Computation offloading optimization method and system based on user association

Abstract

The present invention relates to the field of computer offloading technology, and more particularly to a method and system for optimizing computing offloading based on user association. The method comprises: the user terminal receives a user task request and selects an optimal base station associated with the user terminal from at least one micro base station covering the user terminal according to the task request; the optimal base station converts the non-smooth problem of minimizing system delay and energy consumption into a nonlinear multivariate function minimum problem using a smooth approximation function; the optimal base station solves the nonlinear multivariate function minimum problem using a particle swarm algorithm to obtain an optimal task offloading decision and the optimal base station allocation resource size; the optimal base station returns the optimal task offloading decision and the optimal base station allocation resource size to the corresponding user terminal. The present invention provides a method for improving computing offloading optimization efficiency and task processing efficiency while reducing system delay and energy consumption.

Description

Technical Field

[0001] The present invention relates to the technical field of computer offloading, and in particular to a method and system for optimizing computing offloading based on user association. Background Art

[0002] With the development of mobile communication technology and the popularization of smart terminals, a variety of emerging applications such as augmented reality, virtual reality, and autonomous driving are constantly emerging, and users have higher requirements for service quality and quality of experience. It is expected that by 2030, mobile data traffic will experience explosive growth, with the number of mobile terminals worldwide approaching 100 billion, and possibly reaching 20 billion in China. Running these large-scale data applications on terminal devices requires extensive computing and storage resources, as well as high energy consumption. However, the computing power, storage resources, and battery power of mobile devices are often limited and cannot meet these demands.

[0003] In actual environments, we often face complex and heterogeneous edge networks with various access methods. On the user side, both the type and number of tasks are not single, and different types of tasks have different requirements for computing power, data transmission capabilities, and latency, resulting in different system energy consumption and system latency. Summary of the Invention

[0004] In view of the shortcomings of the prior art described above, the present invention provides a method and system for optimizing computation offloading based on user association, which can ensure optimization of computation offloading problems while reducing system latency and energy consumption.

[0005] To achieve the above-mentioned purpose and other related purposes, the present invention provides a user-association-based computing offloading optimization method, comprising the following steps: applied to a multi-layer heterogeneous network, the multi-layer heterogeneous network including a plurality of user terminals, a plurality of micro base stations and a macro base station; the user-association-based computing offloading optimization method comprises: the user terminal receives a user task request, and selects the best base station associated with the user terminal according to the task request; the best base station uses a smooth approximation function to convert the non-smooth problem of minimizing system delay and energy consumption into a nonlinear multivariate function minimum problem; the best base station uses a particle swarm algorithm to solve the nonlinear multivariate function minimum problem to obtain the best task offloading decision and the best base station allocation resource size; the best base station returns the best task offloading decision and the best base station allocation resource size to the corresponding user terminal; the user terminal executes the task offloading decision returned by the best base station.

[0006] In an optional embodiment of the present invention, the user terminal receives a user task request, and selects the best base station associated with the user terminal according to the task request, including: a task set, wherein the task set is represented as:

[0007]

[0008] Among them, d n Indicates the size of the input calculation data, c n Indicates the computing resource requirements required for unit task input size, Indicates the maximum tolerable delay required to complete the task.

[0009] In an optional embodiment of the present invention, the user terminal receives a user task request, and the step of selecting the best base station associated with the user terminal according to the task request includes: the user terminal receives the user task request, and selects the best base station associated with the user terminal from at least one of the micro base stations covering the user terminal according to the task request; the user terminal calculates the channel gain between the user terminal and each of the micro base stations covering the user terminal and the requested load of each of the micro base stations covering the user terminal; the user terminal calculates the association index between the user terminal and each of the micro base stations covering the user terminal according to the channel gain between the user terminal and each of the micro base stations covering the user terminal and the requested load of each of the micro base stations covering the user terminal; the user terminal selects the micro base station with the largest association index as the best base station associated with the user terminal.

[0010] In an optional embodiment of the present invention, the user terminal calculates the association index between the user terminal and each of the micro base stations covering the user terminal according to the channel gain between the user terminal and each of the micro base stations covering the user terminal and the requested load of each of the micro base stations covering the user terminal, and is implemented by the following formula:

[0011]

[0012] Where: b n,k represents the association index between the user terminal and each micro base station under overlapping coverage, h n,k represents the channel gain between base station k and user n, load k represents the request load of base station k.

