Method for allocating energy harvesting duration and computing tasks in a wireless powered edge network

Abstract

The application discloses a method for allocating energy capture duration and computing task in a wireless energy supply edge network, comprising: outputting a time proportion set by using a one-dimensional exhaustive search method; calculating an energy proportion corresponding to each time proportion in the time proportion set based on the time proportion; calculating the computing rate sum of N wireless devices according to each time proportion and the corresponding energy proportion in the time proportion set, taking the maximum computing rate sum as the optimal time proportion and energy proportion of the wireless device for this time computing; and completing the computing task offloading distribution according to the optimal time proportion and energy proportion. The application optimizes the energy capture duration and computing task allocation, has smaller time consumption, can achieve higher computing rate, and is suitable for time-varying channel states.

Description

Technical Field

[0001] This invention belongs to the field of mobile edge computing and is applicable to the use of Orthogonal Frequency Division Multiple Access (OFDMA) communication mode and partial offloading mode to determine the allocation scheme of wireless energy capture time and computing task offloading in wireless powered edge computing networks, so as to achieve a higher computing rate. Background Technology

[0002] With the development of the Internet of Things (IoT) market, various novel technologies are constantly emerging, and different applications are being updated and iterated with the advent of new technologies. The number of IoT connections has also seen tremendous growth, leading to an explosive increase in the amount of data generated by IoT terminals. Consequently, the energy supply requirements and latency requirements for IoT are also becoming increasingly stringent. In traditional IoT, the energy reserves and computing power of battery power are limited, making it impossible to handle long-duration, high-data-volume computing tasks. Therefore, addressing the high energy supply requirements and limited computing power—these two performance limitations—is a key challenge in current IoT technology research.

[0003] The introduction of Power over Wireless Communication Networks (WPCNs) and edge computing provides technical feasibility for such IoT scenarios. WPCNs are a novel network model that utilizes wireless power transfer (WPT) technology to remotely charge wireless communication devices, significantly improving the performance of traditional battery-powered communication networks, such as higher throughput, longer device lifespan, and lower network costs. Mobile edge computing (MEC) is a network architecture concept that provides information technology and cloud computing capabilities at the edge of mobile networks, offering a service environment with ultra-low latency, high bandwidth, and direct access to real-time network information. Combining WPCNs with mobile edge computing yields Power over Wireless Mobile Edge Computing (WP-MEC) networks. WP-MEC networks possess the advantages of both WPCNs and MECs, providing an ideal solution to the energy and computing resource constraints of wireless devices.

[0004] To obtain ideal solutions to energy and computational resource constraints, various traditional optimization methods have evolved, such as one-dimensional search algorithms, algorithms that use all collected energy to offload computational tasks to ECS (called pure edge computing algorithms), and algorithms that use all harvested energy to perform local computation on IoT nodes (called pure node computing algorithms).

[0005] However, the channel state of wireless devices is time-varying throughout the edge computing task. Therefore, it is crucial to quickly determine the offloading strategy under the current channel state. This is very challenging for traditional optimization methods, as they usually require multiple iterations to obtain a good or optimal solution, resulting in long computational delays and making them unsuitable for scenarios with time-varying channel states. Summary of the Invention

[0006] The purpose of this invention is to provide a method for allocating energy capture time and computing tasks in a wireless power edge network, which optimizes the allocation of energy capture time and computing tasks, has less time consumption, can achieve a higher computing rate, and is suitable for time-varying channel conditions.

[0007] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0008] A method for allocating energy capture time and computation tasks in a wireless power edge network, wherein the wireless power edge network includes one gateway and N wireless devices, the gateway integrating a radio frequency energy transmitter and an edge server, and the N wireless devices operating based on an orthogonal frequency division multiple access (OFDMA) communication mode. The method for allocating energy capture time and computation tasks in the wireless power edge network is implemented in each wireless device, including:

[0009] Step 1: Use a one-dimensional finite search method to output a time percentage set, where the time percentage in the time percentage set is the ratio of energy capture time to the frame length of the time frame;

[0010] Step 2: Based on each time percentage in the set of time percentages, calculate the energy percentage corresponding to the time percentage, where the energy percentage is the ratio of the energy unloaded by the calculation task to the energy captured.

[0011] Step 2.1: Select one time percentage from the set of time percentages;

[0012] Step 2.2: Initialize the energy percentage, upper limit, and lower limit corresponding to the current time percentage;

[0013] Step 2.3: Input the energy ratio into the derivative function to obtain the derivative value, and iteratively optimize the energy ratio based on the derivative value. After iterative optimization, output the energy ratio corresponding to the current time ratio.

