CN112437468A - Task unloading algorithm based on time delay and energy consumption weight calculation - Google Patents

Abstract

The invention provides a task unloading algorithm based on time delay and energy consumption weight calculation aiming at calculating time delay and energy consumption weight coefficients of task execution and solving the problem of task unloading of mobile terminal equipment in mobile edge calculation, and aims to reduce the task execution overhead of the terminal equipment according to user requirements and the electric quantity state of the equipment. The algorithm provides a task unloading and resource allocation strategy based on limited energy and time delay aiming at computation unloading and computation energy consumption coefficient of a task under a single MEC multi-user scene. And solving the optimal allocation resources of the local equipment, solving the resource allocation of the MEC end by utilizing convex optimization, and obtaining the optimal channel allocation through an iterative optimization algorithm on the basis of meeting the time delay and the energy consumption. The unloading scheme provided by the invention can minimize the time delay and energy consumption of the user, meet the user requirements and improve the user experience quality.

Description

Task unloading algorithm based on time delay and energy consumption weight calculation

Technical Field

The invention belongs to the field of task unloading of mobile edge computing, and particularly relates to a task unloading algorithm based on time delay and energy consumption weight computing.

Background

With the development of the internet of things and the arrival of the world of everything interconnection, more and more terminal devices are applied to our lives, but the existing technology and the design limit of the terminal device manufacture limit the computing resources and the battery capacity of the mobile device. The requirements of current emerging mobile applications such as interactive games, augmented reality and the like on calculation and storage are higher and higher, and when the applications run on terminal equipment, the requirements of tasks on processing capacity and cruising capacity are sometimes difficult to meet. When a large amount of calculation programs are operated, the response is slow, the power consumption is fast, and the user experience quality is difficult to meet.

How to solve the contradiction between the limited resources of the mobile terminal equipment and the computing resources required by the application process becomes one of the main problems to be solved urgently in the mobile communication network. The computation offload technique in the mobile edge computation is considered as one of the key techniques to effectively solve the above-described problems. The mobile edge computing does not need to upload the task to a far-end central cloud, but unloads the task to an edge server close to the user terminal, so that transmission delay and energy consumption are greatly reduced, and the problem of resource limitation of the mobile terminal can be effectively solved.

For task unloading, a good unloading method or strategy can not only effectively meet the user requirements, improve the service quality of the user, but also greatly reduce the complexity. Therefore, it is necessary to provide a task offloading algorithm based on time delay and energy consumption weight calculation.

Disclosure of Invention

The invention aims to: aiming at the task unloading problem of mobile terminal equipment in mobile edge calculation, a task unloading algorithm based on time delay and energy consumption weight calculation is provided, and the task execution overhead of the terminal equipment is reduced according to user requirements and the electric quantity state of the equipment. In order to achieve the purpose, the technical scheme adopted by the invention comprises the following parts:

1. a task unloading algorithm based on time delay and energy consumption weight calculation specifically comprises the following implementation steps:

step 1, each terminal device has a task to be processed, the device submits the task, and the time delay and the energy consumption coefficient of the task in the device are respectively calculated.

And 2, performing local optimal resource allocation on the tasks in the equipment.

Step 3, initially unloading all tasks to an MEC server for execution, and setting a for all tasks_i＝1。

And 4, performing optimal resource allocation on the tasks needing to be unloaded to the MEC server for execution.

And 5, carrying out channel allocation according to the algorithm 1.

And 6, making an unloading decision according to the cost of each task at the MEC end and the local end.

And 7, judging whether the unloading decision is changed or not, and if not, stopping the algorithm. Otherwise, the process goes to step 4.

2. The time delay and energy consumption weight calculation of step 1 of claim 1, wherein:

remember REc_iThe residual ratio of the electric quantity of the terminal equipment is as follows:

from which the coefficient of energy consumption Ec is set_iThat is, the smaller the value of the remaining power is in proportion to the maximum power of the terminal device, the stronger the demand for reducing the energy consumption is.

Determining the urgency of the task according to the task deadline, the size of the task and the required computing capacity, and recording as Tcr_iIn this document, the urgency of the task is characterized as the sensitivity of the task to the time delay, and the smaller the numerical value, the higher the sensitivity of the task to the time delay.

