A resource allocation method for wireless power supply edge computing network

Abstract

The present invention discloses a resource allocation method for a wireless power supply edge computing network. According to a set step size, the current power supply duration ratio is selected in turn from the value range of the power supply duration ratio, and iteration is performed to determine the energy ratio and task unloading communication duration ratio of each wireless device for task unloading communication under the current power supply duration ratio, and the power supply duration ratio when the total computing rate of the wireless power supply edge computing network is the maximum is selected, and the radio frequency energy transmitter is set to wirelessly power the wireless device at the beginning of each time frame; each wireless device is configured to perform task unloading communication in each time frame with the task unloading communication duration and energy allocation ratio corresponding to the selected power supply duration ratio. The present invention can quickly calculate the energy allocation and time allocation scheme of partial unloading, and also obtain a higher network computing rate.

Description

Technical Field

[0001] The present application belongs to the field of wireless power supply technology, and in particular, relates to a resource allocation method for a wireless power supply edge computing network. Background Art

[0002] The development of the Internet of Things has led to more and more human-computer interactions in emerging applications, including smart homes and autonomous driving. Many new high-performance applications rely on real-time communication and heavy computing, but IoT nodes usually have limited power and low computing power and cannot support high-performance computing applications. Therefore, solving these two limitations is one of the key issues to improve the performance of IoT applications.

[0003] Radio frequency-based wireless power transmission technology (WPT) provides a feasible method to solve the problem of energy shortage in IoT networks. Mobile edge computing technology (MEC) can provide high-performance computing services for IoT nodes with low computing power. Wireless power edge computing network (WP-MEC) can solve the problems of energy shortage and computing power limitation in IoT networks. Specifically, energy transmitters and edge computing servers are deployed at the edge of the network. Wireless devices capture radio frequency energy and rely on the collected energy to offload computing tasks to edge computing servers and calculate the rest of the tasks locally. The combination of wireless energy transmission and mobile edge computing achieves sustainable network operation, significantly extends the life of wireless networks, and enhances the computing and communication capabilities of wireless edge devices.

[0004] In edge computing networks, it is necessary to consider the offloading strategy of wireless devices, such as the allocation of communication time within a time frame, which will determine whether a network is efficient. Each IoT wireless device follows a partial offloading strategy, that is, the task can be divided, one part of the task is executed locally on the wireless device, and the other part of the task is offloaded to the edge server for calculation. There are also many indicators to evaluate whether a network is efficient, such as minimum delay, maximum rate, etc.

[0005] Currently, most time allocation methods for wireless power edge computing networks use traditional optimization methods, but the solution process is usually complicated and requires multiple iterations to obtain the optimal solution, which consumes a lot of time. This is unacceptable for applications that require high-performance computing services. Summary of the invention

[0006] The purpose of this application is to provide a resource allocation method and device for a wireless power supply edge computing network, which is used to overcome the above technical difficulties and achieve a higher computing rate.

[0007] In order to achieve the above purpose, the technical solution of this application is as follows:

[0008] A method for allocating resources of a wireless power supply edge computing network, the wireless power supply edge computing network comprising a wireless device and a gateway, the gateway comprising a radio frequency energy transmitter and an edge computing server, the method for allocating resources of the wireless power supply edge computing network comprising:

[0009] Step S1: According to the set step size, the current energy supply duration ratio is selected from the value range of the energy supply duration ratio in turn, and the following iterations are performed to determine the energy ratio and the task offloading communication duration ratio of each wireless device under the current energy supply duration ratio;

[0010] Step S1.1, selecting a first wireless device from the wireless power supply edge computing network, and initializing a range of energy proportions used by the first wireless device for task offloading communication;

[0011] Step S1.2: Set the energy proportion x1 used by the first wireless device for task offloading communication to the average of the upper limit and the lower limit of the range, and calculate the energy proportion used by any other wireless device for task offloading communication according to the following formula:

[0012]

[0013] Wherein, x1 is the energy ratio of the first wireless device used for task offloading communication, x i is the energy ratio of the ith wireless device used for task offloading communication, h1 is the channel gain of the first wireless device, and h i is the channel gain of the i-th wireless device;

[0014] Step S1.3, calculate the wireless device task offloading communication duration ratio according to the following formula:

[0015]

[0016] Among them, t i is the ratio of the communication duration of the i-th wireless device task offloading, B is the wireless communication bandwidth, φ is the number of CPU cycles required for the wireless device to locally process a one-bit task, and k e is the calculation energy efficiency coefficient of the wireless device, N0 is the noise power, a is the energy supply duration ratio, μ is the energy capture efficiency, and P is the RF energy transmission power of the RF energy transmitter;

