Method, apparatus and storage medium for optimizing mobile edge network

Abstract

The application discloses a kind of optimization method, device and storage medium of mobile edge network, comprising: construct three-layer mobile edge network model based on the Internet of Things equipment, unmanned aerial vehicle and satellite between;The calculation model related to three-layer mobile edge network model is constructed, wherein the calculation model determines the time delay and energy consumption of three-layer mobile edge network model based on the following parameters: the first variable between the Internet of Things equipment and unmanned aerial vehicle, bandwidth allocation ratio between the Internet of Things equipment and unmanned aerial vehicle, the second variable between the Internet of Things equipment and unmanned aerial vehicle and the first position coordinate of unmanned aerial vehicle;According to the calculation model, determine the joint optimization problem, wherein the joint optimization problem is used to indicate the calculation time delay of minimization;And by block coordinate descent method, the joint optimization problem is divided into different sub-problems, and the first variable, bandwidth allocation ratio, second variable and first position coordinate are respectively optimized based on sub-problem.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of communication, and in particular to a mobile edge network optimization method and device and storage medium. BACKGROUND

[0002] With the continuous development of mobile communication, communication services have higher requirements for communication systems in terms of latency, reliability, and rate. The traditional way is to transfer computing tasks to the cloud for processing, but this way has a long delay. Based on this, the existing mobile edge computing technology is proposed. Mobile edge computing technology is a new computing technology that can sink computing power from the cloud to the edge of the network, thereby further improving data transmission capacity and data transmission efficiency.

[0003] Unmanned aerial vehicle assisted mobile edge computing network refers to giving unmanned aerial vehicles computing power, and enabling resource-constrained Internet of Things devices to offload tasks with large data volume to unmanned aerial vehicles with high computing power, thereby achieving the technical effect of meeting the quality of service requirements of ground users and reducing latency.

[0004] Further, based on the unmanned aerial vehicle assisted mobile edge computing network, a three-layer mobile edge network model based on Internet of Things devices-unmanned aerial vehicles-satellites can be constructed, and the computing tasks of the Internet of Things devices are cached to the unmanned aerial vehicles and offloaded to the satellites for processing, thereby fully utilizing the advantages of the mobile edge computing network.

[0005] In addition, in the three-layer mobile edge network model of Internet of Things devices-unmanned aerial vehicles-satellites, minimizing task latency is one of the most important problems to be solved. However, because there are many variables involved in the three-layer network model composed of Internet of Things devices-unmanned aerial vehicles-satellites, the task latency problem to be solved is relatively complex.

[0006] In view of the technical problem in the prior art that because there are many variables involved in the three-layer mobile edge network model composed of Internet of Things devices-unmanned aerial vehicles-satellites, the task latency problem to be solved is relatively complex, no effective solution has been proposed so far. SUMMARY

[0007] Embodiments of the present disclosure provide a mobile edge network optimization method, device and storage medium to at least solve the technical problem in the prior art that because there are many variables involved in the three-layer mobile edge network model composed of Internet of Things devices-unmanned aerial vehicles-satellites, the task latency problem to be solved is relatively complex.

[0008] According to an aspect of embodiments of the present disclosure, there is provided a method for optimizing a mobile edge network, comprising: constructing a three-layer mobile edge network model based on an Internet of Things device, a drone, and a satellite; constructing a calculation model related to the three-layer mobile edge network model, wherein the calculation model determines a latency and an energy consumption of the three-layer mobile edge network model based on a first variable between the Internet of Things device and the drone, a bandwidth allocation ratio between the Internet of Things device and the drone, a second variable between the Internet of Things device and the drone, and a first position coordinate of the drone; determining a joint optimization problem according to the calculation model, wherein the joint optimization problem is used to indicate a calculation latency to be minimized; and dividing the joint optimization problem into different sub-problems by a block coordinate descent method, and optimizing the first variable, the bandwidth allocation ratio, the second variable, and the first position coordinate based on the sub-problems, respectively.

[0009] According to another aspect of embodiments of the present disclosure, there is also provided a storage medium comprising a stored program, wherein the program, when executed by a processor, performs any of the above methods.

[0010] According to another aspect of embodiments of the present disclosure, there is also provided an apparatus for optimizing a mobile edge network, comprising: a mobile edge network model construction module configured to construct a three-layer mobile edge network model based on an Internet of Things device, a drone, and a satellite; a calculation model construction module configured to construct a calculation model related to the three-layer mobile edge network model, wherein the calculation model determines a latency and an energy consumption of the three-layer mobile edge network model based on a first variable between the Internet of Things device and the drone, a bandwidth allocation ratio between the Internet of Things device and the drone, a second variable between the Internet of Things device and the drone, and a first position coordinate of the drone; a joint optimization problem determination module configured to determine a joint optimization problem according to the calculation model, wherein the joint optimization problem is used to indicate a calculation latency to be minimized; and an optimization module configured to divide the joint optimization problem into different sub-problems by a block coordinate descent method, and optimize the first variable, the bandwidth allocation ratio, the second variable, and the first position coordinate based on the sub-problems, respectively.

[0011] According to another aspect of the embodiments of the present disclosure, there is also provided an optimization device of a mobile edge network, comprising: a processor; and a memory connected with the processor, configured to provide the processor with instructions to process the following processing steps: constructing a three-layer mobile edge network model based on an Internet of Things device, a drone and a satellite; constructing a calculation model related to the three-layer mobile edge network model, wherein the calculation model determines the latency and energy consumption of the three-layer mobile edge network model based on the following parameters: a first variable between the Internet of Things device and the drone, a bandwidth allocation ratio between the Internet of Things device and the drone, a second variable between the Internet of Things device and the drone, and a first position coordinate of the drone; determining a joint optimization problem according to the calculation model, wherein the joint optimization problem is used to indicate the calculation latency to be minimized; and dividing the joint optimization problem into different sub-problems by block coordinate descent method, and optimizing the first variable, the bandwidth allocation ratio, the second variable and the first position coordinate based on the sub-problems respectively.

[0012] The present application provides an optimization method of a mobile edge network. First, a processor constructs a three-layer mobile edge network model based on an Internet of Things device, a drone and a satellite. Then, the processor constructs a calculation model related to the three-layer mobile edge network model. Wherein, the calculation model determines the latency and energy consumption of the three-layer mobile edge network model based on the following parameters: a first variable between the Internet of Things device and the drone, a bandwidth allocation ratio between the Internet of Things device and the drone, a second variable between the Internet of Things device and the drone, and a first position coordinate of the drone. Further, the processor determines a joint optimization problem according to the calculation model. Finally, the processor divides the joint optimization problem into different sub-problems by block coordinate descent method, and optimizes the first variable, the bandwidth allocation ratio, the second variable and the first position coordinate based on the sub-problems respectively.

