Multi-mode terminal adaptive calculation unloading method and device

Abstract

The invention discloses a multi-mode terminal adaptive calculation unloading method and device, and the method comprises the steps: obtaining multi-mode terminal target information which comprises multi-mode terminal equipment information and multi-mode terminal local data information; according to the multi-mode terminal target information, calculating a target calculation unloading mode set and a multi-mode terminal calculation unloading task total time delay; according to the target calculation unloading mode set, calculating the total energy consumption of a multi-mode terminal calculation unloading task; constructing a target function according to the total time delay of the unloading task calculated by the multi-mode terminal and the total energy consumption of the unloading task calculated by the multi-mode terminal; and performing adaptive calculation unloading according to the objective function. According to the method, self-adaptive calculation unloading is realized, and the efficiency and the adaptability are improved. The method can be widely applied to the technical field of mobile communication.

Description

Technical Field

[0001] The present invention relates to the field of mobile communication technology, and in particular to a method and device for adaptive computing unloading of a multi-mode terminal. Background Art

[0002] Computation-intensive applications require huge computing power support, but mobile terminals are limited by size, power consumption and cost, and have limited computing resources. When a large number of users are connected, they face huge computing pressure, making it difficult to meet computing needs and resulting in poor user experience. Traditional computing offloading methods offload computing tasks to cloud data centers to obtain computing results, or offload computing tasks to edge servers to obtain computing results, but the offloading method is single and inefficient. At the same time, mobile terminals usually have multi-mode communication capabilities such as short-range, cellular and satellite. A single offloading method is difficult to adapt to the diversity of services and the variability of network environments, and has low adaptability.

[0003] In summary, the technical problems existing in the relevant technologies need to be improved. Summary of the invention

[0004] The embodiment of the present invention provides a method and device for adaptive calculation offloading of a multi-mode terminal, which effectively improves efficiency and adaptability.

[0005] On the one hand, an embodiment of the present invention provides a multi-mode terminal adaptive calculation offloading method, comprising the following steps:

[0006] Acquire multimode terminal target information, wherein the multimode terminal target information includes multimode terminal device information and multimode terminal local data information;

[0007] Calculate the target computing offloading mode set and the total delay of the multi-mode terminal computing offloading task according to the multi-mode terminal target information;

[0008] Calculate the total energy consumption of the multi-mode terminal computing offloading task according to the target computing offloading mode set;

[0009] Constructing an objective function according to the total delay of the offloading task calculated by the multi-mode terminal and the total energy consumption of the offloading task calculated by the multi-mode terminal;

[0010] According to the objective function, adaptive computation offloading is performed.

[0011] In some embodiments, the calculating, according to the multi-mode terminal target information, a target calculation offloading mode set and a total delay of the multi-mode terminal calculation offloading task includes:

[0012] Determine a target resource pool and terminal status information according to the multimode terminal target information, wherein the target resource pool includes a communication resource pool and a computing resource pool, and the terminal status information includes computing task requirements, resource status, and communication capability information;

[0013] Determining the target computing offloading mode set according to the target resource pool and the terminal status information;

[0014] Calculate the target delay of task offloading according to the mobile edge computing communication bandwidth, the mobile edge computing communication power, the satellite service time, the satellite switching time, the device-to-device communication bandwidth, the device-to-device communication power and the offloading ratio of the task to the neighboring node, wherein the target delay of task offloading includes the delay of task offloading to the wireless access point and the delay of task offloading to the neighboring node;

[0015] The total delay of the multi-mode terminal computing offloading task is calculated according to the target computing offloading mode set, the delay of the task offloading to the wireless access point and the delay of the task offloading to the neighboring node.

[0016] In some embodiments, determining the target resource pool and the terminal status information according to the multi-mode terminal target information includes:

[0017] According to the multi-mode terminal device information, the first unit is virtualized to obtain the communication resource pool, where the first unit includes a modem, a radio frequency unit or an antenna device, and the communication resource pool includes frequency points, communication standards, time domain resources, frequency domain resources, code domain resources or spatial domain resources supported by the multi-mode terminal;

[0018] According to the multi-mode terminal device information, the first processor is virtualized to obtain the computing resource pool, where the first processor includes an application processor or a coprocessor, the coprocessor includes a graphics processing unit, a neural network processing unit or a data processing unit, and the computing resource pool includes floating-point operations per second resources, processor computing power unit resources or instructions executed per second resources;

[0019] Extracting the computing task requirement from the local data information of the multi-mode terminal, the computing task requirement including task size, latency requirement and required computing power;

[0020] Extracting the resource status from the local data information of the multi-mode terminal, the resource status including available energy resources, available computing resources or available communication resources of the multi-mode terminal;

[0021] The communication capability information is extracted from the local data information of the multi-mode terminal, the communication capability information includes the communication mode supported by the multi-mode terminal, the communication mode includes short-range communication, cellular communication or satellite communication, the short-range communication includes wireless local area network communication, Bluetooth communication or star flash communication, the cellular communication includes 4G communication or 5G communication, and the satellite communication includes low-orbit communication or medium-orbit communication.

[0022] In some embodiments, determining the target computing offloading mode set according to the target resource pool and the terminal status information includes:

[0023] Initialize the initial calculation offloading mode set;

[0024] If the target resource pool and the terminal status information meet the business computing requirements, adding the local computing mode to the initial computing offloading mode set;

[0025] If the target resource pool and the terminal status information do not meet the business computing requirements, determining the available network;

[0026] If the available network is an infrastructure network, determining the communication mode;

[0027] If the communication mode is wireless local area network communication, adding a wireless local area network access point offloading mode to the initial calculation offloading mode set;

[0028] If the communication mode is 5G communication, adding the 5G base station offloading mode to the initial calculation offloading mode set;

[0029] If the communication mode is satellite communication, adding a satellite access point offloading mode to the initial calculation offloading mode set;

[0030] If the available network is a neighbor node network, adding a device-to-device offloading mode to the initial computation offloading mode set;

[0031] The updated initial computing offloading mode set is used as the target computing offloading mode set.

[0032] In some embodiments, the calculating the task offloading target delay according to the mobile edge computing communication bandwidth, the mobile edge computing communication power, the satellite service duration, the satellite switching duration, the device-to-device communication bandwidth, the device-to-device communication power and the offloading ratio of the task to the neighboring node includes:

[0033] Calculate a achievable data rate from the multi-mode terminal to the wireless access point according to the mobile edge computing communication bandwidth, the mobile edge computing communication power, the power spectral density of noise, and the channel gain from the multi-mode terminal to the wireless access point;

[0034] Calculate the satellite calculation result return delay according to the calculation result data size, the satellite service time and the satellite switching time;

[0035] When the communication mode is short-range communication or cellular communication, calculating the delay of offloading the task to the wireless access point according to the offloading ratio of the task to the wireless access point, the number of central processor cycles required for the multi-mode terminal to calculate the task, the central processor frequency of the wireless access point, the size of the calculation task, and the reachable data rate from the multi-mode terminal to the wireless access point;

[0036] When the communication mode is satellite communication, calculating the delay of unloading the task to the wireless access point according to the unloading ratio of the task to the wireless access point and the delay of returning the satellite calculation result;

[0037] Calculate a achievable data rate from the multimode terminal to the neighboring node according to the device-to-device communication bandwidth, the device-to-device communication power, the power spectral density of noise, and the channel gain from the multimode terminal to the neighboring node;

[0038] The delay of offloading the task to the neighbor node is calculated based on the offloading ratio of the task to the neighbor node, the number of central processing unit cycles required for the multi-mode terminal computing task, the central processing unit frequency of the neighbor node, the computing task size and the achievable data rate from the multi-mode terminal to the neighbor node.

