Space-air-ground integrated industrial internet task unloading and energy collecting device

By constructing a mobile edge computing model of the integrated industrial Internet of space and earth and the Liyapunov optimization method, the problem of limited computing resources and energy of terminal equipment is solved, efficient optimization of task unloading and energy collection is achieved, and the processing rate and energy of the equipment is improved. Utilization.

CN118963980APending Publication Date: 2024-11-15BEIJING INFORMATION SCI & TECH UNIV
View PDF 0 Cites 1 Cited by

Patent Information

Application Number
CN202310498817.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2023-05-06
Publication Date
2024-11-15

AI Technical Summary

Technical Problem

The computing resources and energy storage of the integrated industrial Internet terminal equipment in the sky and earth are limited, resulting in delays in computing tasks and energy management difficulties, which are difficult to effectively solve in the existing technology.

Method used

A space-space and earth integrated industrial Internet task unloading and energy harvesting device was designed. By constructing a mobile edge computing model, the profit function and constraints of terminal equipment were defined, and the problem was transformed into a problem of minimizing the stable yield difference of queues. It also proposes an iterative algorithm for task offloading and energy harvesting optimization.

Benefits of technology

The calculation task processing rate and energy utilization rate of terminal equipment are improved, efficient utilization of resources is achieved, task accumulation and energy consumption are reduced, and system stability and efficiency are improved.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118963980A_ABST
    Figure CN118963980A_ABST
Patent Text Reader

Abstract

The invention discloses a space-air-ground integrated industrial internet task unloading and energy collecting device, and belongs to the technical field of calculation, reckoning or counting. According to the device, calculation task unloading and energy collection of the space-air-ground integrated industrial internet terminal equipment are jointly considered, and a strategy for solving the problems of storable energy and limited calculation capability of the terminal equipment is provided, so that the terminal equipment can complete a local calculation task as soon as possible; energy is obtained through the radio frequency signal transmitting station and stored in a battery, so that continuous and normal operation of equipment is guaranteed. According to the device, firstly, a space-air-ground integrated industrial internet mobile edge computing model is constructed, the space-air-ground integrated industrial internet mobile edge computing model comprises a computing task unloading part and an energy collection part, and the space-air-ground integrated industrial internet terminal equipment computing task unloading and processing and energy collection modes are described; secondly, defining a terminal equipment revenue function and constraint conditions to combine task unloading and energy collection; and finally, through a Lyapunov optimization method, the problem is converted into a problem of poor stability income of a minimization queue based on a time slot, then an iterative algorithm is proposed to obtain an optimal solution, and a simulation result shows that the proposed algorithm has relatively good performance.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to communication technology, and specifically discloses an air-space-ground integrated industrial Internet task unloading and energy collection device, which belongs to the technical field of calculation, estimation or counting. Background Art

[0002] The popularization of the fifth generation of mobile communication technology has brought more opportunities for the further development of the industrial Internet of Things, and the space-air-ground integration technology has expanded the communicable space to land, sea, sky, and space, further meeting the needs of various users to communicate anytime and anywhere, and realizing wide-area wireless communication coverage around the world. It can be seen that the space-air-ground integration network has greatly changed people's lives, and studying it has great practical significance.

[0003] Although the integrated space-ground network has become an indispensable part of people's lives, the computing resources and storable energy of the terminal devices of the network are limited, which restricts its application. If all computing tasks are left to be executed locally on the device, it may cause excessive backlog of computing tasks, which will bring unbearable delays to the device and pose a huge challenge to the power of the device. Mobile Edge Computing (MEC) technology can offload part or all of the computing tasks of the integrated space-ground network terminal devices to the edge server for processing. The edge server has abundant computing resources and can complete the computing tasks that are difficult for the device to handle more quickly, relieve the pressure on the terminal, reduce service delays, and improve user experience. Therefore, formulating a reasonable computing task offloading strategy for the integrated space-ground terminal devices can further improve the resource utilization of the entire system, reduce the overall system overhead, and make the MEC integrated space-ground network system run more efficiently and stably.