[0017] Step S1.4: When is greater than the search precision ξ, if Then, x1 is used as the upper limit of the range of the energy ratio of the first wireless device used for task offloading communication, the range of the energy ratio of the first wireless device used for task offloading communication is updated and the process returns to step S1.2; otherwise, x1 is used as the lower limit of the range of the energy ratio of the first wireless device used for task offloading communication, the range of the energy ratio of the first wireless device used for task offloading communication is updated and the process returns to step S1.2. When it is less than or equal to the search accuracy ξ, the total computing rate of the wireless power supply edge computing network is calculated and the iteration ends;

[0018] Step S2, select the power supply time ratio when the total computing rate of the wireless power supply edge computing network is the maximum, set the RF energy transmitter to wirelessly power the wireless device at the beginning of each time frame; configure each wireless device to perform task offloading communication within each time frame with the task offloading communication time and energy allocation ratio corresponding to the selected power supply time ratio, and configure each wireless device to perform local task calculation with the remaining energy within each time frame.

[0019] Furthermore, the total computing rate of the wireless power supply edge computing network is calculated as follows:

[0020]

[0021] Wherein, N is the number of wireless devices.

[0022] Furthermore, each wireless device performs task offloading communication in a time division multiplexing manner.

[0023] Furthermore, the energy supply duration ratio has a value range of [0,1], with a step size of 0.01.

[0024] A resource allocation method for a wireless power supply edge computing network proposed in this application, compared with traditional optimization methods, can solve a complex non-convex problem in a wireless power supply edge computing network using TDMA communication mode and partial unloading mode, and can quickly calculate the energy allocation and time allocation plan of partial unloading, while also obtaining a higher network computing rate. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 This is a schematic diagram of a wireless power supply edge computing network based on TDMA;

[0026] Figure 2 This is a flow chart of the resource allocation method for the wireless power supply edge computing network of this application. DETAILED DESCRIPTION

[0027] To make the objectives, technical solutions, and advantages of this application more clear, the following further elaborates on this application in combination with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not used to limit this application.

[0028] As Figure 1 shown, the wireless power supply edge computing network includes wireless devices and a gateway. The gateway includes a radio frequency energy transmitter and an edge computing server. The radio frequency energy transmitter is used to supply power to the wireless devices, and the edge computing server is used to calculate the tasks unloaded from the wireless devices.

[0029] In one embodiment, as Figure 2 shown, a resource allocation method for a wireless power supply edge computing network is provided, including:

[0030] Step S1: Sequentially select the current power supply duration ratio from the value range of the power supply duration ratio according to a set step size, and perform the following iteration to determine the energy ratio and task offloading communication duration ratio used by each wireless device for task offloading communication in the case of the current power supply duration ratio.

[0031] This embodiment is described by taking a wireless power supply edge computing network composed of one gateway and 10 wireless devices as an example. The gateway integrates a radio frequency energy transmitter and an edge computing server. The task calculation of the wireless devices works based on time frames, and T is the length of one time frame.

[0032] At the beginning of each time frame, the gateway wirelessly supplies power to 10 wireless devices. The wireless power supply duration is a times the frame length, that is, the power supply duration is aT, where 0 < a < 1, and a is the power supply duration ratio.

[0033] The energy captured by the i-th wireless device is expressed as E i = μPh i aT, where μ is the energy capture efficiency, P is the radio frequency energy emission power of the radio frequency energy transmitter, and h i is the channel gain of the i-th device in the current time frame, i ∈ {1, 2,..., 10}, and h = [h1, h2,..., h 10 represents the channel gains of 10 wireless devices within one time frame.

[0034] In this embodiment, within each time frame, the i-th device uses (1 - x i )E i of the energy for local calculation for a duration of T, and uses x i E i of the energy for task offloading communication. x i is the energy ratio for task offloading communication, and 1 - x i is the energy ratio for local calculation.

[0035] In each time frame, aT is used to supply energy. In the remaining time (1-a)T, 10 devices use part of the captured energy to perform task offloading based on wireless communication. The task offloading communication of the 10 devices is based on time division multiplexing. The i-th wireless device uses x i E i The energy of 10 wireless devices is used to offload the task. The offload energy allocation ratio of 10 wireless devices is x = [x1, x2, … x 10 ], the communication duration of the i-th device is t i times the frame length, the communication time of 10 wireless devices is t = [t1T, t2T, …t 10 T].

[0036] In this embodiment, the energy supply duration ratio has a value range of [0, 1], with a step size of 0.01. The energy supply duration ratio takes values ​​in [0, 1] in steps of 0.01, that is, different values ​​of the energy supply duration ratio a are 0.01, 0.02, ..., 0.99. The adopted step size can be set to other values ​​according to actual conditions, such as 0.02.