[0039] In some embodiments, the calculating the total delay of the multi-mode terminal computing offloading task according to the target computing offloading mode set, the delay of offloading the task to the wireless access point, and the delay of offloading the task to the neighboring node includes:

[0040] Calculate the local computing delay according to the target computing offloading mode set, the offloading ratio of the task to the wireless access point, the offloading ratio of the task to the neighboring node, the number of CPU cycles required for the multi-mode terminal computing task and the CPU frequency of the multi-mode terminal;

[0041] Calculate the unloading delay of the unloading task in a single communication mode according to the local calculation delay, the delay of unloading the task to the wireless access point, and the delay of unloading the task to the neighboring node;

[0042] The total delay of the multi-mode terminal calculation offloading task is calculated according to the target calculation offloading mode set and the single communication mode offloading delay of the offloading task.

[0043] In some embodiments, the calculating the total energy consumption of the multi-mode terminal computing offloading task according to the target computing offloading mode set includes:

[0044] The energy consumption required for offloading the single communication mode of the calculation task to the wireless access point is calculated based on the chip structure capacitance parameters of the wireless access point, the number of CPU cycles required for the wireless access point computing task, the CPU frequency of the wireless access point, and the mobile edge computing communication power;

[0045] Calculate the energy consumption required for offloading the task in a single communication mode to the neighboring node according to the chip structure capacitance parameters of the neighboring node, the number of CPU cycles required for the neighboring node to calculate the task, the CPU frequency of the neighboring node and the device-to-device communication power;

[0046] Calculate the total energy consumption of unloading the task single communication mode according to the energy consumption required for unloading the task single communication mode to the wireless access point and the energy consumption required for unloading the task single communication mode to the neighboring node;

[0047] Calculate the local computing energy consumption of the offload task according to the target computing offload mode set, the offload ratio of the task offloaded to the wireless access point, the offload ratio of the task offloaded to the neighboring node, the chip structure capacitance parameter of the multi-mode terminal, the number of central processing unit cycles required for the multi-mode terminal computing task and the central processing unit frequency of the multi-mode terminal;

[0048] The total energy consumption of the multi-mode terminal computing offloading task is calculated according to the target computing offloading mode set, the total energy consumption of the task single communication mode offloading and the local computing energy consumption of the offloading task.

[0049] In some embodiments, constructing an objective function according to the multi-mode terminal calculating the total delay of the offloading task and the multi-mode terminal calculating the total energy consumption of the offloading task includes:

[0050] Constructing a first constraint according to an offloading ratio of tasks to the wireless access point, a number of CPU cycles required for the multi-mode terminal computing tasks, and available computing resources of the wireless access point;

[0051] A second constraint is constructed according to the offloading ratio of the task to the neighboring node, the number of CPU cycles required for the multi-mode terminal computing task, and the available computing resources of the neighboring node;

[0052] Constructing a third constraint according to the target computing offloading mode set, the offloading ratio of the task to the wireless access point, the offloading ratio of the task to the neighboring node, the number of CPU cycles required for the multi-mode terminal computing task, and the available computing resources of the multi-mode terminal;

[0053] Calculate the total delay of the offloading task and the maximum delay of completing the offloading task according to the multi-mode terminal, and construct a fourth constraint;

[0054] The objective function is constructed according to the first constraint, the second constraint, the third constraint, the fourth constraint, the calculated unloading delay weight coefficient, the energy consumption performance indicator weight coefficient, the total delay of the multi-mode terminal calculating the unloading task and the total energy consumption of the multi-mode terminal calculating the unloading task.

[0055] In some embodiments, the method further comprises:

[0056] Determine the uninstall mode;

[0057] According to the offloading mode, sending a collaborative offloading request message to a computing platform, the collaborative offloading request message including a task requirement;

[0058] receiving a collaborative offloading response message from the computing platform, the collaborative offloading response message including offloading confirmation information and available resource information;

[0059] Determining an offloading ratio and resource parameters according to the collaborative offloading response message;

[0060] According to the offloading ratio and the resource parameters, the offloading data is sent to the computing platform;

[0061] A computation result is received from the computation platform.

[0062] On the other hand, an embodiment of the present invention provides a multi-mode terminal adaptive calculation unloading device, including:

[0063] The first module is used to obtain multi-mode terminal target information, wherein the multi-mode terminal target information includes multi-mode terminal equipment information and multi-mode terminal local data information;

[0064] The second module is used to calculate the target computing offloading mode set and the total delay of the multi-mode terminal computing offloading task according to the multi-mode terminal target information;

[0065] The third module is used to calculate the total energy consumption of the multi-mode terminal computing offloading task according to the target computing offloading mode set;

[0066] The fourth module is used to construct an objective function according to the total delay of the offloading task calculated by the multi-mode terminal and the total energy consumption of the offloading task calculated by the multi-mode terminal;

[0067] The fifth module is used to perform adaptive computation offloading according to the objective function.

[0068] The beneficial effects of the present invention are as follows:

[0069] The embodiment of the present invention first obtains the target information of the multi-mode terminal, calculates the target computing offloading mode set and the total delay of the multi-mode terminal computing offloading task according to the multi-mode terminal target information, then calculates the total energy consumption of the multi-mode terminal computing offloading task according to the target computing offloading mode set, and then constructs the objective function according to the total delay of the multi-mode terminal computing offloading task and the total energy consumption of the multi-mode terminal computing offloading task, and finally performs adaptive computing offloading according to the objective function, so that adaptive computing offloading can be achieved through the combined optimization of delay, energy consumption and offloading mode, thereby improving efficiency and adaptability.

[0070] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the description and the drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0072] Figure 1 This is a flow chart of a multi-mode terminal adaptive computing offloading method according to an embodiment of the present invention;

[0073] Figure 2 A schematic diagram of a process for determining a set of computing offloading modes according to an embodiment of the present invention;

[0074] Figure 3 A schematic diagram of a system flow of adaptive computing offloading for multi-mode terminals according to an embodiment of the present invention;

[0075] Figure 4 The present invention is a schematic diagram of the structure of a multi-mode terminal adaptive computing offloading device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0076] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application is further described in detail below in conjunction with the accompanying drawings and examples. It should be understood that the specific embodiments described herein are only used to explain the present application and are not intended to limit the present application. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the embodiments of the present application. They are only examples of devices and methods consistent with some aspects of the embodiments of the present application as detailed in the attached claims.

[0077] It is understood that the terms "first", "second", etc. used in this application can be used to describe various concepts in this article, but unless otherwise specified, these concepts are not limited by these terms. These terms are only used to distinguish one concept from another concept. For example, without departing from the scope of the embodiment of the present application, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the words "if" and "if" as used herein can be interpreted as "at the time of" or "when" or "in response to determination".

[0078] The terms "at least one", "multiple", "each", "any", etc. used in this application, at least one includes one, two or more, multiple includes two or more, each refers to each of the corresponding multiple, and any refers to any one of the multiple.

[0079] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application and are not intended to limit this application.