[0013] In an optional embodiment of the present invention, the optimal base station uses a smooth approximation function to convert the non-smooth problem of minimizing system delay and energy consumption into a nonlinear multivariate function minimum problem, which is achieved by the following formula:

[0014]

[0015] Among them, X represents the task offloading decision, F represents the size of the base station allocated resources, Indicates the local computing latency, represents the user terminal under the best base station, Indicates the delay that the user terminal spends on the micro base station, represents the calculation delay of the macro base station, E n represents energy consumption, μ represents the smoothing parameter, and λ represents the weight factor of delay and energy consumption.

[0016] In an optional embodiment of the present invention, the optimal base station uses a particle swarm algorithm to solve the nonlinear multivariate function minimum problem to obtain the optimal task offloading decision and the optimal base station resource allocation size. The steps include: the base station initializes the nonlinear multivariate function minimum problem into a number of particles; the base station searches for the optimal individual particle and the optimal particle swarm through a fitness function during iteration of the particles, and updates the positions and velocities of the particles; the base station obtains the optimal task offloading decision and the optimal base station resource allocation size based on the optimal individual particle and the optimal particle swarm.

[0017] In an optional embodiment of the present invention, the base station searches for the optimal individual particle and the optimal particle swarm through a fitness function during iteration of the plurality of particles, which is achieved by the following formula:

[0018]

[0019] Among them, Fit represents fitness, η represents penalty function, t n represents the delay, λ represents the weight factor of delay and energy consumption, Indicates the calculation delay of the macro base station. represents the delay that user n spends on base station m, E n Indicates energy consumption.

[0020] In an optional embodiment of the present invention, the updating of the positions and velocities of the particles is achieved by the following formula:

[0021] V k =W·V k +C1·R1·(E best -M k )+C2·R2·(G best -M k )

[0022] Among them, M K represents the kth decision-making method, V K represents the speed of the kth particle, C1, C2 are acceleration constants, W is the inertia factor, R1, R2 are random numbers in [0,1], G best Indicates the global optimum found by the entire particle search.

[0023] In an optional embodiment of the present invention, the macro base station further comprises: solving the linear programming problem through an optimization toolbox to obtain the optimal task offloading decision.

[0024] The present invention also includes a user-association-based computing offloading optimization system. In an optional embodiment of the present invention, the user terminal is used to receive user task requests and select the best base station associated with the user terminal according to the task request; the micro base station is used to obtain the best task decision and the resource allocation size of the best base station through a particle swarm algorithm; the macro base station is used to solve the linear programming problem through an optimization toolbox to obtain the best task offloading decision.

[0025] The beneficial effects of the present invention are as follows: on the basis of adopting a multi-layer network, a smooth approximation function and a particle swarm algorithm are added to ensure that on the basis of reducing system delay and energy consumption, the problem of reducing system delay and energy consumption is converted into a user terminal association problem of multiple micro base stations and a computational offloading problem of a single micro base station. In this way, the user terminal can be assigned to the micro base station with the largest association index value as the destination of the offloading task, and the best offloading decision and the best allocated resource size are obtained at the same time to ensure computational offloading optimization.

[0026] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] The accompanying drawings are incorporated into and constitute a part of the specification, illustrate embodiments consistent with the present invention, and together with the specification, are used to explain the principles of the present invention. Obviously, the drawings described below are only some embodiments of the present invention, and it is clear that a person skilled in the art can derive other drawings based on these drawings without inventive effort. In the drawings:

[0028] Figure 1 Schematic diagram of an application scenario of a user association-based computing offloading optimization method provided by an embodiment of the present invention;

[0029] Figure 2 is a flowchart of a method for optimizing computation offloading based on user association provided by an embodiment of the present invention;

[0030] Figure 3 is a flowchart of selecting the best base station associated with the user terminal provided by an embodiment of the present invention;

[0031] Figure 4 A flowchart of obtaining an optimal task offloading decision and the optimal base station allocation resource size provided by an embodiment of the present invention;

[0032] Figure 5is a table of system simulation parameters provided by an embodiment of the present invention;

[0033] Figure 6 This is a specific block diagram of a comparison of system overheads as the number of users varies, as provided by an embodiment of the present invention;

[0034] Figure 7 This is a specific block diagram showing a comparison of system overheads as the number of servers varies, as provided by an embodiment of the present invention;

[0035] Figure 8 This is a specific block diagram of a system overhead comparison for computing task size changes provided by an embodiment of the present invention;