[0014] Step 3: Based on each time percentage and corresponding energy percentage in the time percentage set, calculate the sum of the calculation rates of N wireless devices, and take the one with the largest sum of calculation rates as the optimal time percentage and energy percentage for this calculation by the wireless device.

[0015] Step 4: Based on the optimal time and energy allocation, complete the offloading and allocation of this calculation task.

[0016] Several alternative methods are provided below, but they are not intended as additional limitations on the overall solution above. They are merely further additions or optimizations. Provided there are no technical or logical contradictions, each alternative method can be combined individually with respect to the overall solution above, or multiple alternative methods can be combined with each other.

[0017] Preferably, the step of using a one-dimensional finite search method to output the time percentage set includes: taking values ​​in [0,1] according to a preset step size, taking each value as a time percentage, and taking all the values ​​as a time percentage set.

[0018] Preferably, the initialization of the energy percentage, upper limit of energy percentage, and lower limit of energy percentage corresponding to the current time percentage includes: initializing the upper limit of energy percentage x imax =1, lower limit of energy percentage x imin =0, energy percentage x i Let i be the energy ratio of the i-th wireless device, where i∈{1,2,3,…,N}.

[0019] Preferably, the formula for the derivative function is as follows:

[0020]

[0021] In the formula, f(x) i (x) is the derivative function. i Let a be the energy share of the i-th wireless device, a be the current time share, B be the bandwidth of the orthogonal channel in the wireless power edge network, P be the transmit power of the RF power transmitter, and h be the energy share of the i-th wireless device. i Let ξ be the channel gain of the i-th wireless device, N0 be the noise power, μ be the number of cycles required for the wireless device to compute one bit of task locally, ξ be the energy absorption efficiency, k be the computational energy efficiency coefficient of the wireless device, e be the base of the natural logarithm, and i ∈ {1, 2, 3, ..., N}.

[0022] Preferably, the step of iteratively optimizing the energy ratio based on the derivative value, and outputting the energy ratio corresponding to the current time ratio after iterative optimization, includes:

[0023] If the derivative value is greater than 0, then x imin =x imin x imax =x i , If the derivative value is less than 0, then x imin =x i x imax =x imax , x i x represents the energy percentage.imax x is the upper limit of energy percentage. imin This represents the lower limit of energy percentage, where i∈{1,2,3,…,N};

[0024] Continue to determine the relationship between the derivative value and the search precision. If the absolute value of the derivative value is less than the search precision, output the current energy percentage and end the iterative optimization; otherwise, return to step 2.3 to continue execution.

[0025] Preferably, the calculation of the sum of the calculation rates of the N wireless devices is as follows:

[0026]

[0027] In the formula, Q(x,a) represents the total calculation rate, a represents the time percentage, and x represents the set of energy percentages, x = [x1, x2, ... x i ,…x N ], x i Let μ be the energy percentage of the i-th wireless device, μ be the number of cycles required for the wireless device to locally compute one bit of a task, ξ be the energy absorption efficiency, P be the transmit power of the RF energy transmitter, and h be the energy percentage of the ith wireless device. i Let be the channel gain of the i-th wireless device, k be the calculated energy efficiency coefficient of the wireless device, B be the bandwidth of the orthogonal channel in the wireless power edge network, and N0 be the noise power.

[0028] Preferably, the step of completing the unloading and allocation of this computational task based on the optimal time and energy proportions includes:

[0029] Step 4.1: Divide the current time frame into capture time and unload time according to the time proportions mentioned above;

[0030] Step 4.2: Capture the radio frequency energy emitted by the radio frequency energy transmitter during the capture time, and divide the captured radio frequency energy into offload energy and local calculation energy according to the energy ratio.

[0031] Step 4.3: During the unloading time, the computing task is unloaded to the edge server using the unloading energy, and the computing task is locally computed based on the local storage energy and the local computing energy throughout the entire time frame.

[0032] Preferably, dividing the current time frame into capture time and unload time according to the time proportion includes:

[0033] If the frame length of a time frame is T and the time percentage is a, then the time interval [0, aT] is taken as the capture time and the time interval (aT, T] is taken as the unload time.