[0080] Before describing the embodiments of the present application in detail, some nouns and terms involved in the embodiments of the present application are first described. The nouns and terms involved in the embodiments of the present application are subject to the following explanations.

[0081] Computing offloading: refers to offloading the computing tasks of terminal devices to edge servers or cloud servers for processing to solve problems such as insufficient processing power and limited resources of terminal devices.

[0082] Edge computing: refers to the use of an open platform that integrates network, computing, storage, and application core capabilities to provide the nearest end service on the side close to the object or data source. Its application is initiated on the edge side, resulting in faster network service response, meeting the industry's basic needs in real-time business, application intelligence, security and privacy protection, etc.

[0083] Multi-mode terminal: refers to a terminal with multiple communication modes, such as 4G, 5G, WIFI, and satellite. Different communication modes can enrich the unloading methods.

[0084] Mobile Edge Computing (MEC): is a network architecture concept that aims to provide cloud computing capabilities and IT service environments at the edge of the network.

[0085] Mobile terminal (or mobile communication terminal) refers to a computer device that can be used on the move. In a broad sense, it includes mobile phones, notebooks, tablet computers, POS machines and even car computers.

[0086] In the related technologies, computationally intensive applications such as virtual reality (VR), augmented reality (AR) and mixed reality (MR) require huge computing power support. However, due to the limitations of size, power consumption and cost, mobile terminals have limited computing resources and cannot directly implement and guarantee these application experiences. Computation offloading is an effective technical means. For example, cloud computing offloads computing tasks to data centers for computing and then returns the computing results to mobile terminals to ensure the user's service experience. In order to reduce the latency of offloading, mobile edge computing (MEC) plays an important role in computation offloading, providing high-speed computing capabilities for users at the edge of the network. Through MEC computation offloading, the edge server is closer to the user, which can provide users with a better service experience. MEC can be implemented on different network edge servers, such as the edge of the ground cellular network and the edge of the satellite network. Although MEC has many advantages, when there are a large number of users connected, MEC still faces a lot of computing pressure. Due to resource competition, the user experience will also become very poor. At the same time, due to the mobility of mobile terminals, the diversity of services and the variability of network environments, a single offloading method cannot guarantee user experience. It is necessary to expand other offloading methods, such as offloading to other terminals or satellites. In addition, the application scenarios of mobile terminals are dynamic, and the way of offloading may change, which also requires an adaptive mechanism. In order to reduce the pressure of edge cloud computing and meet the requirements of lower latency, a more effective offloading mechanism has become imminent. Mobile terminals generally have multi-mode communication capabilities such as short-range, cellular and satellite, but for computing-intensive services that are increasingly applied on a large scale, local computing power is insufficient. How to efficiently use multi-mode communication capabilities for business offloading is an urgent problem that needs to be solved.

[0087] In view of this, this embodiment adaptively determines a set of potential offloading modes based on the communication capabilities, wireless environment, and computing power environment of the multi-mode terminal, and determines the offloading ratio and resource allocation as needed to achieve multi-mode parallel offloading, thereby improving the service experience of the multi-mode terminal.

[0088] A multi-mode terminal adaptive computing unloading method provided in an embodiment of the present application relates to the field of mobile communication technology. A multi-mode terminal adaptive computing unloading method provided in an embodiment of the present application can be applied to a terminal, can also be applied to a server, and can also be software running in a terminal or a server. In some embodiments, the terminal can be a smart phone, a tablet computer, a laptop computer, a desktop computer, a smart speaker, a smart watch, and a car terminal, etc., but is not limited to this; the server side can be configured as an independent physical server, or a server cluster or distributed system composed of multiple physical servers, and can also be configured as a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms. The server can also be a node server in a blockchain network; the software can be an application that implements a multi-mode terminal adaptive computing unloading method, etc., but is not limited to the above forms.

[0089] The present application can be used in many general or special computer system environments or configurations. For example: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, etc. The present application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. The present application can also be practiced in distributed computing environments, in which tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media including storage devices.

[0090] The following is a detailed explanation of the embodiments of the present application in conjunction with the accompanying drawings:

[0091] Figure 1 is an optional flowchart of a multi-mode terminal adaptive calculation offloading method provided in an embodiment of the present application. Figure 1 The method may include but is not limited to steps S101 to S105.

[0092] Step S101: Acquire multi-mode terminal target information, where the multi-mode terminal target information includes multi-mode terminal device information and multi-mode terminal local data information;

[0093] Step S102: Calculate the target computing offloading mode set and the total delay of the multi-mode terminal computing offloading task according to the multi-mode terminal target information;

[0094] Step S103, calculating the total energy consumption of the multi-mode terminal computing offloading task according to the target computing offloading mode set;

[0095] Step S104, constructing an objective function according to the total delay of the multi-mode terminal calculation offloading task and the total energy consumption of the multi-mode terminal calculation offloading task;

[0096] Step S105: Perform adaptive computation offloading according to the objective function.

[0097] Steps S101 to S105 shown in the embodiment of the present application implement adaptive computing offloading, thereby improving efficiency and adaptability.

[0098] In step S101 of some embodiments, the multimode terminal target information includes multimode terminal device information and multimode terminal local data information. When acquiring the multimode terminal target information, the multimode terminal device information can be acquired through a device database, and the multimode terminal local data information can be acquired through a terminal configuration information library. The multimode terminal device information and the multimode terminal local data information can also be acquired through other methods, not limited thereto.

[0099] In some embodiments, in step S102, according to the multi-mode terminal target information, calculating the target computing offloading mode set and the total latency of the multi-mode terminal computing offloading task may include but is not limited to the following steps:

[0100] Step S201: Determine a target resource pool and terminal status information according to the multi-mode terminal target information, the target resource pool includes a communication resource pool and a computing resource pool, and the terminal status information includes computing task requirements, resource status, and communication capability information;

[0101] Step S202: Determine a target computing offloading mode set according to the target resource pool and terminal status information;

[0102] Step S203, calculating the target delay of task offloading according to the mobile edge computing communication bandwidth, the mobile edge computing communication power, the satellite service duration, the satellite switching duration, the device-to-device communication bandwidth, the device-to-device communication power and the offloading ratio of the task to the neighboring node, where the target delay of task offloading includes the delay of task offloading to the wireless access point and the delay of task offloading to the neighboring node;

[0103] Step S204: Calculate the total delay of the multi-mode terminal computing offloading task according to the target computing offloading mode set, the delay of offloading the task to the wireless access point, and the delay of offloading the task to the neighboring node.

[0104] In some embodiments, in step S201, determining the target resource pool and the terminal status information according to the multi-mode terminal target information may include but is not limited to the following steps:

[0105] According to the multi-mode terminal device information, the first unit is virtualized to obtain a communication resource pool, where the first unit includes a modem, a radio frequency unit or an antenna device, and the communication resource pool includes frequency points, communication standards, time domain resources, frequency domain resources, code domain resources or air domain resources supported by the multi-mode terminal;

[0106] According to the multi-mode terminal device information, the first processor is virtualized to obtain a computing resource pool, where the first processor includes an application processor or a coprocessor, the coprocessor includes a graphics processing unit, a neural network processing unit or a data processing unit, and the computing resource pool includes floating-point operations per second resources, processor computing power unit resources or instructions executed per second resources;

[0107] Extract computing task requirements from local data information of multi-mode terminals, where computing task requirements include task size, latency requirements, and required computing power;

[0108] Extracting resource status from local data information of the multi-mode terminal, where the resource status includes available energy resources, available computing resources or available communication resources of the multi-mode terminal;

[0109] The communication capability information is extracted from the local data information of the multi-mode terminal. The communication capability information includes the communication modes supported by the multi-mode terminal. The communication modes include short-range communication, cellular communication or satellite communication. Short-range communication includes wireless LAN communication, Bluetooth communication or star flash communication. Cellular communication includes 4G communication or 5G communication. Satellite communication includes low-orbit communication or medium-orbit communication.