[0036] Figure 9 This is a specific block diagram of system performance under different weights provided by an embodiment of the present invention;

[0037] Figure 10 This is a specific block diagram of the terminal uninstall ratio provided by an embodiment of the present invention;

[0038] Figure 11 This is a specific block diagram of the system overhead as the number of iterations changes, as provided by an embodiment of the present invention;

[0039] Figure 12 This is a specific block diagram of the position of the kth particle provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0040] The following describes the embodiments of the present invention with reference to the accompanying drawings and preferred embodiments. Those skilled in the art will readily appreciate the other advantages and benefits of the present invention from the disclosure herein. The present invention may also be implemented or applied through various other specific embodiments, and the various details in this specification may be modified or altered based on different viewpoints and applications without departing from the spirit of the present invention. It should be understood that the preferred embodiments are intended only to illustrate the present invention and are not intended to limit the scope of protection of the present invention.

[0041] It should be noted that the illustrations provided in the following embodiments are merely schematic illustrations of the basic concept of the present invention. Therefore, the illustrations only show components related to the present invention and are not drawn according to the number, shape, and size of components in actual implementation. In actual implementation, the type, quantity, and proportion of each component may be changed arbitrarily, and the component layout may also be more complex.

[0042] In the following description, numerous details are discussed to provide a more thorough explanation of the embodiments of the present invention. However, it will be apparent to those skilled in the art that the embodiments of the present invention may be practiced without these specific details. In other embodiments, well-known structures and devices are shown in block diagram form rather than in detail to avoid obscuring the embodiments of the present invention.

[0043] A heterogeneous network is a type of network composed of computers, network devices, and systems produced by different manufacturers, often running on different protocols to support different functions or applications. Heterogeneity refers to two or more wireless communication systems using different access technologies, or using the same wireless access technology but belonging to different wireless operators. Leveraging the various existing wireless communication systems and integrating them to complement each other's strengths is an effective means of meeting future mobile communication service needs and leveraging the advantages of each system. Because existing wireless access systems overlap in many areas, these overlapping wireless access systems can be intelligently combined. By utilizing the intelligent access methods of multi-mode terminals, multiple different types of networks can be combined to provide users with wireless access anytime, anywhere.

[0044] In other application scenarios, the offloading optimization of computing is configured according to actual conditions, and the embodiments of the present invention are not limited thereto.

[0045] Figure 1 This is a schematic diagram of an application scenario of a user-association-based computational offloading optimization method provided in an embodiment of the present invention. When a user terminal receives a user task request, the user terminal selects a base station with the largest index value associated with the user terminal as the optimal base station based on the user task request. After the optimal base station is selected, the optimal base station uses a smooth approximation function to convert the non-smooth problem of minimizing system delay and energy consumption into a nonlinear multivariate function minimum problem. The optimal base station then uses a particle swarm algorithm to obtain the optimal task offloading decision and the optimal base station allocation resource size.

[0046] Figure 2 This is a flowchart of a method for optimizing computing offloading based on user association provided by an embodiment of the present invention. It should be noted that the present invention is based on user association and heterogeneous networks. The heterogeneous network includes several user terminals, several micro base stations and one macro base station. Since different servers have different computing capabilities and can serve multiple users, and different users may be covered by multiple base stations, users need to select the best base station from the covered base stations for festivals to ensure that they have the destination for unloading. This method can be applied to Figure 1The implementation environment shown is a schematic diagram. It should be understood that the method can also be applied to other exemplary implementation environments and specifically executed by devices in other implementation environments. This embodiment does not limit the implementation environment to which the method is applicable.

[0047] like Figure 2 As shown, the user association-based calculation offloading optimization method of this embodiment includes:

[0048] Step S21: The user terminal receives a user task request and selects the best base station associated with the user terminal from at least one micro base station covering the user terminal according to the task request. It should be noted that the user task request includes a task set, which is represented by:

[0049] Among them, d n Indicates the size of the input calculation data, c n Indicates the computing resource requirements required for unit task input size, Indicates the maximum tolerable delay required to complete the task.

[0050] Step S22: The optimal base station uses a smooth approximation function to convert the non-smooth problem of minimizing system delay and energy consumption into a nonlinear multivariate function minimum problem. It should be noted that this embodiment uses the Matlab simulation platform to test the impact of various indicators on system delay and energy consumption. Figure 5 Summarizes the parameter values ​​used during the simulation.