[0037] When the current energy supply duration ratio is determined, the energy ratio and task offloading communication duration ratio of each wireless device used for task offloading communication are calculated by the following steps:

[0038] Step S1.1: select a first wireless device from a wireless powered edge computing network, and initialize a range of energy proportions used by the first wireless device for task offloading communication.

[0039] For example, initialize the range of the transmission task energy allocation ratio x1 of the first wireless device, x 1max Indicates the upper limit of x1, initialize it to x 1max =1,x 1min Represents the lower limit of x1, initialized to x 1min =0.

[0040] Step S1.2: Set the energy proportion x1 used by the first wireless device for task offloading communication to the average of the upper limit and the lower limit of the range, and calculate the energy proportion used by any other wireless device for task offloading communication according to the following formula:

[0041]

[0042] Wherein, x1 is the energy ratio of the first wireless device used for task offloading communication, x i is the energy ratio of the ith wireless device used for task offloading communication, h1 is the channel gain of the first wireless device, and h i is the channel gain of the i-th wireless device.

[0043] In this embodiment, x1 is initialized to

[0044] according to Get x i The value of , where i∈{2,3,…,10}, can be used to obtain the energy proportion of each wireless device used for task offloading communication.

[0045] Step S1.3, calculate the wireless device task offloading communication duration ratio according to the following formula:

[0046]

[0047] Among them, t i is the ratio of the communication duration of the i-th wireless device task offloading, B is the wireless communication bandwidth, φ is the number of CPU cycles required for the wireless device to locally process a one-bit task, k e is the calculated energy efficiency coefficient of the wireless device, N0 is the noise power, a is the energy supply time ratio, μ is the energy capture efficiency, and P is the RF energy transmission power of the RF energy transmitter.

[0048] By substituting the above parameters into the formula, the ratio of communication duration of each wireless device task offloading can be obtained.

[0049] Step S1.4: When is greater than the search precision ξ, if Then, x1 is used as the upper limit of the range of the energy ratio of the first wireless device used for task offloading communication, the range of the energy ratio of the first wireless device used for task offloading communication is updated and the process returns to step S1.2; otherwise, x1 is used as the lower limit of the range of the energy ratio of the first wireless device used for task offloading communication, the range of the energy ratio of the first wireless device used for task offloading communication is updated and the process returns to step S1.2. When it is less than or equal to the search accuracy ξ, the total computing rate of the wireless power supply edge computing network is calculated and the iteration ends.

[0050] In this embodiment, the number of wireless devices is N. When the search accuracy ξ is set, through the above judgment, When it is greater than the search accuracy ξ, the range of the energy proportion used by the first wireless device for task offloading communication is updated, and the process returns to step S1.2 to recalculate.

[0051] Specifically, if Then set x 1max = x1, return to step S1.2, and calculate again the energy ratio and task offloading communication duration ratio of each wireless device used for task offloading communication. Then set x 1min=x1, return to step S1.2, and calculate again the energy ratio and task offloading communication duration ratio of each wireless device used for task offloading communication.

[0052] until When it is less than or equal to the search accuracy ξ, the total computing rate of the wireless power supply edge computing network is calculated and the iteration ends.

[0053] Step S2, select the power supply time ratio when the total computing rate of the wireless power supply edge computing network is the maximum, set the RF energy transmitter to wirelessly power the wireless device at the beginning of each time frame; configure each wireless device to perform task offloading communication within each time frame with the task offloading communication time and energy allocation ratio corresponding to the selected power supply time ratio, and configure each wireless device to perform local task calculation with the remaining energy within each time frame.

[0054] For any selected power supply duration ratio a, the total computing rate of the wireless power supply edge computing network, as well as the corresponding energy ratio and task offloading communication duration ratio of each wireless device used for task offloading communication can be calculated. Finally, the power supply duration ratio when the total computing rate is the largest, and the corresponding energy ratio and task offloading communication duration ratio of each wireless device used for task offloading communication are selected as the parameters of the wireless edge computing network for operation.

[0055] Specifically, the RF energy transmitter is set to wirelessly supply energy to the wireless device at the beginning of each time frame according to the selected energy supply duration ratio. Within the energy supply duration aT after the start of the current time frame, it belongs to the energy supply time, and the RF energy transmitter wirelessly supplies energy to the wireless device.

[0056] Each wireless device is configured to perform task offloading communication in each time frame with the task offloading communication duration corresponding to the selected energy supply duration ratio and the energy ratio used for task offloading communication. Each wireless device is configured to perform local task calculation in each time frame with the remaining energy.