[0110] In some embodiments, the target resource pool and the terminal status information are determined according to the multi-mode terminal target information. The target resource pool can be determined first according to the multi-mode terminal device information, wherein the target resource pool includes a communication resource pool and a computing resource pool. The first unit can be virtualized according to the multi-mode terminal device information to obtain a communication resource pool, wherein the first unit includes a modem, a radio frequency unit or an antenna device, and the communication resource pool includes the frequency point, communication standard, time domain resources, frequency domain resources, code domain resources or airspace resources supported by the multi-mode terminal. Then, according to the multi-mode terminal device information, the first processor is virtualized to obtain a computing resource pool, wherein the first processor includes an application processor or a coprocessor, and the coprocessor includes a graphics processing unit GPU (Graphics Processing Unit), a neural network processing unit NPU (Neural network Processing Unit) or a data processing unit DPU (Data Processing Unit), and the computing resource pool includes floating-point operations per second (FLOPS), processor computing power unit resources OPS (operations per second) or instructions executed per second (IPS). Then, according to the local data information of the multi-mode terminal, the terminal status information is determined, wherein the terminal status information includes computing task requirements, resource status and communication capability information. The computing task requirements can be first extracted from the local data information of the multi-mode terminal, wherein the computing task requirements include task size, latency requirements and required computing power. Then the resource status is extracted from the local data information of the multi-mode terminal, wherein the resource status includes the available energy resources, available computing resources or available communication resources (such as transmission power) of the multi-mode terminal. Then the communication capability information is extracted from the local data information of the multi-mode terminal, wherein the communication capability information includes the communication mode supported by the multi-mode terminal, and the communication mode includes short-range communication, cellular communication or satellite communication. Short-range communication includes wireless LAN communication, Bluetooth communication or star flash communication. Cellular communication includes 4G communication (4G LTE) or 5G communication (5G NR). Satellite communication includes low-orbit communication or medium-orbit communication.

[0111] In some embodiments, in step S202, determining a target computing offloading mode set according to the target resource pool and the terminal status information may include but is not limited to the following steps:

[0112] Initialize the initial calculation offloading mode set;

[0113] If the target resource pool and terminal status information meet the business computing requirements, the local computing mode is added to the initial computing offloading mode set;

[0114] If the target resource pool and terminal status information do not meet the business computing requirements, the available network is judged;

[0115] If the available network is an infrastructure network, the communication mode is determined;

[0116] If the communication mode is wireless local area network communication, adding the wireless local area network access point offloading mode to the initial calculation offloading mode set;

[0117] If the communication mode is 5G communication, the 5G base station offloading mode is added to the initial calculation offloading mode set;

[0118] If the communication mode is satellite communication, then adding the satellite access point offloading mode to the initial calculation offloading mode set;

[0119] If the available network is a neighbor node network, the device-to-device offloading mode is added to the initial computation offloading mode set;

[0120] The updated initial computation offloading mode set is used as the target computation offloading mode set.

[0121] In some embodiments, the calculation of the unloading mode set determination process is as follows Figure 2 As shown, the initial computing offloading mode set can be initialized first, and then the target resource pool and terminal status information of the multi-mode terminal are judged. If the target resource pool and terminal status information meet the business computing requirements, the local computing mode is added to the initial computing offloading mode set. The mathematical expression is: k =M k ∪{LOCAL},M k is the initial computing offloading mode set, and LOCAL is the local computing mode. If the target resource pool and terminal status information do not meet the business computing requirements, the available network of the communication mode m supported by the multi-mode terminal is judged. If an available network exists, the available network corresponding to the communication mode m is added to the initial computing offloading mode set. If the available network is an infrastructure network, that is, the multi-mode terminal can access the network through a wireless access point, the communication mode is judged. If the communication mode is wireless LAN communication (WiFi communication), the wireless LAN access point offloading mode is added to the initial computing offloading mode set; if the communication mode is 5G communication, the 5G base station offloading mode is added to the initial computing offloading mode set; if the communication mode is satellite communication, the satellite access point offloading mode is added to the initial computing offloading mode set. The mathematical expression is: M k =M k ∪{LOCAL}∪{MEC m,b}, MEC m,bis a wireless access point offloading mode, which may include a wireless LAN access point offloading mode, a 5G base station offloading mode, or a satellite access point offloading mode. If the available network is a neighbor node network, the device-to-device offloading mode is added to the initial calculation offloading mode set, and the mathematical expression is: M k =M k ∪{LOCAL}∪{MEC m,n}, MEC m,n The device-to-device offloading mode enables the task to be offloaded to the corresponding neighbor through cellular communication or short-range communication. Finally, the updated initial computing offloading mode set is used as the target computing offloading mode set.

[0122] In some embodiments, in step S203, the task offloading target delay is calculated according to the mobile edge computing communication bandwidth, the mobile edge computing communication power, the satellite service duration, the satellite switching duration, the device-to-device communication bandwidth, the device-to-device communication power and the offloading ratio of the task to the neighboring node, which may include but is not limited to the following steps:

[0123] Calculate the achievable data rate from the multi-mode terminal to the wireless access point based on the mobile edge computing communication bandwidth, the mobile edge computing communication power, the power spectrum density of the noise, and the channel gain from the multi-mode terminal to the wireless access point;

[0124] Calculate the satellite calculation result return delay based on the calculation result data size, satellite service time and satellite switching time;

[0125] When the communication mode is short-range communication or cellular communication, the latency of offloading the task to the wireless access point is calculated based on the offloading ratio of the task to the wireless access point, the number of CPU cycles required for the multi-mode terminal to calculate the task, the CPU frequency of the wireless access point, the size of the calculation task, and the achievable data rate from the multi-mode terminal to the wireless access point;

[0126] When the communication mode is satellite communication, the delay of offloading the task to the wireless access point is calculated according to the offloading ratio of the task to the wireless access point and the delay of the satellite calculation result return;

[0127] Calculate the achievable data rate from the multimode terminal to the neighboring node based on the device-to-device communication bandwidth, device-to-device communication power, power spectrum density of noise, and channel gain from the multimode terminal to the neighboring node;

[0128] The latency of offloading the task to the neighbor node is calculated based on the offloading ratio of the task to the neighbor node, the number of CPU cycles required for the multimode terminal to calculate the task, the CPU frequency of the neighbor node, the size of the calculation task and the achievable data rate from the multimode terminal to the neighbor node.