[0013] As can be seen from the above, after the processor in the present application determines the joint optimization problem (i.e., the problem of minimizing the calculation latency) according to the calculation model, it divides the joint optimization problem into different sub-problems (i.e., solving the associated variable (i.e., the first variable) and the cache decision variable (i.e., the second variable) given the bandwidth allocation ratio and the first position coordinate; solving the bandwidth allocation ratio given the associated variable, the cache decision variable and the first position coordinate; solving the first position coordinate given the associated variable, the bandwidth allocation ratio and the cache decision variable) by block coordinate descent method. Then, the processor converts each non-convex sub-problem into a convex sub-problem and solves each sub-problem respectively, so as to determine the minimized calculation latency and further ensure the communication quality.

[0014] Further, the technical problem that the task delay problem of solving is relatively complex due to more variables involved in the three-layer mobile edge network model constituted by the Internet of Things device, the unmanned aerial vehicle and the satellite in the prior art is solved. BRIEF DESCRIPTION OF DRAWINGS

[0015] The accompanying drawings, which are included to provide a further understanding of the present disclosure and are incorporated in and constitute a part of this application, illustrate embodiments of the present disclosure and serve to explain the principles of the present disclosure, and do not limit the present disclosure in any manner. In the drawings:

[0016] Figure 1 is a hardware structure block diagram of a computing device for implementing the method according to Embodiment 1 of the present disclosure;

[0017] Figure 2 is a schematic diagram of a three-layer mobile edge network model corresponding to the Internet of Things terminal, the unmanned aerial vehicle and the satellite according to Embodiment 1 of the present disclosure;

[0018] Figure 3 is a flowchart of the optimization method of the mobile edge network according to the first aspect of Embodiment 1 of the present disclosure;

[0019] Figure 4 is a schematic diagram of the optimization device of the mobile edge network according to the first aspect of Embodiment 2 of the present disclosure;

[0020] Figure 5 is a schematic diagram of the optimization device of the mobile edge network according to the first aspect of Embodiment 3 of the present disclosure. DETAILED DESCRIPTION

[0021] In order to enable persons skilled in the art to better understand the technical solutions of the present disclosure, the technical solutions in the embodiments of the present disclosure will be described clearly and completely below with reference to the accompanying drawings in the embodiments of the present disclosure. Obviously, the described embodiments are only a part of the embodiments of the present disclosure, rather than all the embodiments. Based on the embodiments in the present disclosure, all other embodiments obtained by persons skilled in the art without creative labor should fall within the protection scope of the present disclosure.

[0022] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this disclosure are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this disclosure described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0023] Example 1

[0024] According to this embodiment, an implementation of a method for optimizing a mobile edge network is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0025] The method embodiments provided in this example can be executed on mobile terminals, computer terminals, servers, or similar computing devices. Figure 1 A hardware block diagram of an optimized computing device for implementing mobile edge networks is shown. Figure 1 As shown, a computing device may include one or more processors (processors may include, but are not limited to, microprocessors such as MCUs or programmable logic devices such as FPGAs), memory for storing data, transmission devices for communication functions, and input / output interfaces. The memory, transmission devices, and input / output interfaces are connected to the processor via a bus. In addition, it may also include a display, keyboard, and cursor control device connected to the input / output interfaces. Those skilled in the art will understand that... Figure 1 The structure shown is for illustrative purposes only and does not limit the structure of the aforementioned electronic device. For example, a computing device may also include... Figure 1 The more or fewer components shown, or having the same Figure 1 The different configurations shown.

[0026] It should be noted that the one or more processors and / or other data processing circuitry described above can be referred to herein generically as "data processing circuitry." The data processing circuitry can be embodied as, include or otherwise be associated with software, hardware, firmware, or any combination thereof. Moreover, the data processing circuitry can be a single independent processing module or any combination of plural processing modules, all of which are incorporated in whole or in part within the computing device. As referred to in the embodiments of the present disclosure, the data processing circuitry functions as a processor to control, for example, selection of the variable resistance terminal path connected with the interface.

[0027] The memory can be used to store software programs and modules of application software, such as program instructions / data storage means corresponding to the method for optimizing a mobile edge network in the embodiments of the present disclosure. The processor performs various functional applications and data processing by running the software programs and modules stored in the memory, i.e., implements the method for optimizing a mobile edge network of the application program as described above. The memory can include a high-speed random access memory, and can also include a non-volatile memory, such as one or more magnetic storage devices, flash memories, or other non-volatile solid-state memories. In some examples, the memory can further include a memory disposed remotely with respect to the processor, which can be connected to the computing device through a network. Examples of the network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0028] The transmission device is used to receive or send data via a network. Specific examples of the network can include a wireless network provided by a communication provider of the computing device. In one example, the transmission device includes a network adapter (NIC), which can be connected to other network devices through a base station so as to be able to communicate with the Internet. In one example, the transmission device can be a radio frequency (RF) module, which is used to communicate with the Internet in a wireless manner.

[0029] The display can be, for example, a touch screen type liquid crystal display (LCD), which can enable a user to interact with the user interface of the computing device.

[0030] It should be noted that in some optional embodiments, the above Figure 1 The computing device shown can include hardware elements (including circuitry), software elements (including computer code stored on a computer readable medium), or a combination of both hardware and software elements. It should be noted that in some embodiments, the functions of the above Figure 1 is merely an example of a particular implementation and is intended to illustrate the types of components that can be present in the computing device described above.

[0031] Figure 2This is a schematic diagram of the three-layer mobile edge network model corresponding to IoT terminal—drone—satellite as described in this embodiment. (Refer to...) Figure 2 As shown, the model includes: multiple IoT devices Li and multiple drones U. j And satellite S. Where i = 1 to k, j = 1 to n.

[0032] In addition, refer to Figure 2 As shown, UAV j Capable of communicating with multiple IoT devices within the coverage area. i Establish a communication connection to receive signals from multiple IoT devices. i The second computational task to be unloaded.

[0033] Internet of Things (IoT) devices L i With the only drone U j Establish a connection to offload the second computational task to the UAV. j superior.