[0057] This application configures the wireless power supply edge computing network to work according to the calculated power supply time, the communication time of each wireless device task offloading and the occupied energy allocation ratio.

[0058] Each wireless device performs task offloading according to the calculated task offloading communication duration. All wireless devices perform task offloading communication through time division multiplexing. The communication duration of the i-th wireless device is t i T. And the energy is allocated according to the calculated energy ratio for task offloading communication. The i-th wireless device uses x i E i energy to offload communication.

[0059] Each wireless device also performs local task calculations. In the entire time frame, the remaining energy can be used for local task calculations. The i-th device uses (1-x i )E i The remaining energy is calculated locally for T time.

[0060] In a specific embodiment, for a given power supply duration ratio a, the total computing rate Q(x, t) of the wireless power supply edge computing network is:

[0061]

[0062] Where φ is the number of CPU cycles required for the wireless device to process one bit of task locally, k e is the computational energy efficiency coefficient of the wireless device, B is the wireless communication bandwidth, N0 is the noise power; Q(x,t) is a i ,t i A concave function that satisfies 0≤t i ≤1, and 0≤x i ≤1; the existing methods for solving convex optimization problems such as interior point method and Lagrange dual method are used to obtain the energy allocation ratio x and communication time t that maximize Q(x,t) under given a.

[0063] Where N is the number of wireless devices, t i is the task offloading communication duration ratio of the i-th wireless device, and the communication duration of the i-th wireless device is t i T.x i is the proportion of energy used for task offloading communication.

[0064] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.

Claims

1. A resource allocation method for a wireless power supply edge computing network, wherein the wireless power supply edge computing network includes a wireless device and a gateway, wherein the gateway includes a radio frequency energy transmitter and an edge computing server, characterized in that: The resource allocation method of the wireless power supply edge computing network includes: Step S1: According to the set step size, the current energy supply duration ratio is selected from the value range of the energy supply duration ratio in turn, and the following iterations are performed to determine the energy ratio and the task offloading communication duration ratio of each wireless device under the current energy supply duration ratio; Step S1.1, selecting a first wireless device from the wireless power supply edge computing network, and initializing a range of energy proportions used by the first wireless device for task offloading communication; Step S1.2: Set the energy proportion x1 used by the first wireless device for task offloading communication to the average of the upper limit and the lower limit of the range, and calculate the energy proportion used by any other wireless device for task offloading communication according to the following formula: Wherein, x1 is the energy ratio of the first wireless device used for task offloading communication, x i is the energy ratio of the ith wireless device used for task offloading communication, h1 is the channel gain of the first wireless device, and h i is the channel gain of the i-th wireless device; Step S1.3, calculate the wireless device task offloading communication duration ratio according to the following formula: Among them, t i is the ratio of the communication duration of the i-th wireless device task offloading, B is the wireless communication bandwidth, φ is the number of CPU cycles required for the wireless device to locally process a one-bit task, k e is the calculation energy efficiency coefficient of the wireless device, N0 is the noise power, a is the energy supply duration ratio, μ is the energy capture efficiency, and P is the RF energy transmission power of the RF energy transmitter; Step S1.4: When is greater than the search precision ξ, if Then, x1 is used as the upper limit of the range of the energy ratio of the first wireless device used for task offloading communication, the range of the energy ratio of the first wireless device used for task offloading communication is updated and the process returns to step S1.2; otherwise, x1 is used as the lower limit of the range of the energy ratio of the first wireless device used for task offloading communication, the range of the energy ratio of the first wireless device used for task offloading communication is updated and the process returns to step S1.

2. When it is less than or equal to the search accuracy ξ, the total computing rate of the wireless power supply edge computing network is calculated and the iteration ends; Step S2, select the power supply time ratio when the total computing rate of the wireless power supply edge computing network is the maximum, set the RF energy transmitter to wirelessly power the wireless device at the beginning of each time frame; configure each wireless device to perform task offloading communication within each time frame with the task offloading communication time and energy allocation ratio corresponding to the selected power supply time ratio, and configure each wireless device to perform local task calculation with the remaining energy within each time frame.

2. The resource allocation method for a wireless power supply edge computing network according to claim 1, characterized in that: The total computing rate of the wireless power supply edge computing network is calculated as follows: Wherein, N is the number of wireless devices.

3. The resource allocation method for a wireless power supply edge computing network according to claim 1, characterized in that: Each wireless device performs task offloading communication in a time division multiplexing manner.

4. The resource allocation method for a wireless power supply edge computing network according to claim 1, characterized in that: The energy supply duration ratio has a value range of [0,1], and a step size of 0.01.

Images