present invention, and will not be described again in the following specific embodiment of the present invention.
As shown in fig. 3, an embodiment of the present invention further provides a computing offload system in a mobile edge computing network, including:
the decomposition module 11 is configured to decompose a computation offload problem in the mobile edge computing network into two sub-problems, namely, a user offload decision solution and a computation resource allocation solution;
an initialization module 12 for initializing the allocation of computing resources;
the screening module 13 is configured to screen the offloading decisions according to different time delay requirements of the users to obtain a feasible offloading decision set of each user, and determine an optimal offloading decision according to an initialization result;
the determining module 14 is configured to calculate, with the sum of the minimum time delay and the energy consumption as an objective function, to obtain an optimal calculation resource allocation scheme under the current optimal offloading decision;
and the optimization module 15 is configured to substitute the obtained optimal computing resource allocation scheme into an objective function to solve the unloading strategy, determine whether the obtained unloading strategy is the same as the optimal unloading decision, update the optimal unloading strategy if the obtained unloading strategy is different from the optimal unloading decision, and return to the execution determination module to stop the iteration until the optimal unloading strategy is not changed any more or the maximum iteration number is reached.
The computation unloading system in the mobile edge computing network of the embodiment of the invention decomposes the computation unloading problem in the mobile edge computing network into two sub-problems of user unloading decision solving and computation resource allocation solving; initializing computing resource allocation; screening unloading decisions according to different time delay requirements of users to obtain a feasible unloading decision set of each user, and determining an optimal unloading decision according to an initialization result; calculating to obtain an optimal calculation resource allocation scheme under the current optimal unloading decision by taking the sum of the minimized time delay and the energy consumption as a target function; and substituting the obtained optimal calculation resource allocation scheme into an objective function to solve the unloading strategy, judging whether the obtained unloading strategy is the same as the optimal unloading decision, if so, updating the optimal unloading strategy until the optimal unloading strategy does not change any more or the maximum iteration number is reached, and stopping iteration. In this way, by optimizing the computation offload strategy in the mobile edge computing network, the user terminal device can offload the computation task to the edge server closer to the user side for execution, so that the energy consumption and time delay of the user terminal device can be reduced.
It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions.
While the foregoing is directed to the preferred embodiment of the present invention, it will be understood by those skilled in the art that various changes and modifications may be made without departing from the spirit and scope of the invention as defined in the appended claims.
Claims (9)
1. A method of computation offload in a mobile edge computing network, comprising:
s101, decomposing a calculation unloading problem in a mobile edge calculation network into two sub-problems of user unloading decision solving and calculation resource allocation solving;
s102, initializing calculation resource allocation;
s103, screening unloading decisions according to different time delay requirements of users to obtain a feasible unloading decision set of each user, and determining an optimal unloading decision according to an initialization result;
s104, calculating to obtain an optimal calculation resource allocation scheme under the current optimal unloading decision by taking the sum of the minimized time delay and the energy consumption as a target function;
s105, substituting the obtained optimal computing resource allocation scheme into an objective function to solve the unloading strategy, judging whether the obtained unloading strategy is the same as the optimal unloading decision, if so, updating the optimal unloading strategy, and returning to execute S104 until the optimal unloading strategy is not changed any more or the maximum iteration number is reached, and stopping iteration;
wherein the objective function is:
C7:α+β＝1
wherein, A represents an unloading decision matrix if an element a in the matrix A ij =1, then represents the task M of the user i i Selecting an unloading mode j for unloading calculation; r represents a calculation resource allocation matrix, and the element R in R ij Representing a task M i Assigned to task M when executed in offload mode j i The computing resources of (1); s represents a set of unloading patterns, S = {0,1,2.., m }; u represents a user set, U = {1,2.., n }; alpha and beta respectively represent the proportion of time delay and energy consumption in the system overhead;
representing a task M i Execution time when executed in offload mode j;
representing a task M i Energy consumption when executed in offload mode j;
representing a task M i A transmission time of the data;
representing a transmission task M i Energy consumption of (2); t is a unit of i MAX Representing a task M i The maximum delay tolerance value of (c);
the maximum resource quantity which can be allocated by the system when the unloading mode j is selected for unloading calculation; c1, C2, C3, C4, C5, C6, C7 all represent constraints.