[0034] Preferably, the captured radio frequency energy is divided into offload energy and locally computed energy according to the energy ratio, including:

[0035] The radio frequency energy E captured by the i-th wireless device in the current time frame i for:

[0036] E i =ξPh i aT

[0037] In the formula, ξ is the energy absorption efficiency, P is the transmit power of the radio frequency energy transmitter, and h i Let be the channel gain of the i-th wireless device, 'a' be the time percentage, and 'T' be the frame length of the time frame.

[0038] Then the captured radio frequency energy x i E i Part of the captured radio frequency energy will be used as unloading energy (1-x). i E i Part of the energy is used for local computation, x i This represents the energy allocation ratio for the i-th wireless device.

[0039] The present invention provides a method for allocating energy capture time and computing tasks in a wireless power edge network. This method utilizes a one-dimensional exhaustive search to provide multiple candidate time percentages and calculates an optimal energy percentage based on each candidate time percentage. The total computing rate in the entire wireless power edge network is obtained from the time and energy percentages. Based on the maximum total computing rate, the energy capture time and computing task offloading scheme for the wireless device in the current time frame are determined. Compared to traditional optimization methods, this method can solve a complex non-convex problem in wireless power edge computing networks using Orthogonal Frequency Division Multiple Access (OFDMA) communication and partial offloading modes. It can quickly calculate the partial offloading energy and time allocation schemes while also achieving a higher network computing rate. Attached Figure Description

[0040] Figure 1 This is a schematic diagram of the OFDMA-based wireless power edge network model of the present invention;

[0041] Figure 2 This is a flowchart illustrating the method for allocating energy capture time and computational tasks in the wireless power edge network of the present invention. Detailed Implementation

[0042] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0043] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to limit the invention.

[0044] This invention relates to a method for allocating energy capture time and computing tasks in a wireless power edge network. This method is applicable to communication modes and partial offloading modes using orthogonal frequency division multiple access (OFDMA) to determine the wireless energy capture time and computing task offloading scheme of the edge computing network with wireless power transmission, so as to achieve a higher computing rate.

[0045] This embodiment uses Figure 1 The following example illustrates a wireless power edge network. A wireless power edge network includes one gateway and N (e.g., 10) wireless devices. The gateway integrates a radio frequency (RF) power transmitter and an edge server. The gateway broadcasts RF power to the wireless devices (e.g., as shown in the energy stream), and the wireless devices offload tasks to the edge server (e.g., as shown in the data stream), where edge computing is performed.

[0046] A gateway is a computer system or device that acts as a translator. Used between two systems with different communication protocols, data formats, languages, or even completely different architectures, a gateway is a translator. Unlike a bridge, which simply forwards information, a gateway repackages the received information to suit the needs of the destination system.

[0047] N wireless devices (IoT1, IoT2...IoTN) operate based on Orthogonal Frequency Division Multiple Access (OFDMA) communication, and perform task calculations based on time frames. Each wireless device captures radio frequency energy broadcast by a radio frequency energy transmitter, transmits the data with the gateway, and performs local calculations.

[0048] Specifically, such as Figure 2 As shown, this embodiment describes a method for allocating energy capture time and computational tasks in a wireless power edge network. This method is implemented in each wireless device, meaning each wireless device runs the method of this invention in each time frame to complete the allocation of energy capture time and computational tasks in the current time frame. Specifically, it includes the following steps:

[0049] Step 1: Use a one-dimensional finite search method to output the time percentage set.

[0050] The resulting set of time percentages contains multiple time percentages, each representing the proportion of energy capture time to the frame length of a time frame. This embodiment determines the length of time used for energy capture in each time frame using these time percentages. Since energy capture time is a time concept and its fitting and solution methods are relatively complex, this embodiment transforms the solution for energy capture time into a solution for time percentages, which are values ​​between 0 and 1, making the processing relatively simple.

[0051] Since the one-dimensional finite search method is simple and straightforward, this embodiment preferably uses the one-dimensional finite search method to obtain the time percentage set. However, similar methods for obtaining datasets are also within the scope of protection of this invention, such as simply replacing them with the DRL exploration method.

[0052] When implementing the one-dimensional finite search method, values ​​are taken in the range [0,1] according to a preset step size. Each value is used as a time percentage, and all the values ​​taken are used as a set of time percentages. The preset step size is set according to actual needs, for example, it can be 0.001, 0.002, 0.01, etc. In this embodiment, considering data precision, the preset step size is set to 0.001, that is, values ​​are taken in the range [0,1] with a step size of 0.001. That is, different values ​​of 'a' are 0.001, 0.002, ..., 0.999, which are used as the set of time percentages.