[0051] like Figure 6-9 The impact of each indicator on system latency and energy consumption is shown. The comparison objects adopted by the present invention include GA, which uses a binary-coded genetic algorithm to solve the adaptive offloading decision problem, PROMOT, which is an offloading mechanism based on probability priority, and obtains the optimal offloading ratio and transmission power through the genetic algorithm. The probability here is calculated based on the prior information of past offloading feedback, local calculation only (local only in the figure) means that all terminals only perform all tasks locally, partial offloading scheme (partial in the figure) means that all terminals offload to the nearest SBS to complete the task, and macro base station only (MBS in the figure) calculates all terminals and offloads all tasks to the macro base station (MBS in the figure).

[0052] In a preferred embodiment, the system latency and energy consumption of local computing are calculated by the following formula:

[0053]

[0054]

[0055] in, Indicates the local computing latency, represents the local computing energy consumption, represents the probability of MD processing, c n Denotes the computational resource requirement per unit task input size, d n Indicates the size of the input calculation data, Indicates the computing power of user n.

[0056] In another preferred embodiment, the system delay and energy consumption of the micro base station are calculated by the following formula:

[0057]

[0058]

[0059] in, represents the delay that user n spends on micro base station m, a m,n Indicates that user n chooses micro base station m to access. represents the processing probability at the micro base station, d n Indicates the size of the input calculation data, c n Indicates the computing resource requirements required for unit task input size, represents the transmission rate of user n uploading to micro base station m, f m,n represents the processing capacity of user n offloaded to micro base station m, represents the energy consumption of user n on micro base station m, P n represents the transmission power of user n.

[0060] In another preferred embodiment, the system delay and energy consumption of the macro base station are calculated by the following formula:

[0061]

[0062]

[0063] in, Indicates MBS calculation delay, represents MBS computing energy consumption, In MBS treatment probability, d n Indicates the size of the input calculation data, c n P represents the computational resource requirement per unit task input size. n represents the transmission power of user n, r n =(1-κ)r m,n +κr c,n , when user n is covered by SBS, κ=0, otherwise κ=1, Indicates the energy consumption of MBS in each cycle, F c represents the computing capability of MBS, β represents the limited transmission power, and γ is the scaling factor.

[0064] Furthermore, the optimal base station uses a smooth approximation function to convert the non-smooth problem of minimizing system delay and energy consumption into a nonlinear multivariate function minimum problem, which is achieved by the following formula:

[0065]

[0066] Among them, X represents the task offloading decision, F represents the size of the base station allocated resources, Indicates the local computing latency, represents the user terminal under the best base station, Indicates the delay that the user terminal spends on the micro base station, represents the calculation delay of the macro base station, E n represents energy consumption, μ represents the smoothing parameter, and λ represents the weight factor of delay and energy consumption.

[0067] Step S23: The optimal base station solves the nonlinear multivariate function minimum problem using a particle swarm algorithm to obtain an optimal task offloading decision and the optimal base station allocation resource size.

[0068] In one embodiment, if Figure 11 As shown in (a), the particle swarm algorithm is used to solve the nonlinear multivariate function minimization problem of users under the optimal base station deployment. That is, when the user is covered by the micro base station, the user can choose local, micro base station and macro base station at the same time.

[0069] In another embodiment, if Figure 11 As shown in (b), the linear programming problem of users under macro base station deployment is solved by using the optimization toolbox. That is, when the user is only covered by the macro base station, both local and macro base stations can be selected.

[0070] In a specific embodiment, if Figure 10 As shown, the percentage of offloaded tasks is determined according to the increase in the total number of user terminals. When each user terminal offloads the computing task to the mobile edge computing server to save energy consumption and / or completion time, a large number of user terminals will cause fierce competition for computing resources on the mobile edge computing server. Therefore, remote computing will be disadvantageous to the user terminals, so some user terminals need to be executed locally.

[0071] Step S24: the optimal base station returns the optimal task offloading decision and the optimal base station allocated resource size to the corresponding user terminal.

[0072] Step S25: The user terminal executes the task offloading decision returned by the optimal base station.

[0073] like Figure 3As shown, the sub-steps of the user terminal receiving a user task request and selecting the best base station associated with the user terminal according to the task request specifically include:

[0074] Step S31: The user terminal receives a user task request.

[0075] Step S32: The user terminal selects several micro base stations that overlap and cover the user terminal according to the task request.