2. The method of claim 1, wherein initializing a computing resource allocation comprises:
each user is assigned an initial value of available computing resources.
3. The method of claim 1, wherein the screening offloading decisions according to different latency requirements of users to obtain a set of feasible offloading decisions for each user, and determining an optimal offloading decision according to an initialization result comprises:
calculating the time delay when the user selects different unloading modes, comparing the calculated time delay with the maximum time delay tolerance value of the user, if the calculated time delay is less than or equal to the maximum time delay tolerance value, the unloading mode is feasible, and adding the unloading mode into a feasible unloading decision set of the user; if the delay time is larger than the maximum delay tolerance value, the unloading mode is not feasible;
and determining the optimal unloading decision by taking the sum of the minimized time delay and the energy consumption as an objective function according to the initialization result and the feasible unloading decision in the feasible unloading decision set.
4. The method of computing offload in a mobile edge computing network of claim 1,
expressed as:
wherein, C ij Representing a task M i The computational resources required when executed in offload mode j.
5. The method of computation offload in a mobile edge computing network according to claim 4,
expressed as:
where k represents the capacitance coefficient.
6. The method of computation offload in a mobile edge computing network of claim 5,
expressed as:
wherein RATE ij Task M representing user i i The transmission rate of task data when not executing locally; p denotes a power distribution matrix, D ij Representing a task M i And selecting the unloading mode j to perform unloading calculation, wherein the data amount needs to be uploaded.
7. The method of computing offload in a mobile edge computing network according to claim 6,
expressed as:
wherein, the elements in P
Representing the transmit power of user i.
8. Method for computation offload in mobile edge computing network according to claim 6 or 7, characterised in that RATE ij Expressed as:
wherein W represents the channel bandwidth; element P in P l t Represents the transmit power of user l; w represents the noise power; h ij 、H lj Respectively representing the channel gains when the user i and the user l select the unloading mode j; a is lj Indicating that the user l selects the unload mode j for the unload calculation.
9. A computing offload system in a mobile edge computing network, comprising:
the decomposition module is used for decomposing the calculation unloading problem in the mobile edge calculation network into two sub-problems of user unloading decision solving and calculation resource allocation solving;
an initialization module for initializing allocation of computing resources;
the screening module is used for screening the unloading decisions according to different time delay requirements of users to obtain a feasible unloading decision set of each user and determining an optimal unloading decision according to an initialization result;
the determining module is used for calculating to obtain an optimal calculation resource allocation scheme under the current optimal unloading decision by taking the sum of the minimized time delay and the energy consumption as a target function;
the optimization module is used for substituting the obtained optimal computing resource allocation scheme into a target function to solve the unloading strategy, judging whether the obtained unloading strategy is the same as the optimal unloading decision, if the obtained unloading strategy is different from the optimal unloading decision, updating the optimal unloading strategy, and returning to the execution determination module until the optimal unloading strategy is not changed any more or the maximum iteration number is reached, and stopping iteration;
wherein the objective function is:
C7:α+β＝1
wherein A represents an unloading decision matrix, if element a in matrix A ij =1, then represents the task M of the user i i Selecting an unloading mode j for unloading calculation; r represents a calculation resource allocation matrix, and the element R in R ij Representing a task M i Assigned to task M when executed in offload mode j i The computing resources of (1); s represents a set of unloading modes, S = {0,1,2.., m }; u represents a set of users, U = {1,2.., n }; alpha and beta respectively represent the proportion of time delay and energy consumption in the system overhead;
representing a task M i Execution time when executed in offload mode j;
representing a task M i Energy consumption when executed in offload mode j;