[0129] In some embodiments, the target delay of task offloading includes the delay of task offloading to the wireless access point and the delay of task offloading to the neighboring node. The reachable data rate from the multi-mode terminal to the wireless access point can be calculated based on the mobile edge computing communication bandwidth, the mobile edge computing communication power, the power spectrum density of the noise and the channel gain from the multi-mode terminal to the wireless access point, wherein the calculation formula of the reachable data rate from the multi-mode terminal to the wireless access point is: In the formula, is the achievable data rate from multimode terminal k to wireless access point b, W k,m,b The mobile edge computing communication bandwidth allocated for communication mode m, Calculate the communication power of the mobile edge for multimode terminal k in communication mode m, h k,m,b is the channel gain from multimode terminal k to wireless access point b in communication mode m, N 0 is the power spectral density of the noise. Then, according to the calculation result data size, satellite service time and satellite switching time, the satellite calculation result return delay is calculated. The calculation formula of the satellite calculation result return delay is: In the formula, The satellite calculation result return delay, D k To calculate the result data size, is the achievable data rate from multimode terminal k to satellite s, is the offloading ratio of satellite s’ tasks to wireless access point b under communication mode m, C k The number of CPU cycles required to compute tasks for multi-mode terminals, f b is the CPU frequency of wireless access point b, T b,b′ is the satellite switching duration, Length of time in service of the satellite.

[0130] When the communication mode is short-range communication or cellular communication, the return delay of the calculation result can be ignored. The delay of the task offloading to the wireless access point can be calculated based on the offloading ratio of the task to the wireless access point, the number of CPU cycles required for the multi-mode terminal calculation task, the CPU frequency of the wireless access point, the calculation task size, and the reachable data rate from the multi-mode terminal to the wireless access point. The calculation formula for the delay of the task offloading to the wireless access point is: In the formula, is the latency of offloading the task to the wireless access point, is the offloading ratio of tasks of multi-mode terminal k to wireless access point b in communication mode m, |V k | is the size of the calculation task. It is understandable that represents the offload transmission delay to the wireless access point, Represents the calculation delay of the wireless access point. When the communication mode is satellite communication, the return delay of the calculation result cannot be ignored. The delay of the task offloading to the wireless access point can be calculated based on the offloading ratio of the task to the wireless access point and the return delay of the satellite calculation result. The calculation formula for the delay of the task offloading to the wireless access point is: In the formula, is the latency of offloading the task to the wireless access point, Calculates the return latency for the satellite.

[0131] Then, the achievable data rate from the multimode terminal to the neighboring node is calculated based on the device-to-device communication bandwidth, device-to-device communication power, power spectrum density of noise, and channel gain from the multimode terminal to the neighboring node. The calculation formula for the achievable data rate from the multimode terminal to the neighboring node is: In the formula, is the achievable data rate from multimode terminal k to neighbor node n, W k,m,n The device-to-device communication bandwidth allocated for communication mode m, is the device-to-device communication power of multi-mode terminal k in communication mode m, h k,m,n is the channel gain from multimode terminal k to neighbor node n in communication mode m, N 0 is the power spectral density of the noise. Then, according to the offloading ratio of the task to the neighboring node, the number of CPU cycles required for the multi-mode terminal to calculate the task, the CPU frequency of the neighboring node, the size of the calculation task, and the achievable data rate from the multi-mode terminal to the neighboring node, the delay of the task to be offloaded to the neighboring node is calculated. The calculation formula of the delay of the task to be offloaded to the neighboring node is: In the formula, is the latency of offloading the task to the neighboring node, is the offloading ratio of tasks of multimode terminal k to neighbor node n under communication mode m, C k The number of CPU cycles required to compute tasks for multi-mode terminals, f n is the CPU frequency of neighbor node n, |V k | is the size of the calculation task. It is understandable that represents the offloading transmission delay from the node to the neighbor node, Calculates latency on behalf of neighboring nodes.

[0132] In some embodiments, in step S204, the total delay of the multi-mode terminal computing offloading task is calculated according to the target computing offloading mode set, the delay of the task offloading to the wireless access point, and the delay of the task offloading to the neighboring node, which may include but is not limited to the following steps:

[0133] Calculate the local computing delay according to the target computing offloading mode set, the offloading ratio of tasks offloading to wireless access points, the offloading ratio of tasks offloading to neighboring nodes, the number of CPU cycles required for the multi-mode terminal computing tasks, and the CPU frequency of the multi-mode terminal;

[0134] The offloading delay of the offloading task in the single communication mode is calculated according to the local calculation delay, the delay of the task offloading to the wireless access point and the delay of the task offloading to the neighboring node;

[0135] According to the target calculation offloading mode set and the offloading delay of the single communication mode of the offloading task, the total delay of the multi-mode terminal calculation offloading task is calculated.

[0136] In some embodiments, the local computing delay may be calculated based on the target computing offloading mode set, the offloading ratio of the task to the wireless access point, the offloading ratio of the task to the neighboring node, the number of CPU cycles required for the multi-mode terminal computing task, and the CPU frequency of the multi-mode terminal, wherein the calculation formula of the local computing delay is: In the formula, is the local calculation delay, M k Compute the set of offload patterns for the target, is the offloading ratio of tasks of multimode terminal k to neighbor node n under communication mode m, is the offloading ratio of tasks of multi-mode terminal k to wireless access point b under communication mode m, C k The number of CPU cycles required to compute tasks for multi-mode terminals, f k is the CPU frequency of the multi-mode terminal k. Then, according to the local calculation delay, the delay of task offloading to the wireless access point and the delay of task offloading to the neighboring node, the unloading delay of the unloading task single communication mode is calculated, where the calculation formula of the unloading delay of the unloading task single communication mode is: In the formula, is the unloading delay of the unloading task in single communication mode of multi-mode terminal k, is the latency of offloading the task to the wireless access point, is the delay of task offloading to neighboring nodes. Finally, according to the target calculation offloading mode set and the offloading delay of the single communication mode of the offloading task, the total delay of the multi-mode terminal calculation offloading task is calculated. Among them, the calculation formula of the total delay of the multi-mode terminal calculation offloading task is: In the formula, Calculate the total latency of offloaded tasks for multi-mode terminals.

[0137] In some embodiments, in step S103, according to the target computing offloading mode set, the total energy consumption of the multi-mode terminal computing offloading task is calculated, which may include but is not limited to the following steps:

[0138] The energy consumption required for offloading the single communication mode of the calculation task to the wireless access point is calculated based on the chip structure capacitance parameters of the wireless access point, the number of CPU cycles required for the wireless access point computing task, the CPU frequency of the wireless access point, and the mobile edge computing communication power;

[0139] The energy consumption required to offload the task in single communication mode to the neighboring node is calculated based on the chip structure capacitance parameters of the neighboring node, the number of CPU cycles required for the neighboring node to calculate the task, the CPU frequency of the neighboring node, and the device-to-device communication power.

[0140] According to the energy consumption required for unloading the task single communication mode to the wireless access point and the energy consumption required for unloading the task single communication mode to the neighboring node, the total energy consumption of unloading the task single communication mode is calculated;

[0141] The local computing energy consumption of the offloaded task is calculated according to the target computing offload mode set, the offload ratio of the task offloaded to the wireless access point, the offload ratio of the task offloaded to the neighboring node, the chip structure capacitance parameters of the multi-mode terminal, the number of CPU cycles required for the computing task of the multi-mode terminal and the CPU frequency of the multi-mode terminal;

[0142] The total energy consumption of the multi-mode terminal computing offloading task is calculated based on the target computing offloading mode set, the total energy consumption of the task single communication mode offloading and the local computing energy consumption of the offloading task.