[0004] At the same time, the emergence of energy harvesting (EH) technology has solved the problem of limited energy of air-ground integrated terminal devices to a certain extent. The application of EH technology enables devices that do not have wired charging function to collect weak radio frequency energy in the surrounding area, and then convert it into electrical energy and store it in its own battery to maintain the long-term normal use of the device. However, when the collected energy exceeds the battery capacity, this part of the excess energy cannot only be used by the device, but also brings a greater burden to the normal operation of the device. Therefore, how to reasonably collect energy has become one of the research hotspots in the MEC air-ground integrated network. Summary of the invention

[0005] Aiming at the problem that the terminal equipment of space-ground-integrated industrial Internet can store limited energy and computing power, the present invention designs a space-ground-integrated industrial Internet task offloading and energy collection device. Firstly, a space-ground-integrated industrial Internet mobile edge computing model is constructed, which includes two parts: computing task offloading and energy collection. The method of computing task offloading and processing and energy collection of space-ground-integrated industrial Internet terminal equipment is described; secondly, the terminal equipment benefit function and constraint conditions are defined to combine task offloading and energy collection; finally, the problem is transformed into a time slot-based minimization queue stable benefit difference problem through the Lyapunov optimization method, and then an iterative algorithm is proposed to obtain the optimal solution. The simulation results show that the proposed algorithm has good performance.

[0006] The air-ground-space integrated industrial Internet task unloading and energy collection device of the present invention includes the following three steps:

[0007] 1) Construct an air-space-ground integrated industrial Internet mobile edge computing model, which contains D air-space-ground integrated industrial Internet terminal devices, denoted as set D = {SD1, SD2, ..., SD i , …, SD D}, where SD i Represents the i-th air-ground integrated industrial Internet terminal device in the system. Each terminal device is equipped with an antenna, a radio frequency energy harvester and a rechargeable battery, and has a certain computing power. Different devices are heterogeneous, and their distances from each server, CPU cycles for processing each megabit of data, chip switch capacitance, control parameters, etc. are all different. The model contains R radio frequency signal transmission stations, denoted as the set F = {RF1, RF2, ..., RF q , …, RF R}, where RF q represents the qth RF signal transmitting station in the system. These signal transmitting stations will transmit RF signals to the terminal devices, and the RF energy harvesters equipped in the terminal devices will obtain energy from the RF signals and store it for subsequent use. In addition, the system also includes S drones with servers, and the drone servers they carry are recorded as a set S = {US1, US2, …, US j , …, US s}, where US j Represents the jth UAV server in the system. The UAV server has more computing resources and stronger computing power, and can complete the computing tasks unloaded by different terminal devices. At the same time, the model contains a low-orbit satellite server, which has strong computing power. The terminal device can offload the computing tasks to the low-orbit satellite server for execution. Compared with the UAV server, the information transmission energy consumption of offloading the task to the low-orbit satellite server is larger, but the computing power of the low-orbit satellite server is stronger than that of the UAV server.

[0008] According to the data transmission characteristics of the terminal device, the computing task can be divided into data volumes. Due to the limited computing resources of the terminal device, each time slot may have unfinished tasks in the previous time slot. i The task backlog at time slot t is expressed as H i (t) indicates SD i The task stacking relationship between time slot t and time slot t+1 can be expressed as

[0009] H i (t+1)=max{H i (t)-C i (t), 0}+A i (t) (1)

[0010] Among them, A i (t) represents SD i The newly added computing task in time slot t, C i (t) represents the computational tasks completed in time slot t, and H i (t) should also satisfy the following formula

[0011]

[0012] Where T is the duration of each time slot, and E[x] represents the mathematical expectation of x.