representing a task M i A transmission time of the data;
representing a transmission task M i Energy consumption of (2); t is i MAX Representing a task M i A maximum delay tolerance value of;
the maximum resource quantity which can be allocated by the system when the unloading mode j is selected for unloading calculation; c1, C2, C3, C4, C5, C6, C7 all represent constraints.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911082719.3A CN110941667B (en) | 2019-11-07 | 2019-11-07 | Method and system for calculating and unloading in mobile edge calculation network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911082719.3A CN110941667B (en) | 2019-11-07 | 2019-11-07 | Method and system for calculating and unloading in mobile edge calculation network |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110941667A CN110941667A (en) | 2020-03-31 |
| CN110941667B true CN110941667B (en) | 2022-10-14 |
Family
ID=69907436
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911082719.3A Active CN110941667B (en) | 2019-11-07 | 2019-11-07 | Method and system for calculating and unloading in mobile edge calculation network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110941667B (en) |
Families Citing this family (23)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111510957B (en) * | 2020-04-09 | 2023-05-09 | 浪潮通信信息系统有限公司 | Mobile edge server coordinate positioning method based on multi-target constraint optimization |
| CN111669775B (en) * | 2020-05-18 | 2022-07-29 | 南京邮电大学 | Resource allocation method based on support vector machine under heterogeneous network |
| CN112540845B (en) * | 2020-05-27 | 2023-08-04 | 南京邮电大学 | Collaboration system and method based on mobile edge calculation |
| CN111786839B (en) | 2020-07-15 | 2021-09-07 | 南通大学 | Calculation unloading method and system for energy efficiency optimization in vehicle-mounted edge calculation network |
| CN114080058A (en) * | 2020-08-17 | 2022-02-22 | 华为技术有限公司 | Processing method of computing task, communication device and wireless network architecture |
| CN111970154B (en) * | 2020-08-24 | 2022-06-10 | 浙江工商大学 | Unloading decision and resource allocation method based on deep reinforcement learning and convex optimization |
| CN112291335B (en) * | 2020-10-27 | 2021-11-02 | 上海交通大学 | Optimized task scheduling method in mobile edge calculation |
| CN113114714B (en) * | 2020-11-03 | 2022-03-01 | 吉林大学 | Energy-saving method and system for unloading large-scale tasks to 5G edge server |
| CN112689296B (en) * | 2020-12-14 | 2022-06-24 | 山东师范大学 | Edge calculation and cache method and system in heterogeneous IoT network |
| CN112910964B (en) * | 2021-01-18 | 2021-11-30 | 北京理工大学 | LEO satellite network computing unloading method based on hybrid cloud and edge computing |
| CN112995023B (en) * | 2021-03-02 | 2022-04-19 | 北京邮电大学 | Multi-access edge computing network computing unloading system and computing unloading method thereof |
| CN113207136B (en) * | 2021-04-02 | 2022-11-18 | 北京科技大学 | Method and device for joint optimization of computation offloading and resource allocation |
| CN113207150B (en) * | 2021-04-23 | 2022-05-17 | 华中科技大学 | Active and passive hybrid unloading method and device based on backscatter communication |
| CN113342409B (en) * | 2021-04-25 | 2022-10-25 | 山东师范大学 | Delay sensitive task unloading decision method and system for multi-access edge computing system |
| CN113296842B (en) * | 2021-05-14 | 2022-05-06 | 上海交通大学 | Optimized task unloading scheduling method based on correlation in mobile edge calculation |
| CN113371173B (en) * | 2021-06-30 | 2022-11-25 | 湖南工商大学 | High-low altitude cooperative epidemic prevention and control system and method based on edge intelligence |
| CN114125063B (en) * | 2021-08-30 | 2022-07-08 | 国网内蒙古东部电力有限公司 | Power communication network task unloading system, method and application based on service QoS |
| CN113835778A (en) * | 2021-09-14 | 2021-12-24 | 北京信息科技大学 | Task unloading method and device, electronic equipment and storage medium |
| CN113923223B (en) * | 2021-11-15 | 2024-02-06 | 安徽大学 | User allocation method with low time cost in edge environment |
| CN114186416B (en) * | 2021-12-13 | 2024-09-24 | 大连理工大学 | Edge network data age optimization method and system |
| CN114567587B (en) * | 2022-03-14 | 2024-02-20 | 广州穗华能源科技有限公司 | Routing decision method for multi-user access edge computing terminal of power distribution Internet of things |
| CN114915627B (en) * | 2022-04-15 | 2024-04-02 | 浙江工业大学 | Wireless energy supply edge computing network unloading decision method based on FDMA access |
| CN117891613B (en) * | 2024-02-29 | 2024-05-31 | 湘江实验室 | Computing unloading and resource allocation method and system based on Internet of vehicles |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108809695A (en) * | 2018-04-28 | 2018-11-13 | 国网浙江省电力有限公司电力科学研究院 | A kind of distribution uplink unloading strategy towards mobile edge calculations |