[0034] UAV j Capable of communicating with multiple IoT devices within the coverage area. i Establish a communication connection to enable multiple IoT devices within the communication coverage area. i When the second computational task being unloaded is different, the UAV U j The total task that needs to be calculated is for each IoT device L i The sum of the unloaded second computing tasks. Multiple IoT devices L within the communication coverage area. i Under the condition that the second computational task being unloaded is the same, the UAV U j The total computational task required is equal to the second computational task. That is, multiple IoT devices L within the communication coverage area... i Under the condition that the second computational task being unloaded is the same, the UAV U j No need for repeated calculations.

[0035] Satellite S and multiple drones U j Communication connection, so that satellite S can communicate with UAV U j Received from IoT terminal L i The third computational task to be unloaded.

[0036] It should be noted that the hardware structure described above is applicable to IoT devices, drones, and satellites in the system.

[0037] Under the aforementioned operating environment, according to the first aspect of this embodiment, an optimization method for a mobile edge network is provided, the method comprising: Figure 1 The processor implementation shown. Figure 3A flowchart of the method is shown, referring to Figure 3 The method comprises:

[0038] S302: A three-layer mobile edge network model between the Internet of Things devices, the unmanned aerial vehicles, and the satellites is constructed;

[0039] S304: A calculation model related to the three-layer mobile edge network model is constructed, wherein the calculation model determines the latency and the energy consumption of the three-layer mobile edge network model based on the following parameters: a first variable between the Internet of Things devices and the unmanned aerial vehicles, a bandwidth allocation ratio between the Internet of Things devices and the unmanned aerial vehicles, a second variable between the Internet of Things devices and the unmanned aerial vehicles, and a first position coordinate of the unmanned aerial vehicles;

[0040] S306: According to the calculation model, a joint optimization problem is determined, wherein the joint optimization problem is used to indicate the calculation latency to be minimized; and

[0041] S308: The joint optimization problem is divided into different sub-problems by the block coordinate descent method, and the first variable, the bandwidth allocation ratio, the second variable, and the first position coordinate are optimized based on the sub-problems.

[0042] Specifically, referring to Figure 2 As shown in the figure, first, the processor constructs a three-layer mobile edge network model between the Internet of Things devices, the unmanned aerial vehicles, and the satellites (S302). Since each unmanned aerial vehicle is configured with a mobile edge network server, each unmanned aerial vehicle can provide computing services for the Internet of Things devices within its communication coverage. Thus, the Internet of Things devices can choose to process the computing tasks locally, or choose to offload the computing tasks to the mobile edge network server of the unmanned aerial vehicle or the satellite end cloud server for processing.

[0043] Then, a plurality of Internet of Things devices L i on the ground are selected as the first variable, and are expressed as a set L i ={L1, L2, L3,..., L k}. A plurality of unmanned aerial vehicles U j =(U1, U2, U3,..., U n} are selected as the second variable.

[0044] Further, the processor determines the computing task I i of the ith Internet of Things device in the current time slot, and determines the number of CPU cycles required by the ith Internet of Things device to calculate 1 bit, the computing task size D i of the ith Internet of Things device, and the maximum deadline T i of the ith Internet of Things device to complete the computing task.

[0045] The computing task size D ican be divided into three parts, respectively, and wherein, denotes the task computed on the Internet of Things terminal (i.e., the first computing task), denotes the task computed on the unmanned aerial vehicle (i.e., the second computing task), and denotes the task computed on the satellite (i.e., the third computing task). Assuming that the division of the computing task does not result in additional input data, the computing task of the ith Internet of Things device is equal to the sum of the task computed on the Internet of Things terminal, the task computed on the unmanned aerial vehicle, and the task computed on the satellite. That is:

[0046]

[0047] It is worth noting that when the processor sets , the entire computing task is offloaded to the unmanned aerial vehicle.

[0048] Further, the processor constructs a computing model related to the three-layer mobile edge network model. Specifically, first, the processor determines the association variable (i.e., the first variable) a = {a i,j} between the ith Internet of Things device and the jth unmanned aerial vehicle. Wherein, the association variable a i,j is used to indicate whether the Internet of Things device is connected with the unmanned aerial vehicle. In the case that the Internet of Things device is connected with the unmanned aerial vehicle, a i,j = 1; in the case that the Internet of Things device is not connected with the unmanned aerial vehicle, a i,j = 0. Further, since each Internet of Things device can be connected with at most one unmanned aerial vehicle, there is the following formula:

[0049]

[0050] Then, the processor determines the cache decision variable (i.e., the second variable) c = {c i,j} between the ith Internet of Things device and the jth unmanned aerial vehicle. Wherein, the cache decision variable c i,j is used to indicate whether the Internet of Things device decides to cache the computing task to the unmanned aerial vehicle. In the case that the Internet of Things device decides to cache the computing task to the unmanned aerial vehicle, c i,j = 1; otherwise, c i,j = 0.

[0051] Further, the processor determines the bandwidth allocated to the Internet of Things device, and the proportion of the bandwidth of the ith Internet of Things device offloading the computing task to the jth unmanned aerial vehicle b = {b i,j}.

[0052] In addition, the processor determines the computing task allocated to the Internet of Things device Assigning computing tasks to drones and assigning computing tasks to satellites wherein the total computing task

[0053] Finally, the processor determines the first position coordinates of the drones wherein H j represents the flight height of the drones, represents the horizontal position coordinates of the drones.

[0054] The processor determines the association variable a = {a i,j} between the ith Internet of Things device and the jth drone, the cache decision variable c = {c i,j} between the ith Internet of Things device and the jth drone, the bandwidth assigned to the Internet of Things device, the proportion of the bandwidth b = {b i,j} of the ith Internet of Things device to offload computing tasks to the jth drone, and the first position coordinates of the drones and determines the latency and energy consumption of the three-layer mobile edge network model according to the above parameters (S304).

[0055] Then, the processor constructs a joint optimization problem according to the computing model (S306). That is, a formula for minimizing the computing latency is constructed. The specific formula is as follows:

[0056]

[0057] and wherein the constraint conditions corresponding to the above joint optimization problem include:

[0058] C1:

[0059] C2: E i ≤ E max ,

[0060] C3:

[0061] C4:

[0062] C5:

[0063] C6:

[0064] C7: a i,j ∈ {0, 1}, c i,j ∈ {0, 1}

[0065] C8:

[0066] C9: m≠j

[0067] C10: H min ≤ H j ≤ H max

[0068] C11: b i,j ,

[0069] wherein E max represents the maximum available energy of the ith Internet of Things device.