[0053] Step 2: Based on each time percentage in the time percentage set, calculate the energy percentage corresponding to the time percentage. The energy percentage is the proportion of the energy unloaded by the calculation task to the energy captured.

[0054] Step 2.1: Take one time percentage from the set of time percentages.

[0055] Step 2.2: Initialize the energy percentage, upper limit of energy percentage, and lower limit of energy percentage corresponding to the current time percentage.

[0056] Since the task unloading energy is a portion of the capture energy allocation, its energy percentage ranges from 0 to 1. Therefore, in this embodiment, the initial energy percentage upper limit x is set. imax =1, lower limit of energy percentage x imin =0, and initialize the energy percentage separately. x i Let i be the energy ratio of the i-th wireless device, where i∈{1,2,3,…,N}.

[0057] In other embodiments, the initial values ​​of the energy percentage, upper limit of energy percentage, and lower limit of energy percentage can be adjusted according to actual needs. For example, the upper limit of energy percentage can be initialized to 0.9, the lower limit of energy percentage to 0.2, the energy percentage to 0.3, etc.

[0058] Step 2.3: Input the energy percentage into the derivative function to obtain the derivative value, and iteratively optimize the energy percentage based on the derivative value. After iterative optimization, output the energy percentage corresponding to the current time percentage.

[0059] Step 2.3.1: First, calculate the derivative value based on the current time and energy percentages. The formula for the derivative function is given in this embodiment as follows:

[0060]

[0061] In the formula, f(x) i (x) is the derivative function. i Let a be the energy share of the i-th wireless device, a be the current time share, B be the bandwidth of the orthogonal channel in the wireless power edge network, P be the transmit power of the RF power transmitter, and h be the energy share of the i-th wireless device. i Let ξ be the channel gain of the i-th wireless device, N0 be the noise power, μ be the number of cycles required for the wireless device to compute one bit of task locally, ξ be the energy absorption efficiency, k be the computational energy efficiency coefficient of the wireless device, and e be the base of the natural logarithm. Without loss of generality, e is approximately 2.718281828459045, i∈{1,2,3,…,N}.

[0062] Step 2.3.2: When iteratively optimizing the energy ratio, the optimization is mainly based on the derivative value calculated from the derivative function. If the derivative value is greater than 0, then x... imin =x imin x imax =x i , If the derivative value is less than 0, then x imin =x i x imax =x imax ,

[0063] Step 2.3.3: Continue to determine the relationship between the derivative value and the search precision. If the absolute value of the derivative value is less than the search precision, output the current energy percentage and end the iterative optimization; otherwise, return to step 2.3 and continue execution. Specifically, based on the energy percentage obtained in step 2.3.2, return to step 2.3.1 to execute.

[0064] In this embodiment, the search precision is set according to actual needs. If a higher final allocation result is required, a smaller search precision value can be set; conversely, if a faster overall method is needed, a larger search precision value can be set to reduce the number of iterations. For example, values ​​like 0.5 or 0.8 can be used.

[0065] Step 3: Based on each time percentage and its corresponding energy percentage in the time percentage set, calculate the sum of the calculation rates of N wireless devices, and take the one with the largest sum of calculation rates as the optimal time percentage and energy percentage for this calculation by the wireless device.

[0066] Since there are multiple time percentages in the time percentage set, this embodiment introduces the sum of calculation rates as a reference quantity in order to obtain the optimal time percentage. This embodiment calculates the sum of calculation rates of N wireless devices as follows:

[0067]

[0068] In the formula, Q(x,a) represents the total calculation rate, a represents the time percentage, and x represents the set of energy percentages, x = [x1, x2, ... x i ,…x N ], x i Let μ be the energy percentage of the i-th wireless device, μ be the number of cycles required for the wireless device to locally compute one bit of a task, ξ be the energy absorption efficiency, P be the transmit power of the RF energy transmitter, and h be the energy percentage of the ith wireless device. i Let be the channel gain of the i-th wireless device, k be the calculated energy efficiency coefficient of the wireless device, B be the bandwidth of the orthogonal channel in the wireless power edge network, and N0 be the noise power.

[0069] It should be noted that since calculating the total rate requires the energy percentage and channel gain of N wireless devices, this wireless device needs to communicate with the gateway to obtain the latest energy percentage and channel gain of other wireless devices stored on the gateway when performing the total rate calculation. Because there is a certain time interval between data interactions between the wireless device and the gateway, the latest energy percentage and channel gain of other wireless devices stored on the gateway obtained by this wireless device may not be the latest energy percentage and channel gain of those other wireless devices locally. Furthermore, the energy percentage used in the calculation is the energy percentage corresponding to multiple time percentages obtained in step 2, and the channel gain of this wireless device is the latest channel gain obtained within the current time frame.