[0143] In some embodiments, the energy consumption required for unloading the task single communication mode to the wireless access point can be calculated based on the chip structure capacitance parameters of the wireless access point, the number of CPU cycles required for the wireless access point computing task, the CPU frequency of the wireless access point, and the mobile edge computing communication power. The calculation formula for the energy consumption required for unloading the task single communication mode to the wireless access point is: In the formula, The energy consumption required for offloading the single-task communication mode to the wireless access point, K v is the chip structure capacitance parameter of the wireless access point, is the offloading ratio of tasks of multi-mode terminal k to wireless access point b under communication mode m, C b The number of CPU cycles required for the wireless access point computing task, f v is the CPU frequency of wireless access point b, Calculate the communication power for the mobile edge of multimode terminal k in communication mode m, is the achievable data rate from multimode terminal k to wireless access point b, |V k | is the size of the calculation task. It is understandable that represents the computing energy consumption of the wireless access point, Represents the transmission energy consumption of the wireless access point.

[0144] Then, according to the chip structure capacitance parameters of the neighboring node, the number of CPU cycles required for the neighboring node to calculate the task, the CPU frequency of the neighboring node, and the device-to-device communication power, the energy consumption required for unloading the task in single communication mode to the neighboring node is calculated. The calculation formula for the energy consumption required for unloading the task in single communication mode to the neighboring node is: In the formula, is the energy consumption required to offload the task to neighbor nodes in single communication mode, K n is the chip structure capacitance parameter of the neighbor node, is the offloading ratio of tasks of multimode terminal k to neighbor node n under communication mode m, C n The number of CPU cycles required to compute tasks for neighbor nodes, f n is the CPU frequency of neighbor node n, is the device-to-device communication power of multi-mode terminal k in communication mode m, is the achievable data rate from multimode terminal k to neighbor node n. It can be understood that represents the computational energy consumption of neighboring nodes, Represents the transmission energy consumption of neighboring nodes.

[0145] Then, according to the energy consumption required for unloading the task single communication mode to the wireless access point and the energy consumption required for unloading the task single communication mode to the neighboring node, the total energy consumption of unloading the task single communication mode is calculated. The calculation formula of the total energy consumption of unloading the task single communication mode is: In the formula, Unloading total energy consumption for task single communication mode, The energy consumption required to offload the single-task communication mode to the wireless access point, The energy consumption required for offloading the task to the neighboring node in single communication mode. The local computing energy consumption of the offloading task is calculated based on the target computing offloading mode set, the offloading ratio of the task to the wireless access point, the offloading ratio of the task to the neighboring node, the chip structure capacitance parameters of the multi-mode terminal, the number of CPU cycles required for the multi-mode terminal computing task, and the CPU frequency of the multi-mode terminal. The calculation formula for the local computing energy consumption of the offloading task is: In the formula, To offload the task local computing energy consumption, M k Calculate the set of offloading patterns for the target, K k is the chip structure capacitance parameter of the multi-mode terminal, C k The number of CPU cycles required to compute tasks for multi-mode terminals, f kis the CPU frequency of the multi-mode terminal k. Finally, the total energy consumption of the multi-mode terminal computing offload task is calculated according to the target computing offload mode set, the total energy consumption of the task single communication mode offload and the local computing energy consumption of the offload task. The calculation formula of the total energy consumption of the multi-mode terminal computing offload task is: In the formula, Calculate the total energy consumption of offload tasks for multi-mode terminals.

[0146] In some embodiments, in step S104, constructing an objective function according to the total delay of the multi-mode terminal calculating the offloading task and the total energy consumption of the multi-mode terminal calculating the offloading task may include but is not limited to the following steps:

[0147] Constructing a first constraint according to an offloading ratio of tasks to the wireless access point, a number of CPU cycles required for the multi-mode terminal computing tasks, and available computing resources of the wireless access point;

[0148] A second constraint is constructed according to the offloading ratio of the task to the neighboring node, the number of CPU cycles required for the multi-mode terminal computing task, and the available computing resources of the neighboring node;

[0149] A third constraint is constructed according to a target computing offloading mode set, an offloading ratio of tasks offloading to wireless access points, an offloading ratio of tasks offloading to neighboring nodes, a number of CPU cycles required for computing tasks of the multi-mode terminal, and available computing resources of the multi-mode terminal;

[0150] Calculate the total delay of the offloading task and the maximum delay of the offloading task completion according to the multi-mode terminal, and construct the fourth constraint;

[0151] An objective function is constructed according to the first constraint, the second constraint, the third constraint, the fourth constraint, the calculation offloading delay weight coefficient, the energy consumption performance index weight coefficient, the multi-mode terminal calculation offloading task total delay and the multi-mode terminal calculation offloading task total energy consumption.

[0152] In some embodiments, a first constraint may be constructed based on the offloading ratio of tasks to wireless access points, the number of CPU cycles required for multi-mode terminal computing tasks, and the available computing resources of wireless access points, wherein the expression of the first constraint is: In the formula, is the offloading ratio of tasks to wireless access points, C k The number of CPU cycles required to compute tasks for multi-mode terminals, F b,abv is the available computing resources of the wireless access point. Then, according to the offloading ratio of the task to the neighboring node, the number of CPU cycles required for the multi-mode terminal computing task and the available computing resources of the neighboring node, the second constraint is constructed, where the expression of the second constraint is: In the formula, is the ratio of tasks offloaded to neighbor nodes, F n,abv is the available computing resources of the neighboring nodes. The third constraint is constructed according to the target computing offloading mode set, the offloading ratio of tasks to wireless access points, the offloading ratio of tasks to neighboring nodes, the number of CPU cycles required for the multi-mode terminal computing tasks, and the available computing resources of the multi-mode terminal, wherein the expression of the third constraint is: Where M k Calculate the set of offloading patterns for the target, F k,abv is the available computing resource of the multi-mode terminal. According to the total delay of the offloading task and the maximum delay of the offloading task completion calculated by the multi-mode terminal, a fourth constraint is constructed, wherein the expression of the fourth constraint is: In the formula, Calculate the total delay of the offload task for the multi-mode terminal, t k,max The maximum delay for completing the offloading task. Finally, according to the first constraint, the second constraint, the third constraint, the fourth constraint, the calculated offloading delay weight coefficient, the energy consumption performance index weight coefficient, the multi-mode terminal calculated offloading task total delay and the multi-mode terminal calculated offloading task total energy consumption, the objective function is constructed, where the expression of the objective function is: In the formula, Calculate the total energy consumption of the offloading task for the multi-mode terminal, α is the weight coefficient for calculating the offloading delay, β is the weight coefficient of the energy consumption performance index, n * is the optimal neighbor node, b * For the best wireless access point, For the optimal ratio of offloading to wireless access points, is the optimal ratio of offloading to neighboring nodes, For optimal wireless access point bandwidth allocation, is the optimal neighbor node bandwidth allocation, is the optimal wireless access point power, is the optimal neighbor node power.