[0070] wherein constraint condition C1 represents that the time delay of processing the computing task cannot exceed the maximum time delay; constraint condition C2 represents that the total energy consumption of each Internet of Things device cannot exceed the maximum available energy; constraint condition C3 represents that the sum of the computing task allocated to the Internet of Things device, the computing task allocated to the unmanned aerial vehicle and the computing task allocated to the satellite is equal to the total computing task; constraint condition C4 represents that the CPU cycle frequency of the unmanned aerial vehicle cannot exceed the maximum CPU cycle frequency allowed; constraint condition C5 represents that each Internet of Things device can be associated with at most one unmanned aerial vehicle; constraint condition C6 represents that the sum of the bandwidth proportions is equal to 1; constraint condition C7 represents that there is only association or no association between the Internet of Things device and the unmanned aerial vehicle, and there is caching of the computing task to the unmanned aerial vehicle or no caching of the computing task to the unmanned aerial vehicle; constraint condition C8 represents that the computing task cached by the unmanned aerial vehicle cannot exceed the maximum cache capacity; constraint condition C9 represents that the distance between the unmanned aerial vehicles should be greater than the minimum distance allowed so as to avoid collision between the unmanned aerial vehicles; constraint condition C10 represents that the flight height of the unmanned aerial vehicle cannot be less than the minimum height allowed and cannot be greater than the maximum height allowed; constraint condition C11 represents that the bandwidth allocation proportion, the computing resource allocated to the Internet of Things device, the computing resource allocated to the unmanned aerial vehicle, the computing resource allocated to the satellite, the computing task allocated to the Internet of Things device, the computing task allocated to the unmanned aerial vehicle and the computing task allocated to the satellite all need to be greater than 0.

[0071] Finally, the processor divides the joint optimization problem into different sub-problems by block coordinate descent method, and optimizes the association variable, bandwidth allocation proportion, cache decision variable and first position coordinate based on the sub-problems (S308). Specifically, 1. given the bandwidth allocation proportion b = {b i,j} and the first position coordinate of the unmanned aerial vehicle , the association variable a = {a i,j} and the cache decision variable c = {c i,j} are solved; 2. given the association variable a = {a i,j}, the cache decision variable c = {c i,j} and the position of the unmanned aerial vehicle, the bandwidth allocation proportion b = {bi,j 3. Given the bandwidth allocation ratio b = {b i,j}, the associated variable a = {a i,j} and the cache decision variable c = {c i,j}, solve the UAV position optimization The details will be described later, and thus will not be described here.

[0072] As described in the background, based on the UAV-assisted mobile edge computing network, a three-layer mobile edge network model based on Internet of Things devices-UAV-satellite can be constructed, and the computing tasks of Internet of Things devices are cached to UAVs and unloaded to satellites for processing, thereby fully exerting the advantages of mobile edge computing network.

[0073] In addition, in the three-layer mobile edge network model of Internet of Things devices-UAV-satellite, minimizing task delay is one of the most important problems to be solved. However, due to the large number of variables involved in the three-layer network model composed of Internet of Things devices-UAV-satellite, the task delay problem to be solved is relatively complex.

[0074] Therefore, after determining the joint optimization problem (i.e., the problem of minimizing the computing delay) according to the computing model, the processor in the present application divides the joint optimization problem into different sub-problems (i.e., given the bandwidth allocation ratio and the first position coordinate, solve the associated variable and the cache decision variable; given the associated variable, the cache decision variable and the first position coordinate, solve the bandwidth allocation ratio; given the associated variable, the bandwidth allocation ratio and the cache decision variable, solve the first position coordinate) by block coordinate descent method. Then, the processor converts each non-convex sub-problem into a convex sub-problem and solves each sub-problem respectively, so as to determine the minimized computing delay and further ensure the communication quality.

[0075] Further, the technical problem that the task delay problem to be solved is relatively complex due to the large number of variables involved in the three-layer network model composed of Internet of Things devices-UAV-satellite in the prior art is solved.

[0076] Optionally, the operation of constructing the computing model related to the three-layer mobile edge network model comprises: constructing a channel model between the IoT device and the UAV; and constructing a computing model among the IoT device, the UAV and the satellite based on the channel model. Further optionally, the operation of constructing the channel model between the IoT device and the UAV comprises: determining a first position coordinate corresponding to the UAV and a second position coordinate corresponding to the IoT device according to a Euclidean coordinate system model, wherein the first position coordinate comprises a flight height of the UAV and a horizontal position coordinate of the UAV, and the second position coordinate is used to indicate a horizontal position coordinate of the IoT device; determining a distance between the IoT device and the UAV; determining a channel coefficient between the IoT device and the UAV; and determining a data transmission rate of the IoT device to offload the computing task to the UAV according to the determined distance and the channel coefficient.

[0077] Specifically, first, the processor adopts a three-dimensional Euclidean coordinate system model, and assumes that the flight height of the jth UAV is H j , the horizontal position coordinate of the jth UAV is the horizontal position coordinate of the IoT device is Therefore, the distance between the IoT device and the UAV can be determined by the following formula:

[0078]

[0079] The channel coefficient from the ith IoT device to the jth UAV is modeled as

[0080]

[0081] wherein, Δ i,j represents a path loss factor between the ith IoT device and the jth UAV, indicates a small-scale fading between the ith IoT device and the jth UAV.

[0082] Further, the path loss factor can be represented by the following formula:

[0083]

[0084] wherein, h0 represents a reference channel gain when the distance d0 = 1m, α is a path loss exponent, d i,j represents the distance between the IoT device and the UAV.

[0085] In addition, the small-scale fading can be represented by the following formula:

[0086]

[0087] wherein, M represents a Rician fading factor, represents a component satisfying line-of-sight. denotes a non-line of sight component. And wherein, (0, 1).

[0088] If the ith Internet of Things device decides to offload its computing task to the jth unmanned aerial vehicle, the data transmission rate of the computing task can be represented by the following formula:

[0089]

[0090] where B represents bandwidth, P i denotes the Internet of Things device fixed uplink transmission power, N0 represents noise power.