[0070] Furthermore, the sum of rates μ, ξ, P, k, B, and N0 are obtained directly or by fitting based on the properties of the actual device or network.

[0071] Step 4: Based on the optimal time and energy allocation, complete the offloading and allocation of this calculation task.

[0072] Step 4.1: Divide the current time frame into capture time and unload time according to the time proportion.

[0073] Let T be the length of a time frame. Without loss of generality, T = 1 second. At the beginning of each time frame, the gateway provides wireless power to N wireless devices. At the same time, each wireless device executes this method to plan the wireless power capture duration and the computational task offloading scheme.

[0074] In this embodiment, the frame length of the time frame is T, and the time percentage is a; then the time period [0, aT] is used as the capture time, and the time period (aT, T] is used as the offload time. That is, the aT part of each time frame is used for the access point to broadcast radio frequency energy, and the (1-a)T part of the time is used to offload part of the computing work to the edge server.

[0075] Step 4.2: Capture the radio frequency energy emitted by the radio frequency energy transmitter during the capture time, and divide the captured radio frequency energy into offload energy and local calculation energy according to the energy ratio.

[0076] If a wireless device captures radio frequency energy at the beginning of each time frame, then the radio frequency energy E captured by the i-th wireless device at the capture time can be obtained. i for:

[0077] E i =ξPh i aT

[0078] In the formula, ξ is the energy absorption efficiency, and 0 < ξ < 1, P is the transmit power of the radio frequency energy transmitter, and h i Let be the channel gain of the i-th wireless device, 'a' be the frame length multiple, 'T' be the frame length of the time frame, i∈{1,2,…,N}, and the channel gain set h=[h1,h2,…,h N ].

[0079] For the i-th wireless device, the energy acquired will be x i E i A portion of the energy is used as offloading energy to offload computational workloads to edge servers, and the energy acquired is (1-x) i E i Part of it is used as local computing energy for local computation, and x i ∈[0,1].

[0080] Step 4.3: During the unloading time, the computing tasks are unloaded to the edge server using the unloading energy, and the computing tasks are computed locally based on the local storage energy and local computing energy throughout the time frame.

[0081] In this embodiment, within the remaining duration (1-a)T of each time frame, N wireless devices operate based on an orthogonal frequency division multiple access (OFDM) communication mode. Each wireless device has a wire and a rechargeable battery for storing radio frequency (RF) energy. Therefore, the wireless devices store the captured RF energy in the rechargeable battery. Energy in the rechargeable battery other than that captured in the current time frame can be considered locally stored energy, while the energy captured in the current time frame is divided into offload energy and local computation energy.

[0082] During the acquisition time, since the wireless device is still acquiring energy, local stored energy can be used for local task computation. After separating offloading energy from local computation energy, offloading energy is used for task offloading, and local computation energy is used for local computation. If the offloading energy and local computation energy are not fully used after task offloading and local computation are completed in the current time frame, they can be allocated to local stored energy for use in the next time frame.

[0083] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0084] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the appended claims.

Claims

1. A method for allocating energy capture time and computing tasks in a wireless power edge network, characterized in that, The wireless power supply edge network includes one gateway and N wireless devices. The gateway integrates a radio frequency energy transmitter and an edge server. The N wireless devices operate based on an orthogonal frequency division multiple access (OFDMA) communication mode. The method for allocating energy capture time and computational tasks in the wireless power supply edge network is implemented in each wireless device, including: Step 1: Use a one-dimensional finite search method to output a time percentage set, where the time percentage in the time percentage set is the ratio of energy capture time to the frame length of the time frame; Step 2: Based on each time percentage in the set of time percentages, calculate the energy percentage corresponding to the time percentage, where the energy percentage is the ratio of the energy unloaded by the calculation task to the energy captured. Step 2.1: Select one time percentage from the set of time percentages; Step 2.2: Initialize the energy percentage, upper limit, and lower limit corresponding to the current time percentage; Step 2.3: Input the energy percentage into the derivative function to obtain the derivative value, and iteratively optimize the energy percentage based on the derivative value. After iterative optimization, output the energy percentage corresponding to the current time percentage. The formula for the derivative function is as follows: In the formula, The derivative function, To determine the energy ratio with the i-th wireless device, This represents the current time percentage. For the bandwidth of the orthogonal channel in a wireless power edge network, This refers to the transmit power of the radio frequency energy transmitter. For the first Channel gain of each wireless device For noise power, The number of cycles required for a wireless device to locally calculate one bit of a task. For energy absorption efficiency, The coefficient of performance (COP) for wireless devices. Let i be the base of the natural logarithm, i∈{1,2,3,…,N}; Step 3: Based on each time percentage and corresponding energy percentage in the time percentage set, calculate the sum of the calculation rates of N wireless devices, and take the one with the largest sum of calculation rates as the optimal time percentage and energy percentage for this calculation by the wireless device. Step 4: Based on the optimal time and energy allocation, complete the offloading and allocation of this calculation task.