Thereby obtaining the time delay coefficient Tc_i：

Because the weight coefficient of the time delay and the energy consumption needs to satisfy the condition: ec_i+Tc_iFor this purpose, the weight coefficient is corrected to obtain a delay weight coefficient le_iAnd energy consumption weight coefficient lt_i。

3. The cardiac channel allocation of step 5 of claim 1, wherein:

when selecting the channel, the following strategies are adopted:

and sequencing all tasks to be unloaded according to the formula (7), and sequentially calculating the minimum transmission rate required by each task. And calculating the reachable rate of each task on any channel according to the formula (8) to form a rate overhead matrix.

Initial channel selection was performed using the Argmax function:

j＝argmax(R_i,j) (10)

thereby forming a decision matrix. From the second iteration, the channel selection is performed again according to the equations (9) and (10) according to the channel selected by the previous generation.

The task unloading method provided by the invention has the following advantages and beneficial effects: according to the method, the user experience and the self condition of the equipment are considered, the optimal resource allocation of the local equipment is solved, the resource allocation of the MEC end is solved by utilizing convex optimization, the optimal channel allocation is obtained through an iterative optimization algorithm on the basis of meeting the time delay and the energy consumption, the time delay and the energy consumption of a user can be minimized through the unloading scheme, the user requirement is met, and the user experience quality is improved.

Drawings

Fig. 1 is a flowchart of a task offloading algorithm based on delay and energy consumption weight calculation according to the present invention.

Detailed Description

In order to make those skilled in the art better understand the technical problems, technical solutions and technical effects in the present application, a task offloading algorithm based on time delay and energy consumption weight calculation according to the present invention is further described in detail below with reference to the accompanying drawings and the detailed description.

FIG. 1 shows the steps of the present invention:

step 1, each terminal device has a task to be processed, the device submits the task, and the time delay and the energy consumption coefficient of the task in the device are respectively calculated. The set of terminal devices is denoted N ═ SMD₁,...,SMD_i,...,SMD_n]Wherein each terminal device has three attributes, denoted as

f_i ^lRepresenting the computational power of the device, i.e. the number of cycles performed in one second. Elec_iRepresenting the current remaining capacity of the device in joules.

Representing the maximum charge storage capacity of the device in joules. Each terminal device has a computationally intensive task to be processed, denoted T_i＝[expt_i,data_i,c_i]Wherein, expt_iData representing expected completion time_iIndicating the size of the data volume, c_iIndicating the computational power required to complete the task, i.e., the number of cpu cycles required.

Remember REc_iThe residual ratio of the electric quantity of the terminal equipment is as follows:

from which the coefficient of energy consumption Ec is set_iThat is, the smaller the value of the remaining power is in proportion to the maximum power of the terminal device, the stronger the demand for reducing the energy consumption is.

Determining the urgency of the task according to the task deadline, the size of the task and the required computing capacity, and recording as Tcr_iIn this document, the urgency of the task is characterized as the sensitivity of the task to the time delay, and the smaller the numerical value, the higher the sensitivity of the task to the time delay.

Thereby obtaining the time delay coefficient Tc_i：

Because the weight coefficient of the time delay and the energy consumption needs to satisfy the condition: ec_i+Tc_iFor this purpose, the weight coefficient is corrected to obtain a delay weight coefficient le_iAnd energy consumption weight coefficient lt_i。

And 2, performing local optimal resource allocation on the tasks in the equipment.

Each local device resource allocation problem can be formalized as:

it can be seen that the local overhead is only related to the local device computing resources. Order:

wherein the formula (9) satisfies the time delay and remaining energy consumption constraint conditions. Therefore, we have a look at f in the above formula_i ^lAnd (5) derivation to obtain:

obtaining an optimal solution:

when f_i ^l＞f_i ^max，f_i ^l＜f_i ^minAnd the time delay and energy consumption constraints are not satisfied.

When f_i ^min≤f_i ^l≤f_i ^maxIn time, three situations are distinguished, specifically:

step 3, initially unloading all tasks to an MEC server for execution, and setting a for all tasks_i＝1。

And 4, performing optimal resource allocation on the tasks needing to be unloaded to the MEC server for execution according to the formula (13).

And 5, carrying out channel allocation according to the algorithm 1.

When selecting the channel, the following strategies are adopted:

and sequencing all tasks to be unloaded according to the formula (7), and sequentially calculating the minimum transmission rate required by each task. And calculating the reachable rate of each task on any channel according to the formula (8) to form a rate overhead matrix.

Initial channel selection was performed using the Argmax function:

j＝argmax(R_i,j) (10)

thereby forming a decision matrix. From the second iteration, the channel selection is performed again according to the equations (9) and (10) according to the channel selected by the previous generation.

And 6, making an unloading decision according to the cost of each task at the MEC end and the local end. If the local execution overhead is less than or equal to the offload overhead, i.e., if the local execution overhead is less than or equal to the offload overhead

Then a_iWith 0, the task is executed locally.

If the local execution overhead is greater than the offload overhead, i.e., if the local execution overhead is greater than the offload overhead

Then a_iAnd 1, unloading the task to the MEC server for execution.

And 7, judging whether the unloading decision is changed or not, and if not, stopping the algorithm. Otherwise, the process goes to step 4.

The above examples are only used to illustrate the present invention and not to limit the technical solutions described in the present invention, and it should be understood by those skilled in the art that the task offloading algorithm based on the time delay and energy consumption weight calculation disclosed in the above invention may be modified on the basis of the above without departing from the broad scope, and these modifications are also considered as protection of the present invention.

Claims (3)

1. A task unloading algorithm based on time delay and energy consumption weight calculation specifically comprises the following implementation steps:

step 1, each terminal device has a task to be processed, the device submits the task, and the time delay and the energy consumption coefficient of the task in the device are respectively calculated.

And 2, performing local optimal resource allocation on the tasks in the equipment.

Step 3, initially unloading all tasks to an MEC server for execution, and setting a for all tasks_i＝1。

And 4, performing optimal resource allocation on the tasks needing to be unloaded to the MEC server for execution.

And 5, carrying out channel allocation according to the algorithm 1.

And 6, making an unloading decision according to the cost of each task at the MEC end and the local end.

And 7, judging whether the unloading decision is changed or not, and if not, stopping the algorithm. Otherwise, the process goes to step 4.

2. The time delay and energy consumption weight calculation of step 1 of claim 1, wherein:

remember REc_iThe residual ratio of the electric quantity of the terminal equipment is as follows:

from which the coefficient of energy consumption Ec is set_iThat is, the smaller the value of the remaining power is in proportion to the maximum power of the terminal device, the stronger the demand for reducing the energy consumption is.

Determining the urgency of the task according to the task deadline, the size of the task and the required computing capacity, and recording as Tcr_iIn this document, the urgency of the task is characterized as the sensitivity of the task to the time delay, and the smaller the numerical value, the higher the sensitivity of the task to the time delay.

Thereby obtaining the time delay coefficient Tc_i：

Because the weight coefficient of the time delay and the energy consumption needs to satisfy the condition: ec_i+Tc_iFor this purpose, the weight coefficient is corrected to obtain a delay weight coefficient le_iAnd energy consumption weight coefficient lt_i。

3. The cardiac channel allocation of step 5 of claim 1, wherein:

when selecting the channel, the following strategies are adopted:

and sequencing all tasks to be unloaded according to the formula (7), and sequentially calculating the minimum transmission rate required by each task. And calculating the reachable rate of each task on any channel according to the formula (8) to form a rate overhead matrix.

Initial channel selection was performed using the Argmax function:

j＝arg max(R_i,j) (10)

thereby forming a decision matrix. From the second iteration, the channel selection is performed again according to the equations (9) and (10) according to the channel selected by the previous generation.

The task unloading method provided by the invention has the following advantages and beneficial effects: according to the method, the user experience and the self condition of the equipment are considered, the optimal resource allocation of the local equipment is solved, the resource allocation of the MEC end is solved by utilizing convex optimization, the optimal channel allocation is obtained through an iterative optimization algorithm on the basis of meeting the time delay and the energy consumption, the time delay and the energy consumption of a user can be minimized through the unloading scheme, the user requirement is met, and the user experience quality is improved.

Images