[0091] Optionally, the operation of constructing a computing model related to the three-layer mobile edge network model comprises: dividing the total computing task of the physical network device into a first computing task, a second computing task and a third computing task, wherein the first computing task is calculated on the Internet of Things device, the second computing task is calculated on the unmanned aerial vehicle, and the third computing task is calculated on the satellite; determining a first computing delay of the Internet of Things device when calculating the first computing task, a second computing delay of the unmanned aerial vehicle when calculating the second computing task, and a third computing delay of the satellite when calculating the third computing task; determining a first energy consumption of the Internet of Things device when calculating the first computing task, a second energy consumption of the unmanned aerial vehicle when calculating the second computing task, and a third energy consumption of the satellite when calculating the third computing task; determining a total delay when calculating the computing task according to the first computing delay, the second computing delay and the third computing delay; and determining a total energy consumption when calculating the computing task according to the first energy consumption, the second energy consumption and the third energy consumption.

[0092] Specifically, when the computing task is processed on the Internet of Things device, the computing delay is represented by , and the energy consumption is represented by ; when the computing task is processed on the unmanned aerial vehicle, the computing delay is represented by , and the energy consumption is represented by ; when the computing task is processed on the satellite, the computing delay is represented by , and the energy consumption is represented by .

[0093] 1. When the first computing task is processed on the Internet of Things device, the computing delay can be calculated by the following formula:

[0094]

[0095] where denotes the first computing task of the Internet of Things device processor, F i denotes the total number of CPU cycles required to calculate 1 bit, denotes the computing resources allocated to the Internet of Things device.

[0096] The energy consumption of the IoT device in processing the first computing task can be represented by the following equation:

[0097]

[0098] wherein η is a constant representing the energy conversion efficiency.

[0099] 2. When the UAV processes the second computing task, the computing delay can be represented by the following equation:

[0100] wherein R i,j represents the transmission rate of the computing task.

[0101] The energy consumption of the UAV in processing the second computing task can be represented by the following equation:

[0102]

[0103] 3. When the satellite processes the third computing task, the computing delay can be represented by the following equation:

[0104] wherein R s represents the transmission rate between the UAV and the satellite.

[0105] The energy consumption in the process of transmitting the third computing task from the UAV to the satellite can be represented by the following equation:

[0106]

[0107] Therefore, the total delay in processing the computing task can be represented by the following equation:

[0108]

[0109] i.e.,

[0110] If the computing task has been cached to the UAV side, the IoT device does not need to upload the computing task. Therefore, the total delay can be represented by the following equation:

[0111]

[0112] Therefore, the total energy consumption in processing the computing task can be represented by the following equation:

[0113]

[0114] i.e.,

[0115] Optionally, the joint optimization problem is divided into different sub-problems by block coordinate descent method, and the operation of optimizing the first variable and the second variable based on the sub-problems comprises: given the bandwidth allocation ratio and the first position coordinate; introducing a third variable, and rewriting the joint optimization problem into a first sub-problem based on the third variable; replacing the first non-convex constraint condition used for indicating the coupling relationship of the first variable and the second variable into a first convex constraint condition according to the McCormick envelope theory; and solving the first variable and the second variable according to the first convex constraint condition and the first sub-problem.

[0116] Specifically, given the bandwidth allocation ratio and the first position coordinate (i.e., {b, U U}), the associated variable a i,j ∈{(0, 1)} is relaxed into a continuous variable 0≤a i,j ≤1, and the associated decision variable c i,j ∈{(0, 1)} is relaxed into a continuous variable 0≤c i,j ≤1.

[0117] In the objective function of the optimization problem (i.e., the above formula 16), the cache decision variable c i,j and the associated variable a i,j are tightly coupled in the form of product, which leads to a difficult problem to solve. In order to solve this difficult problem, a new variable (i.e., the third variable) z i,j is introduced, i∈K, j∈N. Wherein, z i,j =(1-c i,j )a i,j , and z is defined as {z i,j}. z i,j is used to indicate the coupling relationship of the cache decision variable c i,j and the associated variable a i,j . Thus, the joint optimization problem can be rewritten into a first sub-problem:

[0118]

[0119] s.t.C1-C3, C6-C9

[0120] z i,j =(1-c i,j )a i,j , i∈K, j∈N (19)

[0121] Further, according to the McCormick envelope theory, the non-convex constraint condition z i,j =(1-c i,j )a i,j can be replaced by its McCormick convex relaxation condition, which can be specifically represented as:

[0122] z i,j ≥a i,j -c i,j , i e K, j e N (20)

[0123] z i,j ≥0, i e N, j e N (21)

[0124] z i,j ≤a i,j , i e N, j e N (22)

[0125] z i,j ≤1-c i,j , i e N, j e N (23)

[0126] where the above constraints indicate that max(a i,j -c i,j , 0) < z i,j ≤ min(a i,j , 1-c i,j ). In particular, due to the binary nature of the optimization variables, z i,j = (1-c i,j )a i,j can be proven to be equivalent to the above constraints.

[0127] Therefore, the first sub-problem can be further expressed as:

[0128]

[0129] s.t. C1-C3, C6-C11

[0130] (2-20)-(2-23)

[0131] The above problem can be solved by various convex optimization toolboxes.

[0132] Optionally, the joint optimization problem is divided into different sub-problems by block coordinate descent method, and the operation of optimizing the bandwidth allocation ratio based on the sub-problems includes: given the first variable, the second variable and the first position coordinate; rewriting the joint optimization problem as a second sub-problem; determining the second convex constraint condition corresponding to the second sub-problem; and solving the bandwidth allocation ratio according to the second convex constraint condition and the second sub-problem.

[0133] Specifically, given the bandwidth allocation ratio {b}, the first position coordinate {U u} and the associated variable {a}, the joint optimization problem can be rewritten as a second sub-problem:

[0134]

[0135] s.t.

[0136] C12: i∈K

[0137] C13: i∈K

[0138] C14:

[0139] C15: b i,j ,

[0140] C16:

[0141] C17:

[0142] where C16 is transformed from and C17 is transformed from According to the convex optimization theory, the perspective function of a concave function is also a concave function. Obviously, is the perspective function of the concave function Therefore, is proved to be a concave function of b i,j In addition, since is convex, the constraint conditions C12 and C13 are convex. In combination with other linear constraints and the objective function, it can be seen that the above-mentioned second sub-problem is a convex problem.

[0143] Therefore, it can be solved by various convex optimization toolboxes.

[0144] Alternatively, by block coordinate descent method, the joint optimization problem is divided into different sub-problems, and the operation of optimizing the position coordinates based on the sub-problems, including: given the first variable, the second variable and the bandwidth allocation ratio; rewriting the joint optimization problem as a third sub-problem; determining the second non-convex constraint condition corresponding to the third sub-problem, and converting the second non-convex constraint condition into a third convex constraint condition; solving the first position coordinate according to the third convex constraint condition and the third sub-problem.

[0145] Specifically, given the bandwidth allocation ratio {b}, the cache decision variable {c} and the associated variable {a}, the joint optimization problem can be rewritten as a third sub-problem:

[0146]

[0147] s.t.

[0148] C18:

[0149] C19:

[0150] C9: m≠j

[0151] C10: H min ≤H j ≤H max

[0152] C11:

[0153] C17:

[0154] To solve the non-convex constraints C18 and C19, first introduce auxiliary variables {x i}, and re-express C18 as:

[0155]

[0156]

[0157] Based on the successive convex approximation method, the lower bound of the left side can be derived by using the first-order Taylor approximation at a given local point :

[0158]

[0159] Then, the non-convex constraint C10 needs to be solved. At the given local point and we get the lower bound of the left side of C10:

[0160]

[0161] By replacing the left sides of C18, C19 and C10 with their lower bounds at the given local feasible point, we can get the following convex problem:

[0162]

[0163] s.t. (31)

[0164] C20:

[0165] C21:

[0166] C22:

[0167] C23:

[0168] C10: H min ≤Hj ≤H max

[0169] The above can be solved using convex optimization toolboxes, such as CVX. In the k-th iteration, we use... and To calculate the solution Since the feasible set of problem 31 is a subset of the original problem 26, the upper bound solution of problem 26 can be found by solving the approximation problem of problem 26.

[0170] Simulation results show that the improved algorithm proposed in this application significantly improves the time efficiency of minimizing computational tasks in mobile edge network models compared to other benchmark algorithms.

[0171] In addition, refer to Figure 1 As shown, according to a second aspect of this embodiment, a storage medium is provided. The storage medium includes a stored program, wherein, when the program is executed, a processor performs any of the methods described above.

[0172] Therefore, according to this embodiment, after determining the joint optimization problem (i.e., the problem of minimizing computational latency) based on the computational model, the processor in this application divides the joint optimization problem into different sub-problems using the block coordinate descent method (i.e., given the bandwidth allocation ratio and the first position coordinate, solving for the correlation variables and cache decision variables; given the correlation variables, cache decision variables, and the first position coordinate, solving for the bandwidth allocation ratio; given the correlation variables, bandwidth allocation ratio, and cache decision variables, solving for the first position coordinate). Then, the processor transforms each non-convex sub-problem into a convex sub-problem and solves each sub-problem separately, thereby determining the minimum computational latency and ensuring communication quality.

[0173] This solves the technical problem in existing technologies where the large number of variables involved in the three-layer mobile edge network model consisting of IoT devices, drones, and satellites leads to complex task latency issues.

[0174] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that the present invention is not limited to the described order of actions, because according to the present invention, some steps can be performed in other orders or simultaneously. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to the present invention.

[0175] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods according to the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0176] Example 2

[0177] Figure 4 An optimization apparatus 400 for a mobile edge network according to a first aspect of this embodiment is shown, which corresponds to the method described according to the first aspect of Embodiment 1. Reference Figure 4 As shown, the device 400 includes: a mobile edge network model construction module 410, used to construct a three-layer mobile edge network model based on IoT devices, drones, and satellites; a computational model construction module 420, used to construct a computational model related to the three-layer mobile edge network model, wherein the computational model determines the latency and energy consumption of the three-layer mobile edge network model based on the following parameters: a first variable between IoT devices and drones, a bandwidth allocation ratio between IoT devices and drones, a second variable between IoT devices and drones, and the first position coordinates of the drones; a joint optimization problem determination module 430, used to determine a joint optimization problem based on the computational model, wherein the joint optimization problem is used to indicate the computational latency to be minimized; and an optimization module 440, used to divide the joint optimization problem into different sub-problems using the block coordinate descent method, and optimize the first variable, bandwidth allocation ratio, second variable, and first position coordinates based on the sub-problems respectively.

[0178] Optionally, the computational model building module 420 includes: a channel model building module for building a channel model between IoT devices and drones; and a computational model building submodule for building a computational model between IoT devices, drones, and satellites based on the channel model.

[0179] Optionally, the channel model construction module comprises: a position coordinate determination module configured to determine, according to a Euclidean coordinate system model, a first position coordinate corresponding to the UAV and a second position coordinate corresponding to the IoT device, wherein the first position coordinate comprises a flight height of the UAV and a horizontal position coordinate of the UAV, and the second position coordinate is used to indicate a horizontal position coordinate of the IoT device; a distance determination module configured to determine a distance between the IoT device and the UAV; a channel coefficient determination module configured to determine a channel coefficient between the IoT device and the UAV; and a data transmission rate determination module configured to determine, according to the determined distance and the channel coefficient, a data transmission rate of the IoT device for offloading a computing task to the UAV.

[0180] Optionally, the computing model construction module 420 comprises: a computing task division module configured to divide a total computing task of the physical network device into a first computing task, a second computing task and a third computing task, wherein the first computing task is calculated on the IoT device, the second computing task is calculated on the UAV, and the third computing task is calculated on the satellite; a computing time delay determination module configured to determine a first computing time delay of the IoT device for calculating the first computing task, a second computing time delay of the UAV for calculating the second computing task, and a third computing time delay of the satellite for calculating the third computing task; an energy consumption determination module configured to determine a first energy consumption of the IoT device for calculating the first computing task, a second energy consumption of the UAV for calculating the second computing task, and a third energy consumption of the satellite for calculating the third computing task; a total time delay determination module configured to determine a total time delay for calculating the computing task according to the first computing time delay, the second computing time delay and the third computing time delay; and a total energy consumption determination module configured to determine a total energy consumption for calculating the computing task according to the first energy consumption, the second energy consumption and the third energy consumption.

[0181] Optionally, the optimization module 440 comprises: a first given module configured to give the bandwidth allocation ratio and the first position coordinate; a first rewriting module configured to introduce a third variable and rewrite the joint optimization problem into a first sub-problem based on the third variable; a first replacement module configured to replace a first non-convex constraint condition into a first convex constraint condition according to the McCormick envelope theory, wherein the first non-convex constraint condition is used to indicate a coupling relationship between the first variable and the second variable; and a first solving module configured to solve the first variable and the second variable according to the first convex constraint condition and the first sub-problem.

[0182] Optionally, the optimization module 440 comprises: a second given module configured to give the first variable, the second variable and the first position coordinate; a second rewriting module configured to rewrite the joint optimization problem into a second sub-problem; a first determination module configured to determine a second convex constraint condition corresponding to the second sub-problem; and a second solving module configured to solve the bandwidth allocation ratio according to the second convex constraint condition and the second sub-problem.

[0183] Optionally, the optimization module 440 includes: a third given module for giving the first variable, the second variable, and the bandwidth allocation ratio; a third rewriting module for rewriting the joint optimization problem into a third subproblem; a second determining module for determining the second non-convex constraint condition corresponding to the third subproblem and transforming the second non-convex constraint condition into a third convex constraint condition; and a third solving module for solving the position coordinates based on the third convex constraint condition and the third subproblem.

[0184] Therefore, according to this embodiment, after determining the joint optimization problem (i.e., the problem of minimizing computational latency) based on the computational model, the processor in this application divides the joint optimization problem into different sub-problems using the block coordinate descent method (i.e., given the bandwidth allocation ratio and the first position coordinate, solving for the correlation variables and cache decision variables; given the correlation variables, cache decision variables, and the first position coordinate, solving for the bandwidth allocation ratio; given the correlation variables, bandwidth allocation ratio, and cache decision variables, solving for the first position coordinate). Then, the processor transforms each non-convex sub-problem into a convex sub-problem and solves each sub-problem separately, thereby determining the minimum computational latency and ensuring communication quality.

[0185] Example 3

[0186] Figure 5 An optimization apparatus 500 for a mobile edge network according to a first aspect of this embodiment is shown, which corresponds to the method described according to the first aspect of Embodiment 1. Reference Figure 5 As shown, the device 500 includes: a processor 510; and a memory 520 connected to the processor 510, for providing the processor 510 with instructions to perform the following processing steps: constructing a three-layer mobile edge network model based on IoT devices, drones, and satellites; constructing a computational model related to the three-layer mobile edge network model, wherein the computational model determines the latency and energy consumption of the three-layer mobile edge network model based on the following parameters: a first variable between the IoT device and the drone, a bandwidth allocation ratio between the IoT device and the drone, a second variable between the IoT device and the drone, and the first position coordinates of the drone; determining a joint optimization problem according to the computational model, wherein the joint optimization problem is used to indicate the computational latency to be minimized; and dividing the joint optimization problem into different sub-problems using the block coordinate descent method, and optimizing the first variable, bandwidth allocation ratio, second variable, and first position coordinates based on the sub-problems respectively.

[0187] Therefore, according to the embodiment, the processor in the application divides the joint optimization problem (i.e., the problem of minimizing the calculation delay) into different sub-problems (i.e., solving the associated variables and the cache decision variables given the bandwidth allocation ratio and the first position coordinates; solving the bandwidth allocation ratio given the associated variables, the cache decision variables and the first position coordinates; solving the first position coordinates given the associated variables, the bandwidth allocation ratio and the cache decision variables) by the block coordinate descent method after determining the joint optimization problem according to the calculation model. Then, the processor converts each non-convex sub-problem into a convex sub-problem and solves each sub-problem respectively, so as to determine the minimized calculation delay and further ensure the communication quality.

[0188] The above-mentioned serial numbers of the embodiments of the application are only for description, and do not represent the advantages and disadvantages of the embodiments.

[0189] In the above-mentioned embodiments of the application, the description of each embodiment has its own focus, and the parts not described in detail in a certain embodiment can be referred to the related description of other embodiments.

[0190] In several embodiments provided in the present application, it should be understood that the disclosed technical contents can be implemented by other ways. Among them, the above-mentioned device embodiments are only schematic, for example, the division of the units is only a logical function division, and actual implementation can have another division way, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the units or modules shown or discussed can be indirect coupling or communication connection through some interfaces, units or modules, which can be electrical or other forms.

[0191] The units described as separate components can or can not be physically separated, and the components shown as units can or can not be physical units, that is, they can be located in one place or distributed on a plurality of network units. According to actual needs, part or all of the units can be selected to achieve the purpose of the embodiment.

[0192] In addition, each functional unit in each embodiment of the application can be integrated in one processing unit, or each unit can exist physically, or two or more units can be integrated in one unit. The above integrated unit can be realized in the form of hardware or in the form of software functional unit.

[0193] The integrated unit, if implemented in the form of a software function unit and sold or used as an independent product, can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application, essentially or in other words, the part that contributes to the prior art or the whole or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, including a number of instructions to make a computer device (which can be a personal computer, a server or a network device, etc.) execute all or part of the steps of the methods described in various embodiments of the present application. The aforementioned storage medium includes: a U disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a mobile hard disk, a magnetic disk or an optical disk, and various media that can store program codes.

[0194] The above is only the preferred embodiment of the present application, and it should be pointed out that for those skilled in the art, without departing from the principles of the present application, a number of improvements and refinements can be made, and these improvements and refinements should be considered as the protection scope of the present application.

Claims

1. An optimization method for mobile edge networks, characterized in that, include: Construct a three-layer mobile edge network model based on IoT devices, drones, and satellites; A computational model is constructed related to the three-layer mobile edge network model, wherein the computational model determines the latency and energy consumption of the three-layer mobile edge network model based on the following parameters: a first variable between the IoT device and the drone, a bandwidth allocation ratio between the IoT device and the drone, a second variable between the IoT device and the drone, and a first location coordinate of the drone, wherein the first variable is an association variable and is used to indicate whether the IoT device is connected to the drone, and the second variable is a caching decision variable and is used to indicate whether the IoT device decides to cache the computing task to the drone; Based on the computational model, a joint optimization problem is determined, wherein the joint optimization problem is used to indicate the computational delay to be minimized; as well as The joint optimization problem is divided into different sub-problems using the block coordinate descent method. Based on these sub-problems, optimizations are performed on the first variable, the bandwidth allocation ratio, the second variable, and the first position coordinate. Each sub-problem includes: a first sub-problem for solving the first variable and the second variable given the bandwidth allocation ratio and the first position coordinate; a second sub-problem for solving the bandwidth allocation ratio given the associated variable, the cache decision variable, and the first position coordinate; and a third sub-problem for optimizing the first position coordinate given the associated variable, the cache decision variable, and the bandwidth allocation ratio. The operation of optimizing the bandwidth allocation based on these sub-problems includes: Given the first variable, the second variable, and the first position coordinates; The joint optimization problem is rewritten as the second subproblem; Determine the second convex constraint condition corresponding to the second subproblem; and The bandwidth allocation is solved based on the second convex constraint and the second subproblem.