[0153] In some embodiments, in step S105, adaptive computing offloading is performed according to the objective function. When only the local computing mode exists in the target computing offloading mode set, the task performed locally does not require a collaborative node, and the computing resources are determined by the multi-mode terminal itself. For tasks offloaded to an available network, the multi-mode terminal needs to select a suitable collaborative offloading mode to collaboratively complete the computing task. The collaborative offloading mode of task offloading includes the mobile edge computing communication mode MEC and the device-to-device communication mode D2D available to the multi-mode terminal, the collaborative node includes the wireless access point and neighbor node of the available communication mode, and the wireless resources include the power and spectrum resources of the offloading transmission. Exemplarily, the offloading ratio, power and spectrum of each mode can be optimized by minimizing the delay and power consumption of the offloading. The offloading parameters include the service offloading ratio, the participating collaborative nodes and the wireless resources. The multi-mode terminal determines the offloading parameters based on the offloading benefit. Using the objective function, with the goal of maximizing the offloading benefit, the optimal parameter combination is obtained, and the offloading ratio and resource allocation are determined according to the parameter combination, and the adaptive computing offloading is performed.

[0154] In some embodiments, the method further comprises:

[0155] Determine the uninstall mode;

[0156] According to the offloading mode, a collaborative offloading request message is sent to the computing platform, where the collaborative offloading request message includes task requirements;

[0157] receiving a collaborative offloading response message from the computing platform, the collaborative offloading response message including offloading confirmation information and available resource information;

[0158] Determine the offloading ratio and resource parameters according to the collaborative offloading response message;

[0159] According to the offloading ratio and resource parameters, the offloading data is sent to the computing platform;

[0160] Receive computation results from the computation platform.

[0161] In some embodiments, the offloading mode may be determined first, and according to the offloading mode, a collaborative offloading request message may be sent to the computing platform, wherein the collaborative offloading request message includes the task requirements. Then, a collaborative offloading response message may be received from the computing platform, wherein the collaborative offloading response message includes the offloading confirmation information and the available resource information, and the available resource information may include the available energy resources, computing resources, and communication resources. Then, according to the collaborative offloading response message, the offloading ratio and resource parameters may be determined, and according to the offloading ratio and resource parameters, the offloading data may be sent to the computing platform, and finally, the computing results may be received from the computing platform.

[0162] In some embodiments, the system process of adaptive computing offloading for multi-mode terminals is as follows: Figure 3As shown, the multi-mode terminal determines a potential set of computing offloading modes based on computing task requirements, local resource status, and communication capabilities. When collaborative offloading is required, the multi-mode terminal sends a collaborative offloading request message to the available network, which carries the offloading task requirements, and then the available network returns a collaborative offloading response message, which carries its available resource information, wherein the available resource information includes available energy, computing, and communication resources. The multi-mode terminal determines the offloading ratio and resource parameters based on the confirmation information and available resource information of the available network. Based on the offloading ratio and resource parameters, the sending of offloading data and the receiving of calculation results are completed.

[0163] The beneficial effects of implementing the embodiments of the present invention include: the embodiments of the present invention first obtain the target information of the multi-mode terminal, calculate the target computing offloading mode set and the total delay of the multi-mode terminal computing offloading task according to the multi-mode terminal target information, then calculate the total energy consumption of the multi-mode terminal computing offloading task according to the target computing offloading mode set, and then construct the objective function according to the total delay of the multi-mode terminal computing offloading task and the total energy consumption of the multi-mode terminal computing offloading task, and finally perform adaptive computing offloading according to the objective function, so that adaptive computing offloading can be achieved through the combined optimization of delay, energy consumption and offloading mode, thereby improving efficiency and adaptability.

[0164] like Figure 4 As shown, the embodiment of the present invention also provides a multi-mode terminal adaptive calculation unloading device, including:

[0165] On the other hand, an embodiment of the present invention provides a multi-mode terminal adaptive calculation unloading device, including:

[0166] The first module 801 is used to obtain multi-mode terminal target information, where the multi-mode terminal target information includes multi-mode terminal device information and multi-mode terminal local data information;

[0167] The second module 802 is used to calculate the target computing offloading mode set and the total delay of the multi-mode terminal computing offloading task according to the multi-mode terminal target information;

[0168] The third module 803 is used to calculate the total energy consumption of the multi-mode terminal computing offloading task according to the target computing offloading mode set;

[0169] The fourth module 804 is used to construct an objective function according to the total delay of the multi-mode terminal calculation offloading task and the total energy consumption of the multi-mode terminal calculation offloading task;

[0170] The fifth module 805 is used to perform adaptive computation offloading according to the objective function.

[0171] The contents of the above method embodiments are all applicable to the present device embodiments. The functions specifically implemented by the present device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0172] The preferred embodiments of the present invention are described above with reference to the accompanying drawings, but the scope of the rights of the present invention is not limited thereto. Any modification, equivalent substitution and improvement made by a person skilled in the art without departing from the scope and essence of the present invention should be within the scope of the rights of the present invention.

Claims

1. A multi-mode terminal adaptive computing offloading method, characterized in that: The following steps are involved: Acquire multimode terminal target information, wherein the multimode terminal target information includes multimode terminal device information and multimode terminal local data information; Calculate the target computing offloading mode set and the total delay of the multi-mode terminal computing offloading task according to the multi-mode terminal target information; Calculate the total energy consumption of the multi-mode terminal computing offloading task according to the target computing offloading mode set; Constructing an objective function according to the total delay of the offloading task calculated by the multi-mode terminal and the total energy consumption of the offloading task calculated by the multi-mode terminal; According to the objective function, adaptive computation offloading is performed.

2. The method according to claim 1, characterized in that: The calculating, according to the multi-mode terminal target information, a target calculation offloading mode set and a total delay of the multi-mode terminal calculation offloading task includes: Determine a target resource pool and terminal status information according to the multimode terminal target information, wherein the target resource pool includes a communication resource pool and a computing resource pool, and the terminal status information includes computing task requirements, resource status, and communication capability information; Determining the target computing offloading mode set according to the target resource pool and the terminal status information; Calculate the target delay of task offloading according to the mobile edge computing communication bandwidth, the mobile edge computing communication power, the satellite service time, the satellite switching time, the device-to-device communication bandwidth, the device-to-device communication power and the offloading ratio of the task to the neighboring node, wherein the target delay of task offloading includes the delay of task offloading to the wireless access point and the delay of task offloading to the neighboring node; The total delay of the multi-mode terminal computing offloading task is calculated according to the target computing offloading mode set, the delay of the task offloading to the wireless access point and the delay of the task offloading to the neighboring node.

3. The method according to claim 2, characterized in that The step of determining a target resource pool and terminal status information according to the multi-mode terminal target information includes: According to the multi-mode terminal device information, the first unit is virtualized to obtain the communication resource pool, where the first unit includes a modem, a radio frequency unit or an antenna device, and the communication resource pool includes frequency points, communication standards, time domain resources, frequency domain resources, code domain resources or spatial domain resources supported by the multi-mode terminal; According to the multi-mode terminal device information, the first processor is virtualized to obtain the computing resource pool, where the first processor includes an application processor or a coprocessor, the coprocessor includes a graphics processing unit, a neural network processing unit or a data processing unit, and the computing resource pool includes floating-point operations per second resources, processor computing power unit resources or instructions executed per second resources; Extracting the computing task requirement from the local data information of the multi-mode terminal, the computing task requirement including task size, latency requirement and required computing power; Extracting the resource status from the local data information of the multi-mode terminal, the resource status including available energy resources, available computing resources or available communication resources of the multi-mode terminal; The communication capability information is extracted from the local data information of the multi-mode terminal, the communication capability information includes the communication mode supported by the multi-mode terminal, the communication mode includes short-range communication, cellular communication or satellite communication, the short-range communication includes wireless local area network communication, Bluetooth communication or star flash communication, the cellular communication includes 4G communication or 5G communication, and the satellite communication includes low-orbit communication or medium-orbit communication.