[0013] In time slot t, SD i The latency of executing local computing tasks is

[0014]

[0015] in, The superscript l indicates local computing, the superscript e indicates the execution of computing tasks, and the subscript i indicates SD i . Indicates SD i The amount of task data executed locally, λ i Indicates SD i The number of CPU cycles required for each Mbit of data in the computation task, f i l (t) represents SD i The CPU cycle frequency meets the constraint where f i min and f i max Respectively represent the minimum and maximum values ​​of the local CPU cycle frequency.

[0016] SD i The local computing energy consumption can be expressed as

[0017]

[0018] Among them, κ i Indicates SD i The effective switch capacitance value of the chip used.

[0019] SD i The benefit of executing local computing tasks can be expressed as

[0020]

[0021] Among them, i Indicates SD i benefit parameters.

[0022] The computing tasks offloaded by the terminal device will be transmitted from the device to the drone server through the wireless channel. At the same time, assuming that different channels are used for information transmission between different terminal devices, there will be no interference between information transmission. i and US j The transfer rate of computing tasks between

[0023]

[0024] in, For SD i and US j The path loss of information transmission between u represents the channel bandwidth between the UAV server and the terminal device, N represents the variance of the noise, and P i d Indicates SD i The superscript d represents the transmission power of the terminal device.

[0025] SD i The data volume is The computing tasks are offloaded to US j The transmission delay can be calculated by the following formula

[0026]

[0027] in, The superscript o in represents task offloading.

[0028] The amount of data to be uninstalled is The computing tasks to US j When SD i The energy required can be calculated by the following formula:

[0029]

[0030] SD iThe data volume is The computing tasks are offloaded to US j The communication cost is

[0031]

[0032] in, Indicates that the j The transmission communication cost required for each Mbit of data during the communication process.

[0033] For US receiving the unloading task j , US j The amount of data processed is The time required for the calculation task can be calculated by the following formula

[0034]

[0035] in, Indicates US j Assigned to Offload from SD i The CPU cycle frequency of the computing task.

[0036] US j The amount of data processed is The energy consumed by the computing task can be calculated by the following formula

[0037]

[0038] Among them, κ j Indicates US j The effective switch capacitance value of the chip used.

[0039] SD i Offloading computational tasks to US j The benefit is expressed as

[0040]

[0041] The computing tasks offloaded by the terminal device are also transmitted from the device to the low-orbit satellite server through the wireless channel. i The channel gain of information transmission between the low-orbit satellite server and the low-orbit satellite server can be calculated by Calculate, where G s represents the receiving antenna gain of the low-orbit satellite, Indicates SD i The transmitting antenna gain, L represents the wavelength of the signal, Indicates SD i The communication distance between the low-orbit satellite server and the rain Indicates rainfall decline.

[0042] SD iThe computing task transmission rate between the low-orbit satellite server is

[0043]

[0044] Among them, W s Represents low-orbit satellite server and SD i The channel bandwidth between .

[0045] SD i The data volume is The transmission delay of the task offloaded to the low-orbit satellite server can be calculated by the following formula

[0046]

[0047] The amount of data to be uninstalled is When the computing task is sent to the low-orbit satellite server, SD i The energy required can be calculated by the following formula:

[0048]

[0049] SD i The data volume is The communication cost of offloading the computing task to the low-orbit satellite server is

[0050]

[0051] Among them, μ s It represents the transmission communication cost per Mbit of data during the communication with the low-orbit satellite server.

[0052] The amount of data processed by the low-orbit satellite server is The time required for the calculation task can be calculated by the following formula

[0053]

[0054] in, Indicates that the low-orbit satellite server is allocated to the unloaded SD i The CPU cycle frequency of the computing task.

[0055] The amount of data processed by the low-orbit satellite server is The energy consumed by the computing task can be calculated by the following formula

[0056]

[0057] Among them, κ s Represents the effective switching capacitance value of the chip used in the low-orbit satellite server.

[0058] SD iThe benefit of offloading computing tasks to low-orbit satellite servers can be expressed as

[0059]

[0060] Since the energy collection process is independent in different time slots and all comes from R RF signal transmitting stations, the SD in time slot t can be obtained according to the Friis formula i The collected energy is calculated as

[0061]

[0062] in, Indicates RF q The transmission power, the superscript r represents the RF signal transmitting station, and the subscript q represents the RF q . Indicates RF q The transmit antenna gain, f q Indicates RF q The transmission frequency, Indicates SD i The energy collection receiving antenna gain, c represents the propagation speed of electromagnetic waves, d i,q Indicates SD i With RF q The distance between them, η represents the energy collection efficiency, T i h Indicates SD i The energy harvesting time.

[0063] However, only a part of the collected energy can be charged into the battery of the terminal device within the time slot t, so the actual amount of energy charged into the SD i Battery Energy Must exist SD i The energy consumed in time slot t is mainly used to perform local computing tasks and offload tasks to the UAV server and low-orbit satellite server, so it can be concluded that SD i The total energy consumption in time slot t is as follows

[0064]

[0065] SD between time slot t and time slot t+1 i The change relationship of the remaining battery energy can be expressed by the following formula

[0066]

[0067] 2) Define the revenue function of the integrated space-ground industrial Internet terminal equipment as

[0068]

[0069] in, For US j The unit energy cost, ω s is the unit energy cost of low-orbit satellite servers.

[0070] The actual power consumption of each space-ground integrated industrial Internet terminal device is constrained to The energy consumption constraint of the terminal device is in and SD i The minimum and maximum energy consumption of the terminal device task accumulation constraint is The CPU cycle frequency constraint of the terminal device is The CPU cycle frequency constraint of the drone server is The CPU cycle frequency constraint of the low-orbit satellite server is The stability constraint of task stacking is

[0071] 3) By introducing a virtual energy queue, the queue stability constraint is written into the objective function of the original optimization problem, thereby converting the optimization problem into a time-slot-based problem of minimizing the queue stability benefit difference through the Lyapunov optimization method, and this problem is divided into two sub-optimization problems: task offloading and energy collection. Then an iterative algorithm is proposed to jointly solve the optimal solutions of the two sub-optimization problems using a linear method and a particle swarm optimization algorithm. Finally, the proposed algorithm is selected for simulation experiments compared with the local execution algorithm only. The simulation results show that the proposed algorithm can complete the locally accumulated computing tasks at a faster speed and has a higher average energy utilization rate.

[0072] The present invention adopts the above-mentioned technical scheme and has the following beneficial effects: the present invention comprehensively considers the computing task offloading and energy collection of the integrated air-space-ground industrial Internet terminal equipment, takes maximizing the terminal equipment profit as the objective function, and takes the actual amount of power charged by the terminal equipment, energy consumption, task accumulation amount, stability of task accumulation, CPU cycle frequency and CPU cycle frequency of the drone server and the low-orbit satellite server as constraints, thereby improving the local task processing rate and average energy utilization of the terminal equipment and realizing efficient resource utilization. BRIEF DESCRIPTION OF THE DRAWINGS

[0073] In order to make the purpose, technical solutions and advantages of the present invention more clear, the present invention will be further described in detail below with reference to the accompanying drawings, in which:

[0074] Figure 1 It is a mobile edge computing model for the integrated air-ground-space industrial Internet;

[0075] Figure 2 This is a schematic diagram comparing the task accumulation of the proposed algorithm and the algorithm executed locally only;

[0076] Figure 3 Schematic diagram of the average energy utilization of the proposed algorithm. DETAILED DESCRIPTION

[0077] The specific embodiments of the present invention are further described in detail below in conjunction with the accompanying drawings.

[0078] Figure 1 The mobile edge computing model of the integrated air-space-ground industrial Internet. The model contains D integrated air-space-ground industrial Internet terminal devices, which are recorded as a set D = {SD1, SD2, ..., SD i , …, SD D}, where SD i Represents the i-th air-ground integrated industrial Internet terminal device in the system. Each terminal device is equipped with an antenna, a radio frequency energy harvester and a rechargeable battery, and has a certain computing power. Different devices are heterogeneous, and their distances from each server, CPU cycles for processing each megabit of data, chip switch capacitance, control parameters, etc. are all different. The model contains R radio frequency signal transmission stations, denoted as the set F = {RF1, RF2, ..., RF q , …, RF R}, where RF q represents the qth RF signal transmitting station in the system. These signal transmitting stations will transmit RF signals to the terminal devices, and the RF energy harvesters equipped in the terminal devices will obtain energy from the RF signals and store it for subsequent use. In addition, the system also includes S drones with servers, and the drone servers they carry are recorded as a set S = {US1, US2, …, US j , …, US s}, where US j Represents the jth UAV server in the system. The UAV server has more computing resources and stronger computing power, and can complete the computing tasks unloaded by different terminal devices. At the same time, the model contains a low-orbit satellite server, which has strong computing power. The terminal device can offload the computing tasks to the low-orbit satellite server for execution. Compared with the UAV server, the information transmission energy consumption of offloading the task to the low-orbit satellite server is larger, but the computing power of the low-orbit satellite server is stronger than that of the UAV server.

[0079] In order to ensure the continuous operation of the air-ground integrated industrial Internet terminal equipment and make rational use of computing resources, the present invention constructs an optimization problem with the goal of maximizing the sum of the benefits of the local terminal equipment, the UAV server, and the low-orbit satellite server performing tasks, minus the sum of the communication cost of offloading tasks to other servers and the energy cost consumed by other servers to perform tasks, which can be expressed as

[0080]

[0081]

[0082]

[0083]

[0084]

[0085]

[0086]

[0087]

[0088] Constraint (25) indicates that the amount of power actually charged into each terminal device will be less than the energy collected by each device and not less than zero. Constraint (26) indicates that the total energy consumption of each terminal device is not less than the minimum energy consumption and is not greater than the remaining energy at the beginning of the time slot and the maximum energy consumption. Constraint (27) indicates that the sum of the computing tasks processed locally by each terminal device and the computing tasks decided to be offloaded to the UAV server and the low-orbit satellite server does not exceed the tasks accumulated in the time slot. Constraint (28) is the CPU cycle frequency constraint of the terminal device. Constraint (29) is the CPU cycle frequency constraint when the UAV server processes the offloaded tasks. Constraint (30) is the CPU cycle frequency constraint when the low-orbit satellite server processes the offloaded tasks. Constraint (31) ensures the stability of task accumulation.

[0089] The problem is transformed into a time slot-based problem of minimizing the stable revenue difference of the queue through the Lyapunov optimization method; that is, in order to prevent excessive consumption of battery power and eliminate the coupling relationship between battery power in different time slots, a virtual energy queue is introduced, and the queue stability constraint is written into the objective function of the original optimization problem, thereby transforming the optimization problem into:

[0090]

[0091]

[0092]

[0093]

[0094]

[0095]

[0096]

[0097] This optimization problem can be divided into two sub-optimization problems: task offloading and energy harvesting. The energy harvesting sub-problem is expressed as

[0098]

[0099]

[0100] The task offloading subproblem is expressed as

[0101]

[0102]

[0103]

[0104]

[0105]

[0106]

[0107] The above two sub-optimization problems can be solved by linear method and particle swarm optimization algorithm respectively; therefore, an iterative algorithm is proposed. First, the energy collection sub-problem is solved by linear method; secondly, constraints (42) and (43) are converted into penalty terms and written into the optimization objective function (41) of the task offloading sub-problem to construct the fitness function of the particle swarm optimization algorithm, and the idea of ​​the particle swarm optimization algorithm is used to iteratively solve the converted unconstrained optimization problem.

[0108] Figure 2The figure is a comparison diagram of the task accumulation of the proposed algorithm and the local execution algorithm. The figure shows the change of the task accumulation of three heterogeneous terminal devices obtained by the proposed algorithm and the local execution algorithm over time when the initial task accumulation is 500Mbit and there is no new computing task added in each time slot. The horizontal axis in the figure represents time, the vertical axis represents the task accumulation of the terminal device, and the solid line and the dotted line represent the task accumulation of the proposed algorithm and the local execution algorithm respectively. It can be seen from the figure that the device task accumulation of the two algorithms will gradually decrease with the increase of time and eventually approach zero, which indicates that the local accumulated tasks of the terminal device will be gradually processed until the local accumulated tasks are close to zero. By comparing the curves of the proposed algorithm and the local execution algorithm, it can be seen that when there is no new computing task added, the proposed algorithm can complete all the accumulated computing tasks at a faster speed, and its completion time can be reduced by up to 85.08% compared with the local execution algorithm. The proposed algorithm is effective.

[0109] Figure 3 The figure is a schematic diagram of the average energy utilization of the proposed algorithm. The diagram shows the cumulative energy collected and consumed by the proposed algorithm at the end of each time slot, when the initial task accumulation is 500Mbit and the new computing task amount added by the terminal device in each time slot is 17Mbit. In the figure, the horizontal axis represents time, the vertical axis represents the cumulative collected energy and the cumulative consumed energy, the solid line represents the cumulative collected energy, and the histogram represents the cumulative consumed energy. From this figure, it can be seen that at the end of any time slot of the proposed algorithm, the cumulative energy consumed by the device is less than the cumulative collected energy, which is in line with the causal principle of energy collection. And for all 200 time slots, the average energy utilization (cumulative energy consumption / cumulative energy collection) of the proposed algorithm can reach 97.36%

[0110] The above specific embodiments further illustrate the invention objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above specific embodiments are only exemplary descriptions and are not used to limit the protection scope of the present invention. Any modifications, equivalent substitutions and transformations made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. Establish an integrated air-ground-space industrial Internet task unloading and energy collection device, characterized in that: The following steps are involved: 1) Construct an integrated space-ground-industrial Internet mobile edge computing model, including computing task offloading and energy collection, and describe the computing task offloading and processing and energy collection methods of the integrated space-ground-industrial Internet terminal equipment; 2) Define the terminal device benefit function and constraints to jointly offload tasks and harvest energy; 3) The problem is transformed into a time-slot-based minimization queue stable revenue difference problem through the Lyapunov optimization method, and then an iterative algorithm is proposed to obtain the optimal solution. The simulation results show that the proposed algorithm has good performance.

2. The device for establishing an integrated space-ground-air industrial Internet task unloading and energy collection device according to claim 1 is characterized in that: In step 1), an air-space-ground integrated industrial Internet mobile edge computing model is constructed, which includes D air-space-ground integrated industrial Internet terminal devices, denoted as a set D = {SD1, SD2, ..., SD i ,…,SD D }, where SD i represents the i-th air-space-ground integrated industrial Internet terminal device in the system. Each terminal device is equipped with an antenna, a radio frequency energy harvester and a rechargeable battery, and has a certain computing power. Different devices are heterogeneous, and their distances from each server, CPU cycles for processing each megabit of data, chip switch capacitance, control parameters, etc. are all different. The model contains R radio frequency signal transmission stations, which are recorded as the set F = {RF1, RF2, …, RF q , …, RF R }, where RF q represents the qth RF signal transmitting station in the system; these signal transmitting stations will transmit RF signals to the terminal devices, and the RF energy harvesters equipped in the terminal devices will obtain energy from the RF signals and store it for subsequent use; in addition, the system also includes S drones with servers, and the drone servers they carry are recorded as the set S = {US1, US2, …, US j , …, US S }, where US j represents the jth UAV server in the system. The UAV server has more computing resources and stronger computing power, and can complete the computing tasks unloaded by different terminal devices. At the same time, the model contains a low-orbit satellite server, which has strong computing power. The terminal device can offload the computing tasks to the low-orbit satellite server for execution. Compared with the UAV server, the information transmission energy consumption of offloading the tasks to the low-orbit satellite server is larger, but the computing power of the low-orbit satellite server is stronger than that of the UAV server. According to the data transmission characteristics of the terminal device, the computing task can be divided into data volumes; due to the limited computing resources of the terminal device, each time slot may have unfinished tasks in the previous time slot; i The task backlog at time slot t is expressed as H i (t) indicates SD i The task stacking relationship between time slot t and time slot t+1 can be expressed as H i (t+1)=max{H i (t)-C i (t),0}+A i (t) (1) Among them, A i (t) represents SD i The newly added computing task in time slot t, C i (t) represents the computational tasks completed in time slot t, and H i (t) should also satisfy the following formula Where T is the duration of each time slot, and E[x] represents the mathematical expectation of x; In time slot t, SD i The latency of executing local computing tasks is in, The superscript l indicates local computing, the superscript e indicates the execution of computing tasks, and the subscript i indicates SD i ; Indicates SD i The amount of task data executed locally, λ i Indicates SD i The number of CPU cycles required for each Mbit of data in the computation task is Indicates SD i The CPU cycle frequency meets the constraint where f i min and f i max Respectively represent the minimum and maximum values ​​of the local CPU cycle frequency; SD i The local computing energy consumption can be expressed as Among them, κ i Indicates SD i The effective switch capacitance value of the chip used; SD i The benefit of executing local computing tasks can be expressed as Among them, i Indicates SD i The benefit parameters; The computing tasks offloaded by the terminal device will be transmitted from the device to the drone server through the wireless channel. At the same time, assuming that different channels are used for information transmission between different terminal devices, there will be no interference between information transmissions; SD i and US j The computing task transfer rate between in, For SD i and US j The path loss of information transmission between u represents the channel bandwidth between the UAV server and the terminal device, N represents the variance of the noise, and P i d Indicates SD i The transmission power of the terminal device is represented by the superscript d. SD i The data volume is The computing tasks are offloaded to US j The transmission delay can be calculated by the following formula in, The superscript o in indicates task offloading; The amount of data to be uninstalled is The computing tasks to US j When i The energy required can be calculated by the following formula: SD i The data volume is The computing tasks are offloaded to US j The communication cost is in, Indicates that the j The transmission communication cost required for each Mbit of data during the communication process; For US receiving the unloading task j , US j The amount of data processed is The time required for the calculation task can be calculated by the following formula in, Indicates US j Assigned to Offload from SD i The CPU cycle frequency of the computing task; US j The amount of data processed is The energy consumed by the computing task can be calculated by the following formula Among them, κ j Indicates US j The effective switch capacitance value of the chip used; SD i Offloading computational tasks to US j The benefit is expressed as The computing tasks offloaded by the terminal device are also transmitted from the device to the low-orbit satellite server through the wireless channel; i The channel gain of information transmission between the low-orbit satellite server and the low-orbit satellite server can be calculated by