| CN109413724A (en) * | 2018-10-11 | 2019-03-01 | 重庆邮电大学 | A kind of task unloading and Resource Allocation Formula based on MEC |
| CN109814951A (en) * | 2019-01-22 | 2019-05-28 | 南京邮电大学 | The combined optimization method of task unloading and resource allocation in mobile edge calculations network |
| CN110035410A (en) * | 2019-03-07 | 2019-07-19 | 中南大学 | Federated resource distribution and the method and system of unloading are calculated in a kind of vehicle-mounted edge network of software definition |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20190208449A1 (en) * | 2017-12-29 | 2019-07-04 | Industrial Technology Research Institute | Mobile edge platform servers and device and message management methods of v2x service thereof |
-
2019
- 2019-11-07 CN CN201911082719.3A patent/CN110941667B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108809695A (en) * | 2018-04-28 | 2018-11-13 | 国网浙江省电力有限公司电力科学研究院 | A kind of distribution uplink unloading strategy towards mobile edge calculations |
| CN109413724A (en) * | 2018-10-11 | 2019-03-01 | 重庆邮电大学 | A kind of task unloading and Resource Allocation Formula based on MEC |
| CN109814951A (en) * | 2019-01-22 | 2019-05-28 | 南京邮电大学 | The combined optimization method of task unloading and resource allocation in mobile edge calculations network |
| CN110035410A (en) * | 2019-03-07 | 2019-07-19 | 中南大学 | Federated resource distribution and the method and system of unloading are calculated in a kind of vehicle-mounted edge network of software definition |
Non-Patent Citations (1)
| Title |
|---|
| "移动边缘计算中的计算卸载和资源管理方案";李邱苹 等;《电信科学》;20190319;第35卷(第3期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110941667A (en) | 2020-03-31 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110941667B (en) | Method and system for calculating and unloading in mobile edge calculation network | |
| CN110928654B (en) | Distributed online task unloading scheduling method in edge computing system | |
| Cui et al. | A novel offloading scheduling method for mobile application in mobile edge computing | |
| CN111010684B (en) | Internet of vehicles resource allocation method based on MEC cache service | |
| Shi et al. | Toward energy-efficient federated learning over 5G+ mobile devices | |
| CN110069341B (en) | Method for scheduling tasks with dependency relationship configured according to needs by combining functions in edge computing | |
| CN110557769A (en) | C-RAN calculation unloading and resource allocation method based on deep reinforcement learning | |
| CN111132191A (en) | Method for unloading, caching and resource allocation of joint tasks of mobile edge computing server | |
| CN112540845B (en) | Collaboration system and method based on mobile edge calculation | |
| CN110519370B (en) | Edge computing resource allocation method based on facility site selection problem | |
| CN109618399A (en) | Distributed energy management solutions optimization method in the mobile edge calculations system of multi-user | |
| CN111836284B (en) | Energy consumption optimization calculation and unloading method and system based on mobile edge calculation | |
| CN111130911A (en) | Calculation unloading method based on mobile edge calculation | |
| Li et al. | Distributed task offloading strategy to low load base stations in mobile edge computing environment | |
| CN111522657A (en) | Distributed equipment collaborative deep learning reasoning method | |
| Wu et al. | Residual energy maximization for wireless powered mobile edge computing systems with mixed-offloading | |
| CN111711962A (en) | Cooperative scheduling method for subtasks of mobile edge computing system | |
| Tuli et al. | SplitPlace: AI augmented splitting and placement of large-scale neural networks in mobile edge environments | |
| Chen et al. | Joint optimization of task offloading and resource allocation via deep reinforcement learning for augmented reality in mobile edge network | |
| Qu et al. | Stochastic cumulative DNN inference with RL-aided adaptive IoT device-edge collaboration | |
| Cheng et al. | Task offloading for automatic speech recognition in edge-cloud computing based mobile networks | |
| CN113159539B (en) | Method for combining green energy scheduling and dynamic task allocation in multi-layer edge computing system | |
| Li | Optimization of task offloading problem based on simulated annealing algorithm in MEC | |
| CN117858109A (en) | User association, task unloading and resource allocation optimization method based on digital twin | |
| CN116828534B (en) | Intensive network large-scale terminal access and resource allocation method based on reinforcement learning |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |