阅读说明：本技术 一种面向mec时延最小的任务功率联合优化方法和系统 (Task power joint optimization method and system for minimum MEC time delay ) 是由 朱佳 李亚利 邹玉龙 于 2021-09-28 设计创作，主要内容包括：本发明公开了一种面向MEC时延最小的任务功率联合优化方法和系统,包括：选定中继；确定用户直接卸载部分任务到MEC的传输速率、用户将部分任务发送给选定中继的传输速率和选定中继转发部分任务到MEC的传输速率；根据传输速率计算用户直传卸载时间和用户-中继卸载时间,将两者中的最大值作为用户卸载时间；比较用户本地计算时间和用户-MEC卸载计算时间,其中的最大值为用户任务计算总时延,所述用户-MEC卸载计算时间为用户卸载时间与MEC计算时间之和；以最小化用户任务计算总时延为目标,建立任务功率联合优化问题并求解获得用户部分任务卸载方案。本发明最小化用户任务计算总时延,获得时延更短的用户部分任务卸载方案。(The invention discloses a method for making a paper-making machine MEC The task power joint optimization method and system with the minimum time delay comprise the following steps: selecting a relay; determining that a user is directly offloading a portion of a task to MEC The transmission rate at which the user sends part of the task to the selected relay, and the selected relay forwards part of the task to MEC The transmission rate of (c); calculating user direct transmission unloading time and user-relay unloading time according to the transmission rate, and taking the maximum value of the user direct transmission unloading time and the user-relay unloading time as the user unloading time; comparing user local computation time with user- MEC Unloading the computation time, wherein the maximum value is the total time delay of the user task computation, the user- MEC The uninstalling calculation time is the user uninstalling time and MEC calculating the sum of the time; establishing task power association with the aim of minimizing the total time delay calculated by user taskAnd optimizing the problem and solving to obtain a user part task unloading scheme. The invention minimizes the total time delay of the user task calculation and obtains the user part task unloading scheme with shorter time delay.)

1. TowardsMECThe task power joint optimization method with the minimum time delay is characterized by comprising the following steps:

selecting a relay for task forwarding;

determining that a user is directly offloading a portion of a task toMECThe transmission rate at which the user sends part of the task to the selected relay, and the selected relay forwards part of the task toMECThe transmission rate of (c); calculating user direct transmission unloading time and user-relay unloading time according to the transmission rate, and taking the maximum value of the user direct transmission unloading time and the user-relay unloading time as the user unloading time;

obtaining user local computation time andMECcalculating time;

comparing user local computation time with user-MECUnloading the computation time, wherein the maximum value is the total time delay of the user task computation, the user-MECThe uninstalling calculation time is the user uninstalling time andMECcalculating the sum of the time;

aiming at minimizing the total time delay calculated by the user task, establishing a task power joint optimization problem; and solving the task power joint optimization problem to determine the optimal task allocation proportion and power allocation proportion, thereby obtaining a user part task unloading scheme.

2. A device as claimed in claim 1MECThe task power joint optimization method with the minimum time delay is characterized in that the relay for task forwarding is selected, and the method comprises the following steps:

acquiring user to send part of task to each relay and each relay to forward part of task toMECSmall-scale fading and large-scale transmission loss;

respectively determining corresponding channel gains according to the small-scale fading and the large-scale transmission loss, wherein the channel gains are respectively obtained by the following formulas:

wherein the content of the first and second substances,sending part of a task to a relay for a userThe channel gain of (a) is determined,is a relayForwarding part of the task toMECThe channel gain of (a) is determined,sending part of a task to a relay for a userThe small-scale fading of the signal is reduced,sending part of a task to a relay for a userThe large-scale transmission loss of the optical fiber,is a relayForwarding part of the task toMECThe small-scale fading of the signal is reduced,is a relayForwarding part of the task toMECThe large-scale transmission loss of the optical fiber,，Nis a positive integer;

sending part of a task to a relay based on a userChannel gain and relaying ofForwarding part of the task toMECThe optimal relay for task forwarding is selected by the following formula:

wherein the content of the first and second substances,rin order to perform the selected relay for the task forwarding,is an optional relay set.

3. A device according to claim 2MECThe task power joint optimization method with minimum time delay is characterized in that a user sends part of tasks to a relayLarge scale transmission loss ofAnd relayForwarding part of the task toMECLarge scale transmission loss ofAre respectively obtained by the following formulas:

in the above-mentioned formula,for users to relayThe distance of (a) to (b),is a relayToMECThe distance of (a) to (b),indicating that the user sent part of the task to the relayThe path loss factor of (a) is,is a relayForwarding part of the task toMECThe path loss factor of (2).

4. A device as claimed in claim 1MECTask power joint optimization method with minimum time delay is characterized in that the user directly unloads partial tasks toMECThe transmission rate at which the user sends part of the task to the selected relay, and the selected relay forwards part of the task toMECThe transmission rates of (a) are respectively obtained by the following formulas:

in the above-mentioned formula,offloading part of the task directly to the userMECThe rate of transmission of (a) is,the transmission rate at which the user sends part of the task to the selected relay,forwarding part of the tasks to the selected relayMECThe rate of transmission of (a) is,for the user toMECThe bandwidth of the channel(s) of (c),for the user to forward to through the selected relayMECThe bandwidth of the channel(s) of (c),in order to select the transmit power of the relay,is the transmit power of the user or users,represent the user toMECThe power distribution ratio of (a) to (b),indicating the power allocation ratio of the user to the selected relay,offloading part of the task directly to the userMECThe small-scale fading of the signal is reduced,offloading part of the task directly to the userMECThe large-scale transmission loss of the optical fiber,small scale fading to send part of the task to the selected relay for the user,large scale transmission losses for users sending part of the mission to selected relays,forwarding part of the tasks to the selected relayMECThe small-scale fading of the signal is reduced,forwarding part of the tasks to the selected relayMECThe large-scale transmission loss of the optical fiber,representing gaussian white noise power.

5. A device as claimed in claim 1MECThe task power joint optimization method with the minimum time delay is characterized in that the calculation method of the user direct transmission unloading time is shown as the following formula:

wherein the content of the first and second substances,indicating the user's direct transfer offload time,Da task representing a user is presented to the user,represent the user toMECThe task allocation ratio of (1) is,offloading part of the task directly to the userMECThe transmission rate of (c).

6. A device as claimed in claim 1MECThe task power joint optimization method with minimum time delay is characterized in that the user-relay unloading time comprises the time when the user sends part of tasks to the selected relay and the time when the selected relay forwards part of tasks to the selected relayMECWherein the time calculation method for the user to send part of the task to the selected relay is shown in the following formula:

whereinIndicating the time at which the user sent part of the task to the selected relay,Da task representing a user is presented to the user,indicating that the user is forwarded to by the selected relayMECThe task allocation ratio of (1) is,a transmission rate at which part of the tasks are sent to the selected relay for the user;

selecting a relay to forward part of the task toMECThe time calculation method of (2) is shown by the following formula:

whereinIndicating selected relay forwarding part of task toMECThe time of (a) is,forwarding part of the tasks to the selected relayMECThe transmission rate of (c).

7. A device as claimed in claim 1MECThe task power joint optimization method with minimum time delay is characterized in that the sum of time and power is calculated locally by a userMECThe calculated times are respectively determined by the following formulas:

in the above-mentioned formula,which represents the time of the local calculation of the user,to representMECThe time is calculated and the time is calculated,Da task representing a user is presented to the user,indicating the assignment proportion of tasks calculated locally by the user,indicating user offload toMECThe task allocation ratio of (1) is,indicating the number of CPU cycles required for the user to calculate 1bit,which represents the CPU frequency of the user,to representMECThe number of CPU cycles required to calculate 1bit,to representMECThe CPU frequency of (c).

8. A device as claimed in claim 1MECThe task power joint optimization method with minimum time delay is characterized in that the task power joint optimization problem is formed by the following formulaRepresents:

in the above-mentioned formula,indicating that the user task calculates the total delay,which represents the time of the local calculation of the user,indicating the time of the user's uninstallation,to representMECThe time is calculated and the time is calculated,indicating the user-relay offload time and,indicating the user's direct transfer offload time,represent the user toMECThe task allocation ratio of (1) is,indicating that the user is forwarded to by the selected relayMECThe task allocation ratio of (1) is,indicating user offload toMECThe task allocation ratio of (1) is,represent the user toMECThe power distribution ratio of (1).

9. TowardsMECThe task power joint optimization system with the minimum time delay is characterized by comprising the following steps: the system comprises a relay selection module, a transmission rate determination module, a time calculation module and an optimization problem solving module;

the relay selection module is used for selecting a relay for task forwarding;

the transmission rate determining module is used for determining that the user directly unloads part of tasks toMECThe transmission rate at which the user sends part of the task to the selected relay, and the selected relay forwards part of the task toMECThe transmission rate of (c);

the time calculation module is used for calculating the user direct transmission unloading time and the user-relay unloading time according to the transmission rate determined by the transmission rate determination module, and taking the maximum value of the user direct transmission unloading time and the user-relay unloading time as the user unloading time; obtaining user local computationsTime andMECcalculating time; comparing user local computation time with user-MECUnloading the computation time, wherein the maximum value is the total time delay of the user task computation, the user-MECThe uninstalling calculation time is the user uninstalling time andMECcalculating the sum of the time;

the optimization problem solving module is used for establishing a task power joint optimization problem by taking the total time delay of the user task calculation determined by the time calculation module as a target; and solving the task power joint optimization problem to determine the optimal task allocation proportion and power allocation proportion, thereby obtaining a user part task unloading scheme.

10. A device according to claim 9MECThe task power joint optimization system with the minimum time delay is characterized in that the relay selection module is specifically configured to execute the following steps:

acquiring user to send part of task to each relay and each relay to forward part of task toMECSmall-scale fading and large-scale transmission loss;

respectively determining corresponding channel gains according to the small-scale fading and the large-scale transmission loss, wherein the channel gains are respectively obtained by the following formulas:

wherein the content of the first and second substances,sending part of a task to a relay for a userThe channel gain of (a) is determined,is a relayForwarding part of the task toMECThe channel gain of (a) is determined,sending part of a task to a relay for a userThe small-scale fading of the signal is reduced,sending part of a task to a relay for a userThe large-scale transmission loss of the optical fiber,is a relayForwarding part of the task toMECThe small-scale fading of the signal is reduced,is a relayForwarding part of the task toMECThe large-scale transmission loss of the optical fiber,，Nis a positive integer;

sending part of a task to a relay based on a userChannel gain and relaying ofForwarding part of the task toMECThe optimal relay for task forwarding is selected by the following formula:

wherein the content of the first and second substances,rin order to perform the selected relay for the task forwarding,is an optional relay set.

Technical Field

The invention belongs to the technical field of wireless communication, and particularly relates to a mobile edge-oriented calculation (MEC，Mobile Edge Computing) Task power joint optimization method with minimum time delayAnd a system.

Background

With the rapid development of wireless communication technology, the requirements of users on network experience are higher and higher. The mobile edge computing sinks the computing capability to the mobile edge node, thereby greatly reducing the communication time delay and improving the network experience of users. The mobile edge calculation has a strong application scene, is widely used in the communication field, and has important practical significance for the communication technology.

However, when the user arrivesMECWhen the direct transmission channel gain is poor, the user can not well unload part of the task directly toMECAnd the calculation results in heavy local calculation task, long time and poor user experience.

Disclosure of Invention

The invention aims to provide a method for improving the stability of the traditional Chinese medicineMECThe task power combined optimization method with the minimum time delay can obtain a task unloading scheme of a user part with reasonable task and power distribution and smaller time delay when the user direct transmission channel gain is poor.

In order to solve the technical problems, the invention adopts the following technical scheme:

in one aspect, the present invention provides a method of operating a computer systemMECThe task power joint optimization method with the minimum time delay comprises the following steps:

selecting a relay for task forwarding;

determining that a user is directly offloading a portion of a task toMECThe transmission rate at which the user sends part of the task to the selected relay, and the selected relay forwards part of the task toMECThe transmission rate of (c); calculating user direct transmission unloading time and user-relay unloading time according to the transmission rate, and taking the maximum value of the user direct transmission unloading time and the user-relay unloading time as the user unloading time;

obtaining user local computation time andMECcalculating time;

comparing user local computation time with user-MECUnloading the computation time, wherein the maximum value is the total time delay of the user task computation, the user-MECThe uninstalling calculation time is the user uninstalling time andMECcalculating the sum of the time;

aiming at minimizing the total time delay calculated by the user task, establishing a task power joint optimization problem; and solving the task power joint optimization problem to determine the optimal task allocation proportion and power allocation proportion, thereby obtaining a user part task unloading scheme.

Further, the user directly unloads a portion of the task toMECThe user sends part of tasks to each relay and each relay forwards part of tasks toMECThe large-scale transmission loss of (a) is obtained by the following equations (1) to (3), respectively:

（1）

（2）

（3）

in the above-mentioned formula,offloading part of the task directly to the userMECThe large-scale transmission loss of the optical fiber,sending part of a task to a relay for a userThe large-scale transmission loss of the optical fiber,is a relayForwarding part of the task toMECThe large-scale transmission loss of the optical fiber,dfor the user toMECThe distance of (a) to (b),for users to relayThe distance of (a) to (b),is a relayToMECThe distance of (a) to (b),offloading part of the task directly to the userMECThe path loss factor of (a) is,indicating that the user sent part of the task to the relayThe path loss factor of (a) is,is a relayForwarding part of the task toMECThe path loss factor of (a) is,，Nis a positive integer.

Further, the acquisition user directly unloads part of the task toMECThe user sends part of tasks to each relay and each relay forwards part of tasks toMECThe corresponding channel gains are obtained by the following equations (4) to (6), respectively:

（4）

（5）

（6）

wherein the content of the first and second substances,offloading part of the task directly to the userMECThe channel gain of (a) is determined,sending part of a task to a relay for a userThe channel gain of (a) is determined,is a relayForwarding part of the task toMECThe channel gain of (a) is determined,offloading part of the task directly to the userMECThe small-scale fading of the signal is reduced,offloading part of the task directly to the userMECThe large-scale transmission loss of the optical fiber,sending part of a task to a relay for a userThe small-scale fading of the signal is reduced,sending part of a task to a relay for a userThe large-scale transmission loss of the optical fiber,is a relayForwarding part of the task toMECThe small-scale fading of the signal is reduced,is a relayForwarding part of the task toMECThe large-scale transmission loss of the optical fiber,，Nis a positive integer;

sending part of a task to a relay based on a userChannel gain and relaying ofForwarding part of the task toMECThe optimal relay for task forwarding is selected by the following formula:

（7）

wherein the content of the first and second substances,rin order to perform the selected relay for the task forwarding,is an optional relay set.

Further, it is determined that the user directly offloads a portion of the task toMECThe transmission rate at which the user sends part of the task to the selected relay, and the selected relay forwards part of the task toMECThe transmission rates of (a) are obtained by the following equations (8) to (10), respectively:

（8）

（9）

（10）

in the above-mentioned formula,offloading part of the task directly to the userMECThe rate of transmission of (a) is,the transmission rate at which the user sends part of the task to the selected relay,forwarding part of the tasks to the selected relayMECThe rate of transmission of (a) is,for the user toMECThe bandwidth of the channel(s) of (c),for the user to forward to through the selected relayMECThe bandwidth of the channel(s) of (c),in order to select the transmit power of the relay,is the transmit power of the user or users,represent the user toMECThe power distribution ratio of (a) to (b),indicating the power allocation ratio of the user to the selected relay,offloading part of the task directly to the userMECThe small-scale fading of the signal is reduced,offloading part of the task directly to the userMECThe large-scale transmission loss of the optical fiber,small scale fading to send part of the task to the selected relay for the user,large scale transmission losses for users sending part of the mission to selected relays,forwarding part of the tasks to the selected relayMECThe small-scale fading of the signal is reduced,forwarding part of the tasks to the selected relayMECThe large-scale transmission loss of the optical fiber,representing gaussian white noise power.

Further, the calculation method of the user direct-transmission unloading time is shown in formula (11):

（11）

wherein the content of the first and second substances,indicating the user's direct transfer offload time,Da task representing a user is presented to the user,represent the user toMECThe task allocation ratio of (1) is,offloading part of the task directly to the userMECThe transmission rate of (c).

Further, the user-relay offload time includes a time when the user sends a portion of the task to the selected relay and the selected relay forwards the portion of the task toMECWherein the time calculation method for the user to send part of the task to the selected relay is shown in formula (12):

（12）

whereinIndicating the time at which the user sent part of the task to the selected relay,Da task representing a user is presented to the user,indicating that the user is forwarded to by the selected relayMECThe task allocation ratio of (1) is,a transmission rate at which part of the tasks are sent to the selected relay for the user;

selecting a relay to forward part of the task toMECThe time calculation method of (2) is shown in equation (13):

（13）

whereinIndicating selected relay forwarding part of task toMECThe time of (a) is,forwarding part of the tasks to the selected relayMECThe transmission rate of (c).

Further, the user calculates the sum of time locallyMECThe calculated times are respectively determined by the following formulas:

(16)

(17)

in the above-mentioned formula,which represents the time of the local calculation of the user,to representMECThe time is calculated and the time is calculated,Da task representing a user is presented to the user,indicating the assignment proportion of tasks calculated locally by the user,indicating user offload toMECThe task allocation ratio of (1) is,indicating the number of CPU cycles required for the user to calculate 1bit,which represents the CPU frequency of the user,to representMECThe number of CPU cycles required to calculate 1bit,to representMECThe CPU frequency of (c).

Further, the task power joint optimization problem is represented by the following formula:

（19）

in the above-mentioned formula,indicating that the user task calculates the total delay,indicating user localityThe time is calculated and the time is calculated,indicating the time of the user's uninstallation,to representMECThe time is calculated and the time is calculated,indicating the user-relay offload time and,indicating the user's direct transfer offload time,represent the user toMECThe task allocation ratio of (1) is,indicating that the user is forwarded to by the selected relayMECThe task allocation ratio of (1) is,indicating user offload toMECThe task allocation ratio of (1) is,represent the user toMECThe power distribution ratio of (1).

In a second aspect, the invention also provides a method for making a paper-making articleMECThe task power joint optimization system with the minimum time delay comprises the following steps: the method comprises the following steps: the system comprises a relay selection module, a transmission rate determination module, a time calculation module and an optimization problem solving module;

the relay selection module is used for selecting a relay for task forwarding;

the transmission rate determining module is used for determining that the user directly unloads part of tasks toMECThe transmission rate at which the user sends part of the task to the selected relay, and the selected relay forwards part of the task toMECThe transmission rate of (c);

the time calculation module is used for calculating the user direct transmission unloading time and the user-relay unloading time according to the transmission rate determined by the transmission rate determination module, and taking the maximum value of the user direct transmission unloading time and the user-relay unloading time as the user unloading time; obtaining user local computation time andMECcalculating time; comparing user local computation time with user-MECUnloading the computation time, wherein the maximum value is the total time delay of the user task computation, the user-MECThe uninstalling calculation time is the user uninstalling time andMECcalculating the sum of the time;

the optimization problem solving module is used for establishing a task power joint optimization problem by taking the total time delay of the user task calculation determined by the time calculation module as a target; and solving the task power joint optimization problem to determine the optimal task allocation proportion and power allocation proportion, thereby obtaining a user part task unloading scheme.

Further, the relay selection module is specifically configured to perform the following steps:

acquiring user to send part of task to each relay and each relay to forward part of task toMECSmall-scale fading and large-scale transmission loss;

respectively determining corresponding channel gains according to the small-scale fading and the large-scale transmission loss, wherein the channel gains are respectively obtained by the following formulas:

wherein the content of the first and second substances,sending part of a task to a relay for a userThe channel gain of (a) is determined,is a relayForwarding part of the task toMECThe channel gain of (a) is determined,sending part of a task to a relay for a userThe small-scale fading of the signal is reduced,sending part of a task to a relay for a userThe large-scale transmission loss of the optical fiber,is a relayForwarding part of the task toMECThe small-scale fading of the signal is reduced,is a relayForwarding part of the task toMECThe large-scale transmission loss of the optical fiber,，Nis a positive integer;

sending part of a task to a relay based on a userChannel gain and relaying ofForwarding part of the task toMECThe optimal relay for task forwarding is selected by the following formula:

wherein the content of the first and second substances,rin order to perform the selected relay for the task forwarding,is an optional relay set.

The invention has the following beneficial technical effects: the invention provides a method for realizing the aimMECRelay selection and task power combined optimization method with minimum time delay from user to userMECWhen the direct transmission channel gain is poor, partial tasks are unloaded in the direct transmission of the user through relay selectionMECWhile calculating, selecting relay to forward partial task to unloading partMECAnd calculating, and performing task and power combined optimization on the basis to achieve the optimal distribution of tasks and power, and the invention minimizes the total time delay of user task calculation and can obtain a user part task unloading scheme with shorter time delay.

The invention directly unloads part of tasks to according to usersMECThe user sends part of tasks to each relay and each relay forwards part of tasks toMECSelecting a relay capable of providing the best channel gain to forward part of the task toMECAnd the total time delay of the task calculation of the user is further shortened by calculation.

Drawings

FIG. 1 is a system model diagram according to an embodiment of the present invention;

FIG. 2 is a schematic flow chart of an embodiment of the present invention;

FIG. 3 is a simulation diagram of an embodiment of the present invention.

Detailed Description

The invention is further described with reference to specific examples. The following examples are only for illustrating the technical solutions of the present invention more clearly, and should not be taken as limiting the scope of the present invention.

As shown in figures 1 and 2, the present invention is designed in a way thatMECThe relay selection and task power combined optimization method with the minimum time delay comprises a base station 1,MEC2. Multiple relays 3 and users 4, with simultaneous local computation and partial task offloading toMECComputing, in which part of the task is offloaded toMECComputing involves a user directly offloading a portion of a task toMECComputation and user-through-relay forwarding offload of part of tasks toMECTwo calculation modes specifically comprise the following steps:

obtaining user direct offload of partial task toMECThe user sends part of task to each relay and each relay forwards part of task toMECSmall-scale fading and large-scale transmission loss;

performing relay selection, optionally, combining the obtained small-scale fading and large-scale transmission loss on the channel, and selecting the relay capable of providing the optimal channel gain to perform task forwarding;

according to the result of relay selection, the user directly unloads part of tasksMECThe user sends part of the task to the selected relay and the selected relay forwards part of the task toMECThe transmission rate of (c);

calculating the time to offload based on the transmission rate, and considering that only offloaded tasks will be forwarded by the relay after all of them reach the relay, the user-relay offload time thus includes the time for the user to send part of the tasks to the selected relay and the time for the selected relay to forward part of the tasks to the selected relayMECTime of (d). In view of unloading toMECThe task of the computation is all reachedMECAfter that, the air conditioner is started to work,MECthe calculation is started and thus, the maximum value of the user-direct offload time and the user-relay offload time is taken as the user offload time.

Calculating the total time delay of the user task by consideringMECThe amount of result data returned by the calculation is small,MECthe result return delay of the server is negligible. Therefore, users-MECThe uninstalling calculation time is the user uninstalling time andMECcalculating the sum of time, comparing the user's local calculated timeAnd user-MECUnloading the calculation time, wherein the maximum value is the total time delay calculated by the user task;

and the total time delay of the user task calculation is minimized through a task and power combined optimization method.

The large-scale transmission loss in the invention is obtained by the following formula:

（1）

（2）

（3）

in the above-mentioned formula,offloading part of the task directly to the userMECThe large-scale transmission loss of the optical fiber,sending part of a task to a relay for a userThe large-scale transmission loss of the optical fiber,is a relayForwarding part of the task toMECThe large-scale transmission loss of the optical fiber,dfor the user toMECThe distance of (a) to (b),for users to relayThe distance of (a) to (b),is a relayToMECThe distance of (a) to (b),offloading part of the task directly to the userMECThe path loss factor of (a) is,indicating that the user sent part of the task to the relayThe path loss factor of (a) is,is a relayForwarding part of the task toMECThe path loss factor of (a) is,，Nis a positive integer.

The relay selection of the invention is characterized in that the system is provided with a plurality of relays, the relays which can provide the best channel gain are selected to carry out task forwarding by combining the obtained small-scale fading and large-scale transmission loss, and the method for selecting the relays for task forwarding comprises the following steps:

obtaining user direct offload of partial task toMECThe user sends part of tasks to each relay and each relay forwards part of tasks toMECThe corresponding channel gains are obtained by the following equations (4) to (6), respectively:

（4）

（5）

（6）

wherein the content of the first and second substances,offloading part of the task directly to the userMECThe channel gain of (a) is determined,sending part of a task to a relay for a userThe channel gain of (a) is determined,is a relayForwarding part of the task toMECThe channel gain of (a) is determined,offloading part of the task directly to the userMECThe small-scale fading of the signal is reduced,offloading part of the task directly to the userMECThe large-scale transmission loss of the optical fiber,sending part of a task to a relay for a userThe small-scale fading of the signal is reduced,sending part of a task to a relay for a userThe large-scale transmission loss of the optical fiber,is a relayForwarding part of the task toMECThe small-scale fading of the signal is reduced,is a relayForwarding part of the task toMECThe large-scale transmission loss of the optical fiber,，Nis a positive integer.

Sending part of a task to a relay based on a userChannel gain and relaying ofForwarding part of the task toMECThe optimal relay for task forwarding is selected by the following formula:

（7）

wherein the content of the first and second substances,rin order to perform the selected relay for the task forwarding,is an optional relay set.

In the invention, after relay selection, a user directly unloads part of tasks toMECThe user sends part of the task to the selected relay and the selected relay forwards part of the task toMECIs obtained by the following formula:

（8）

（9）

（10）

in the above-mentioned formula,offloading part of the task directly to the userMECThe rate of transmission of (a) is,the transmission rate at which the user sends part of the task to the selected relay,forwarding part of the tasks to the selected relayMECThe rate of transmission of (a) is,for the user toMECThe bandwidth of the channel(s) of (c),for the user to forward to through the selected relayMECThe bandwidth of the channel(s) of (c),transmitting for selected relaysThe power of the electric motor is controlled by the power controller,is the transmit power of the user or users,represent the user toMECThe power distribution ratio of (a) to (b),indicating the power allocation ratio of the user to the selected relay,offloading part of the task directly to the userMECThe small-scale fading of the signal is reduced,offloading part of the task directly to the userMECThe large-scale transmission loss of the optical fiber,small scale fading to send part of the task to the selected relay for the user,large scale transmission losses for users sending part of the mission to selected relays,forwarding part of the tasks to the selected relayMECThe small-scale fading of the signal is reduced,forwarding part of the tasks to the selected relayMECThe large-scale transmission loss of the optical fiber,representing gaussian white noise power.

The user unloading time in the invention is obtained by the following formula:

（11）

（12）

（13）

（14）

（15）

in the above-mentioned formula,indicating the user's direct transfer offload time,indicating the time at which the user sent part of the task to the selected relay,indicating selected relay forwarding part of task toMECThe time of (a) is,indicating the user-relay offload time and,indicating the time of the user's uninstallation,Da task representing a user is presented to the user,represent the user toMECThe task allocation ratio of (1) is,indicating that the user is forwarded to by the selected relayMECThe task allocation ratio of (1) is,offloading part of the task directly to the userMECThe rate of transmission of (a) is,the transmission rate at which the user sends part of the task to the selected relay,forwarding part of the tasks to the selected relayMECThe transmission rate of (c).

The user task in the invention calculates the total time delay, and is obtained by the following formula:

（16）

（17）

（18）

in the above-mentioned formula,which represents the time of the local calculation of the user,to representMECThe time is calculated and the time is calculated,indicating the user-relay offload time and,indicating the user's direct transfer offload time,indicating the time of the user's uninstallation,indicating that the user task calculates the total delay,Da task representing a user is presented to the user,indicating the assignment proportion of tasks calculated locally by the user,indicating user offload toMECThe task allocation ratio of (1) is,indicating the number of CPU cycles required for the user to calculate 1bit,which represents the CPU frequency of the user,to representMECThe number of CPU cycles required to calculate 1bit,to representMECThe CPU frequency of (c).

In the invention, by the task power joint optimization method, the total time delay of the user task calculation is minimized, and a task power joint optimization problem is established, wherein the optimization problem (namely an objective function) is obtained by the following formula:

（19）

the objective function formula can be treated by the following four optimization problems:

in the above-mentioned formula,indicating that the user task calculates the total delay,which represents the time of the local calculation of the user,indicating the time of the user's uninstallation,to representMECThe time is calculated and the time is calculated,indicating the user-relay offload time and,indicating the user's direct transfer offload time,represent the user toMECThe task allocation ratio of (1) is,indicating that the user is forwarded to by the selected relayMECThe task allocation ratio of (1) is,indicating user offload toMECThe task allocation ratio of (1) is,represent the user toMECThe power distribution ratio of (1).

The invention provides a method for realizing the aimMECMinimum delayRelay selection and task power joint optimization method in user-to-userMECWhen the direct transmission channel gain is poor, partial tasks are unloaded in the direct transmission of the user through relay selectionMECWhile calculating, selecting relay to forward partial task to unloading partMECAnd calculating, and performing task and power combined optimization on the basis to achieve optimal task and power distribution, and the invention minimizes the total time delay of user task calculation and can obtain a user part task unloading scheme with shorter time delay.

An example of the implementation of the invention on a computer using MATLAB language simulation is given below. User directly uninstalls part of task toMECThe user sends part of task to each relay and each relay forwards part of task toMECAre subject to rayleigh fading, and have a variance of 1,dfor the user toMECThe distance between the relay nodes is 500m, the number of the optional relays is 6,the values of the distances from the user to each relay are respectively 100m, 98m, 102m, 104m, 96m and 105m,for each relay toMECThe distances of (a) are respectively 401m, 403m, 399m, 398m, 406m and 396m,offloading part of the task directly to the userMECThe path loss factor of (a) is,representing the path loss factor at which the user sends part of the task to each relay,for each relay forwarding part of the task toMECThe values of the path loss factors of (1) are all 4,，the number of CPU cycles required for the user to calculate 1bit is 1000cycles/bit,for the CPU frequency of the user, take the value ofcycles/s，Is composed ofMECThe number of CPU cycles required for calculating 1bit is 1000cycles/bit,is composed ofMECHas a CPU frequency ofcycles/s, user transmit powerIs 1w, the relay transmission power is selectedIs 10w, noise powerIs composed ofw, user toMECOf the channel bandwidthAt 1MHz, the users are forwarded to via relaysMECOf the channel bandwidthIs 1 MHz. FIG. 3 shows a scheme proposed by the present inventionThe optimization task allocation under the relay selection equal power allocation and the optimization task allocation under the relay selection water filling algorithm power allocation are compared, independent variables in the graph are tasks, and the graph shows that the scheme provided by the invention is smaller than the two comparison schemes in time delay, and the advantage is gradually obvious along with the increase of the tasks.

The invention directly unloads part of tasks to according to usersMECThe user sends part of tasks to each relay and each relay forwards part of tasks toMECSmall scale fading and large scale transmission loss, selecting a relay forwarding offload component task that provides the best channel gain toMECAnd the total time delay of the task calculation of the user is further shortened by calculation.

An aspect similar to that provided in the above embodimentMECCorrespondingly, the invention also provides a task power joint optimization method oriented to the minimum time delayMECThe task power joint optimization system with the minimum time delay comprises the following steps: the system comprises a relay selection module, a transmission rate determination module, a time calculation module and an optimization problem solving module;

the relay selection module is used for selecting a relay for task forwarding;

the transmission rate determining module is used for determining that the user directly unloads part of tasks toMECThe transmission rate at which the user sends part of the task to the selected relay, and the selected relay forwards part of the task toMECThe transmission rate of (c);

the time calculation module is used for calculating the user direct transmission unloading time and the user-relay unloading time according to the transmission rate determined by the transmission rate determination module, and taking the maximum value of the user direct transmission unloading time and the user-relay unloading time as the user unloading time; acquiring local calculation time of a user and MEC calculation time; comparing user local computation time with user-MECUnloading the computation time, wherein the maximum value is the total time delay of the user task computation, the user-MECThe uninstalling calculation time is the user uninstalling time andMECcalculating the sum of the time;

the optimization problem solving module is used for establishing a task power joint optimization problem by taking the total time delay calculated by the minimized user task as a target; and solving the task power joint optimization problem to determine the optimal task allocation proportion and power allocation proportion, thereby obtaining a user part task unloading scheme.

Further, the relay selection module is specifically configured to perform the steps of the method:

obtaining user direct offload of partial task toMECThe user sends part of tasks to each relay and each relay forwards part of tasks toMECThe corresponding channel gains are obtained by the following equations (4) to (6), respectively:

（4）

（5）

（6）

wherein the content of the first and second substances,offloading part of the task directly to the userMECThe channel gain of (a) is determined,sending part of a task to a relay for a userThe channel gain of (a) is determined,is a relayForwarding part of the task toMECThe channel gain of (a) is determined,offloading part of the task directly to the userMECThe small-scale fading of the signal is reduced,offloading part of the task directly to the userMECThe large-scale transmission loss of the optical fiber,sending part of a task to a relay for a userThe small-scale fading of the signal is reduced,sending part of a task to a relay for a userThe large-scale transmission loss of the optical fiber,is a relayForwarding part of the task toMECThe small-scale fading of the signal is reduced,is a relayForwarding part of the task toMECThe large-scale transmission loss of the optical fiber,，Nis a positive integer;

sending part of a task to a relay based on a userChannel gain and relaying ofForwarding part of the task toMECThe optimal relay for task forwarding is selected by the following formula:

（7）

wherein the content of the first and second substances,rin order to perform the selected relay for the task forwarding,for the set of alternative relays to be a positive integer,sending part of a task to a relay for a userThe channel gain of (a) is determined,is a relayForwarding part of the task toMECThe channel gain of (1).

It should be noted that, for convenience and brevity of description, it may be clearly understood by those skilled in the art that the specific working processes of the above-described systems, apparatuses and units may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.

As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.

The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

While the present invention has been described with reference to the embodiments shown in the drawings, the present invention is not limited to the embodiments, which are illustrative and not restrictive, and it will be apparent to those skilled in the art that various changes and modifications can be made therein without departing from the spirit and scope of the invention as defined in the appended claims.