2. The method for allocating energy capture time and computing tasks in a wireless power edge network as described in claim 1, characterized in that, The method of using a one-dimensional finite search to output the time percentage set includes: taking values ​​in [0,1] according to a preset step size, taking each value as a time percentage, and taking all the values ​​as a time percentage set.

3. The method for allocating energy capture time and computing tasks in a wireless power edge network as described in claim 1, characterized in that, The initialization of the energy percentage, upper limit of energy percentage, and lower limit of energy percentage corresponding to the current time percentage includes: initializing the upper limit of energy percentage x imax =1, lower limit of energy percentage x imin =0, energy percentage , Let i be the energy ratio of the i-th wireless device, where i∈{1,2,3,…,N}.

4. The method for allocating energy capture time and computing tasks in a wireless power edge network as described in claim 1, characterized in that, The step of iteratively optimizing the energy ratio based on the derivative value, and outputting the energy ratio corresponding to the current time ratio after iterative optimization, includes: If the derivative value is greater than 0, then , , If the derivative value is less than 0, then , , , As a percentage of energy, This represents the upper limit of the energy percentage. This represents the lower limit of energy percentage, where i∈{1,2,3,…,N}; Continue to determine the relationship between the derivative value and the search precision. If the absolute value of the derivative value is less than the search precision, output the current energy percentage and end the iterative optimization; otherwise, return to step 2.3 to continue execution.

5. The method for allocating energy capture time and computing tasks in a wireless power edge network as described in claim 1, characterized in that, The calculation of the sum of the computing rates of N wireless devices is as follows: In the formula, To calculate the sum of rates, For time percentages, Let x be the set of energy percentages, x = [x1, x2, ... x]. i ,…x N ], For the first Energy percentage of each wireless device The number of cycles required for a wireless device to locally calculate one bit of a task. For energy absorption efficiency, This refers to the transmit power of the radio frequency energy transmitter. For the first Channel gain of each wireless device The coefficient of performance (COP) for wireless devices. For the bandwidth of the orthogonal channel in a wireless power edge network, This represents noise power.

6. The method for allocating energy capture time and computing tasks in a wireless power edge network as described in claim 1, characterized in that, The process of unloading and allocating the computational task based on the optimal time and energy allocation includes: Step 4.1: Divide the current time frame into capture time and unload time according to the time proportions mentioned above; Step 4.2: Capture the radio frequency energy emitted by the radio frequency energy transmitter during the capture time, and divide the captured radio frequency energy into offload energy and local calculation energy according to the energy ratio. Step 4.3: During the unloading time, the computing task is unloaded to the edge server using the unloading energy, and the computing task is locally computed based on the local storage energy and the local computing energy throughout the entire time frame.

7. The method for allocating energy capture time and computing tasks in a wireless power edge network as described in claim 6, characterized in that, The step of dividing the current time frame into capture time and unload time according to the time proportion includes: Let the frame length of the time frame be T, and the time percentage be... Then the time period [0, ] as the capture time, the time period ( , [This refers to the uninstallation time.] 8. The method for allocating energy capture time and computing tasks in a wireless power edge network as described in claim 6, characterized in that, The captured radio frequency energy is divided into offload energy and local computation energy according to the energy ratio, including: No. Radio frequency energy captured by a wireless device in the current time frame for: In the formula, For energy absorption efficiency, This refers to the transmit power of the radio frequency energy transmitter. For the first Channel gain of each wireless device T represents the time percentage, and T is the frame length of the time frame. Then the captured radio frequency energy Part of the captured radio frequency energy is used as offload energy. Part of it is used for local computing power. For the first The energy allocation ratio of each wireless device.

Images