4. The method according to claim 2, characterized in that: The determining the target computing offloading mode set according to the target resource pool and the terminal status information includes: Initialize the initial calculation offloading mode set; If the target resource pool and the terminal status information meet the business computing requirements, adding the local computing mode to the initial computing offloading mode set; If the target resource pool and the terminal status information do not meet the business computing requirements, determining the available network; If the available network is an infrastructure network, determining the communication mode; If the communication mode is wireless local area network communication, adding a wireless local area network access point offloading mode to the initial calculation offloading mode set; If the communication mode is 5G communication, adding the 5G base station offloading mode to the initial calculation offloading mode set; If the communication mode is satellite communication, adding a satellite access point offloading mode to the initial calculation offloading mode set; If the available network is a neighbor node network, adding a device-to-device offloading mode to the initial computation offloading mode set; The updated initial computing offloading mode set is used as the target computing offloading mode set.

5. The method according to claim 2, characterized in that: The calculating of the task offloading target delay according to the mobile edge computing communication bandwidth, the mobile edge computing communication power, the satellite service duration, the satellite switching duration, the device-to-device communication bandwidth, the device-to-device communication power and the offloading ratio of the task to the neighboring node includes: Calculate a achievable data rate from the multi-mode terminal to the wireless access point according to the mobile edge computing communication bandwidth, the mobile edge computing communication power, the power spectral density of noise, and the channel gain from the multi-mode terminal to the wireless access point; Calculate the satellite calculation result return delay according to the calculation result data size, the satellite service time and the satellite switching time; When the communication mode is short-range communication or cellular communication, calculating the delay of offloading the task to the wireless access point according to the offloading ratio of the task to the wireless access point, the number of central processor cycles required for the multi-mode terminal to calculate the task, the central processor frequency of the wireless access point, the size of the calculation task, and the reachable data rate from the multi-mode terminal to the wireless access point; When the communication mode is satellite communication, calculating the delay of unloading the task to the wireless access point according to the unloading ratio of the task to the wireless access point and the delay of returning the satellite calculation result; Calculate a achievable data rate from the multimode terminal to the neighboring node according to the device-to-device communication bandwidth, the device-to-device communication power, the power spectral density of noise, and the channel gain from the multimode terminal to the neighboring node; The delay of offloading the task to the neighbor node is calculated based on the offloading ratio of the task to the neighbor node, the number of central processing unit cycles required for the multi-mode terminal computing task, the central processing unit frequency of the neighbor node, the computing task size and the achievable data rate from the multi-mode terminal to the neighbor node.

6. The method according to claim 2, characterized in that The calculating the total delay of the multi-mode terminal computing offloading task according to the target computing offloading mode set, the delay of the task offloading to the wireless access point, and the delay of the task offloading to the neighboring node includes: Calculate the local computing delay according to the target computing offloading mode set, the offloading ratio of the task to the wireless access point, the offloading ratio of the task to the neighboring node, the number of CPU cycles required for the multi-mode terminal computing task and the CPU frequency of the multi-mode terminal; Calculate the unloading delay of the unloading task in a single communication mode according to the local calculation delay, the delay of unloading the task to the wireless access point, and the delay of unloading the task to the neighboring node; The total delay of the multi-mode terminal calculation offloading task is calculated according to the target calculation offloading mode set and the single communication mode offloading delay of the offloading task.

7. The method according to claim 1, characterized in that The calculating the total energy consumption of the multi-mode terminal computing offloading task according to the target computing offloading mode set includes: The energy consumption required for offloading the single communication mode of the calculation task to the wireless access point is calculated based on the chip structure capacitance parameters of the wireless access point, the number of CPU cycles required for the wireless access point computing task, the CPU frequency of the wireless access point, and the mobile edge computing communication power; Calculate the energy consumption required for offloading the task in a single communication mode to the neighboring node according to the chip structure capacitance parameters of the neighboring node, the number of CPU cycles required for the neighboring node to calculate the task, the CPU frequency of the neighboring node and the device-to-device communication power; Calculate the total energy consumption of unloading the task single communication mode according to the energy consumption required for unloading the task single communication mode to the wireless access point and the energy consumption required for unloading the task single communication mode to the neighboring node; Calculate the local computing energy consumption of the offload task according to the target computing offload mode set, the offload ratio of the task offloaded to the wireless access point, the offload ratio of the task offloaded to the neighboring node, the chip structure capacitance parameter of the multi-mode terminal, the number of central processing unit cycles required for the multi-mode terminal computing task and the central processing unit frequency of the multi-mode terminal; The total energy consumption of the multi-mode terminal computing offloading task is calculated according to the target computing offloading mode set, the total energy consumption of the task single communication mode offloading and the local computing energy consumption of the offloading task.

8. The method according to claim 1, characterized in that: The constructing of an objective function according to the total delay of the offloading task calculated by the multi-mode terminal and the total energy consumption of the offloading task calculated by the multi-mode terminal comprises: Constructing a first constraint according to an offloading ratio of tasks to the wireless access point, a number of CPU cycles required for the multi-mode terminal computing tasks, and available computing resources of the wireless access point; Constructing a second constraint according to the offloading ratio of the task to the neighboring node, the number of CPU cycles required for the multi-mode terminal computing task, and the available computing resources of the neighboring node; Constructing a third constraint according to the target computing offloading mode set, the offloading ratio of the task to the wireless access point, the offloading ratio of the task to the neighboring node, the number of CPU cycles required for the multi-mode terminal computing task, and the available computing resources of the multi-mode terminal; Calculate the total delay of the offloading task and the maximum delay of completing the offloading task according to the multi-mode terminal, and construct a fourth constraint; The objective function is constructed according to the first constraint, the second constraint, the third constraint, the fourth constraint, the calculated unloading delay weight coefficient, the energy consumption performance indicator weight coefficient, the total delay of the multi-mode terminal calculating the unloading task and the total energy consumption of the multi-mode terminal calculating the unloading task.

9. The method according to claim 1, characterized in that: The method further comprises: Determine the uninstall mode; According to the offloading mode, sending a collaborative offloading request message to a computing platform, the collaborative offloading request message including a task requirement; receiving a collaborative offloading response message from the computing platform, the collaborative offloading response message including offloading confirmation information and available resource information; Determining an offloading ratio and resource parameters according to the collaborative offloading response message; According to the offloading ratio and the resource parameters, the offloading data is sent to the computing platform; A computation result is received from the computation platform.

10. A multi-mode terminal adaptive computing unloading device, characterized in that: include: The first module is used to obtain multi-mode terminal target information, wherein the multi-mode terminal target information includes multi-mode terminal equipment information and multi-mode terminal local data information; The second module is used to calculate the target computing offloading mode set and the total delay of the multi-mode terminal computing offloading task according to the multi-mode terminal target information; The third module is used to calculate the total energy consumption of the multi-mode terminal computing offloading task according to the target computing offloading mode set; The fourth module is used to construct an objective function according to the total delay of the offloading task calculated by the multi-mode terminal and the total energy consumption of the offloading task calculated by the multi-mode terminal; The fifth module is used to perform adaptive computation offloading according to the objective function.

Images