2. The method according to claim 1, characterized in that, The operations for constructing the computational model associated with the three-layer mobile edge network model include: Construct a channel model between the IoT device and the drone; and A computational model is constructed based on the channel model to connect the IoT device, the drone, and the satellite.

3. The method according to claim 1, characterized in that, The operation of constructing the channel model between the IoT device and the drone includes: Based on the Euclidean coordinate system model, a first position coordinate corresponding to the drone and a second position coordinate corresponding to the IoT device are determined, wherein the first position coordinate includes the drone's flight altitude and the drone's horizontal position coordinate, and the second position coordinate is used to indicate the IoT device's horizontal position coordinate; Determine the distance between the IoT device and the drone; Determine the channel coefficient between the IoT device and the drone; and Based on the determined distance and the channel coefficient, the data transmission rate at which the IoT device offloads computing tasks to the drone is determined.

4. The method according to claim 1, characterized in that, The operations for constructing the computational model associated with the three-layer mobile edge network model include: The total computing task of the IoT device is divided into a first computing task, a second computing task, and a third computing task, wherein the first computing task is performed on the IoT device, the second computing task is performed on the drone, and the third computing task is performed on the satellite. The first computing latency of the IoT device when calculating the first computing task, the second computing latency of the UAV when calculating the second computing task, and the third computing latency of the satellite when calculating the third computing task are determined. Determine the first energy consumption of the IoT device when calculating the first computing task, the second energy consumption of the drone when calculating the second computing task, and the third energy consumption of the satellite when calculating the third computing task; The total latency for the computation task is determined based on the first computation latency, the second computation latency, and the third computation latency; and The total energy consumption for the computation task is determined based on the first energy consumption, the second energy consumption, and the third energy consumption.

5. The method according to claim 1, characterized in that, The joint optimization problem is divided into different subproblems using the block coordinate descent method, and the optimization of the first and second variables based on the subproblems includes: Given the bandwidth allocation ratio and the first location coordinates; A third variable is introduced, and the joint optimization problem is rewritten into a first subproblem based on the third variable, wherein the third variable is used to indicate the coupling relationship between the first variable and the second variable. According to McCormick's envelope theory, the first non-convex constraint is replaced with a first convex constraint, wherein the first non-convex constraint is used to indicate the coupling relationship between the first variable and the second variable; and Solve for the first variable and the second variable based on the first convex constraint and the first subproblem.

6. The method according to claim 1, characterized in that, The joint optimization problem is divided into different sub-problems using the block coordinate descent method, and the operation of optimizing the first position coordinates based on the sub-problems includes: Given the first variable, the second variable, and the bandwidth allocation ratio; The joint optimization problem is rewritten as a third subproblem; Determine the second non-convex constraint condition corresponding to the third subproblem, and transform the second non-convex constraint condition into the third convex constraint condition; The position coordinates are solved based on the third convex constraint and the third subproblem.

7. A storage medium, characterized in that, The storage medium includes a stored program, wherein, when the program is executed, the method described in any one of claims 1 to 6 is performed by a processor.

8. An optimization device for a mobile edge network, characterized in that, include: The Mobile Edge Network Model Building Module is used to build a three-layer mobile edge network model based on IoT devices, drones, and satellites. A computational model building module is used to construct a computational model related to the three-layer mobile edge network model. The computational model determines the latency and energy consumption of the three-layer mobile edge network model based on the following parameters: a first variable between the IoT device and the drone, a bandwidth allocation ratio between the IoT device and the drone, a second variable between the IoT device and the drone, and the first position coordinates of the drone. The first variable is an association variable, which is used to indicate whether the IoT device is connected to the drone. The second variable is a caching decision variable, which is used to indicate whether the IoT device decides to cache the computational task to the drone. A joint optimization problem determination module is used to determine a joint optimization problem based on the computational model, wherein the joint optimization problem is used to indicate the computational delay to be minimized; as well as An optimization module is used to divide the joint optimization problem into different sub-problems using a block coordinate descent method, and to optimize the first variable, the bandwidth allocation ratio, the second variable, and the first position coordinate based on the sub-problems respectively. The sub-problems include: a first sub-problem for solving the first variable and the second variable given the bandwidth allocation ratio and the first position coordinate; a second sub-problem for solving the bandwidth allocation ratio given the associated variable, the cache decision variable, and the first position coordinate; and a third sub-problem for optimizing the first position coordinate given the associated variable, the cache decision variable, and the bandwidth allocation ratio. Furthermore, the optimization module includes: a second given module, used to give the first variable, the second variable, and the first position coordinates; The second rewriting module is used to rewrite the joint optimization problem into the second subproblem; A first determining module is used to determine the second convex constraint condition corresponding to the second subproblem; and The second solution module is used to solve the bandwidth allocation ratio based on the second convex constraint and the second subproblem.

9. An optimization device for a mobile edge network, characterized in that, include: processor; as well as A memory, connected to the processor, for providing the processor with instructions to perform the following processing steps: Construct a three-layer mobile edge network model based on IoT devices, drones, and satellites; A computational model is constructed related to the three-layer mobile edge network model, wherein the computational model determines the latency and energy consumption of the three-layer mobile edge network model based on the following parameters: a first variable between the IoT device and the drone, a bandwidth allocation ratio between the IoT device and the drone, a second variable between the IoT device and the drone, and a first location coordinate of the drone, wherein the first variable is an association variable and is used to indicate whether the IoT device is connected to the drone, and the second variable is a caching decision variable and is used to indicate whether the IoT device decides to cache the computing task to the drone; Based on the computational model, a joint optimization problem is determined, wherein the joint optimization problem is used to indicate the computational delay to be minimized; as well as The joint optimization problem is divided into different sub-problems using the block coordinate descent method. Based on these sub-problems, optimizations are performed on the first variable, the bandwidth allocation ratio, the second variable, and the first position coordinate. Each sub-problem includes: a first sub-problem for solving the first variable and the second variable given the bandwidth allocation ratio and the first position coordinate; a second sub-problem for solving the bandwidth allocation ratio given the associated variable, the cache decision variable, and the first position coordinate; and a third sub-problem for optimizing the first position coordinate given the associated variable, the cache decision variable, and the bandwidth allocation ratio. The operation of optimizing the bandwidth allocation based on these sub-problems includes: Given the first variable, the second variable, and the first position coordinates; The joint optimization problem is rewritten as the second subproblem; Determine the second convex constraint condition corresponding to the second subproblem; and solve the bandwidth allocation based on the second convex constraint condition and the second subproblem.

Images