Calculate, where G s represents the receiving antenna gain of the low-orbit satellite, Indicates SD i The transmitting antenna gain, L represents the wavelength of the signal, Indicates SD i The communication distance between the low-orbit satellite server and the rain Indicates rainfall decline; SD i The computing task transmission rate between the low-orbit satellite server is Among them, W s Represents low-orbit satellite server and SD i The channel bandwidth between SD i The data volume is The transmission delay of the task offloaded to the low-orbit satellite server can be calculated by the following formula The amount of data to be uninstalled is When the computing task is sent to the low-orbit satellite server, SD i The energy required can be calculated by the following formula: SD i The data volume is The communication cost of offloading the computing task to the low-orbit satellite server is Among them, μ s Indicates the transmission communication cost required for each Mbit of data during communication with the low-orbit satellite server; The amount of data processed by the low-orbit satellite server is The time required for the calculation task can be calculated by the following formula Among them, f i s (t) represents the low-orbit satellite server allocated to the unloaded SD i The CPU cycle frequency of the computing task; The amount of data processed by the low-orbit satellite server is The energy consumed by the computing task can be calculated by the following formula Among them, κ s Indicates the effective switch capacitance value of the chip used in the low-orbit satellite server; SD i The benefit of offloading computing tasks to low-orbit satellite servers can be expressed as Since the energy collection process is independent in different time slots and all comes from R RF signal transmitting stations, the SD in time slot t can be obtained according to the Friis formula i The collected energy is calculated as in, Indicates RF q The transmission power, the superscript r represents the RF signal transmitting station, and the subscript q represents the RF q ; Represents RF q The transmit antenna gain, f q Represents RF q The transmission frequency, Indicates SD i The energy collection receiving antenna gain, c represents the propagation speed of electromagnetic waves, d i,q Indicates SD i With RF q The distance between them, η represents the energy collection efficiency, Indicates SD i The time of energy harvesting; However, only a part of the collected energy can be charged into the battery of the terminal device within the time slot t, so the actual amount of energy charged into the SD i Battery Energy Must exist SD i The energy consumed in time slot t is mainly used to perform local computing tasks and offload tasks to the UAV server and low-orbit satellite server, so it can be concluded that SD i The total energy consumption in time slot t is as follows SD between time slot t and time slot t+1 i The change relationship of the remaining battery energy can be expressed by the following formula 3. The device for establishing an integrated space-ground-air industrial Internet task unloading and energy collection device according to claim 1 is characterized in that: The revenue function of the integrated space-ground industrial Internet terminal device defined in step 2) is the sum of the benefits of the local, drone server, and low-orbit satellite server performing tasks, minus the sum of the communication cost of offloading tasks to other servers and the energy cost consumed by other servers to perform tasks, expressed as in, For US j The unit energy cost, ω s is the unit energy cost of the low-orbit satellite server; The actual amount of power charged to each terminal device is constrained to The energy consumption constraint of the terminal device is in and SD i minimum and maximum energy consumption; The task stacking constraint is The CPU cycle frequency constraints of terminal devices, drone servers, and low-orbit satellite servers are The stability constraint of task stacking is 4. The device for establishing an integrated space-ground-air industrial Internet task unloading and energy collection device according to claim 1 is characterized in that: Step 3) The problem is transformed into a time slot-based minimization queue stability benefit difference problem through the Lyapunov optimization method; that is, in order to prevent excessive consumption of battery power and eliminate the coupling relationship between battery power in different time slots, a virtual energy queue is introduced, and the queue stability constraint is written into the objective function of the original optimization problem, thereby transforming the optimization problem into: This optimization problem can be divided into two sub-optimization problems: task offloading and energy harvesting. The energy harvesting sub-problem is expressed as The task offloading subproblem is expressed as The above two sub-optimization problems can be solved by linear method and particle swarm optimization algorithm respectively; therefore, an iterative algorithm is proposed. First, the energy collection sub-problem is solved by linear method; secondly, constraints (41) and (42) are converted into penalty terms and written into the optimization objective function (40) of the task offloading sub-problem to construct the fitness function of the particle swarm optimization algorithm, and the idea of ​​the particle swarm optimization algorithm is used to solve the converted unconstrained optimization problem; finally, the algorithm is verified by simulation.

Citation Information

Cited By

  • Unmanned cluster assisted multi-strategy network service dynamic migration method and device

    CN119653395A