[0076] Step S33: The user terminal calculates the channel gain between the user terminal and each of the micro base stations covering the user terminal and the requested load of each of the micro base stations covering the user terminal. Figure 8 As shown in the figure, when the user terminal is in the overlapping coverage area, the user terminal will access several micro base stations. In order to reduce the computational complexity, the user terminal is associated and selected, that is, the user terminal is assigned to the nearest micro base station. This is because the shorter the distance, the higher the channel gain. However, when the user terminal is in the overlapping coverage area, the base station with a high user terminal density will be overloaded.

[0077] Step S34: The user terminal calculates the association index between the user terminal and each of the micro base stations covering the user terminal based on the channel gain between the user terminal and each of the micro base stations covering the user terminal and the requested load of each of the micro base stations covering the user terminal.

[0078] In a preferred embodiment, the association index between the user terminal and each of the micro base stations covering the user terminal is calculated by the following formula:

[0079]

[0080] Where: b n,k represents the association index between the user terminal and each micro base station under overlapping coverage, h n,k represents the channel gain between base station k and user n, load k represents the request load of base station k.

[0081] Step S35: The user terminal selects the micro base station with the largest association index as the optimal base station associated with the user terminal.

[0082] like Figure 4 As shown, the optimal base station uses a particle swarm algorithm to solve the nonlinear multivariate function minimum problem to obtain the optimal task offloading decision and the optimal base station allocation resource size. The sub-steps specifically include:

[0083] Step S41: the base station initializes the nonlinear multivariate function minimum problem into a number of particles.

[0084] Step S42: the base station searches for the optimal individual particle and the optimal particle group through the fitness function during the iteration of the particles, and updates the positions and velocities of the particles.

[0085] In a preferred embodiment, the base station searches for the optimal individual particle and the optimal particle swarm through a fitness function during iteration of the plurality of particles, which is achieved by the following formula:

[0086]

[0087] Among them, Fit represents fitness, η represents penalty function, t n represents the delay, λ represents the weight factor of delay and energy consumption, Indicates the calculation delay of the macro base station. represents the delay that user n spends on base station m, E n Indicates energy consumption.

[0088] In a specific embodiment, if Figure 12 As shown, when When the kth decision method M k ={(0.2, 0.5, 7),(0.1, 0.6, 8),(0.15, 0.7, 10)}, where 0.2 and 0.5 in (0.2, 0.5, 7) represent the offloading probabilities of user 1 locally and on SBSm, respectively, and 7 represents the resource size allocated to user 1 by SBSm. Similarly, the velocity of the kth particle above represents V k ={v k1 ,v k2 ,v k3}, where: v k1 is a three-dimensional vector representing the movement speed of user 1, and so on. The updating of its own speed and position is achieved by the following formula:

[0089] V k =W·V k +C1·R1·(E best -M k )+C2·R2·(G best -M k )

[0090] M k =M k +V k

[0091] Where: C1, C2 are acceleration constants, W is the inertia factor, R1, R2 are random numbers in [0,1], E best represents the optimal position searched by the kth particle, G best Indicates the global optimum searched by the entire particle.

[0092] Step S43: the base station obtains the optimal task offloading decision and the optimal base station allocation resource size according to the individual particle optimum and the particle swarm optimum.

[0093] To sum up, the user-association-based computing offloading optimization method provided by the present invention can achieve the problem of reducing system latency and energy consumption and ensuring the optimization of computing offloading, while achieving the purpose of providing task processing efficiency. It can meet the computing offloading optimization needs of most functional scenarios, and can also ensure the purpose of improving task offloading efficiency and task processing efficiency.

[0094] The present invention also provides a user-association-based computation offloading optimization system comprising a plurality of user terminals, a plurality of micro base stations, and a macro base station; the plurality of user terminals, the plurality of micro base stations, and the macro base station are configured such that: the user terminal receives a user task request and, based on the task request, selects an optimal base station associated with the user terminal from at least one of the micro base stations covering the user terminal; the optimal base station converts the non-smooth problem of minimizing system delay and energy consumption into a nonlinear multivariate function minimum problem using a smooth approximation function;

[0095] The optimal base station uses a particle swarm algorithm to solve the nonlinear multivariate function minimum problem to obtain the optimal task offloading decision and the optimal base station allocation resource size; the optimal base station returns the optimal task offloading decision and the optimal base station allocation resource size to the corresponding user terminal; the user terminal executes the task offloading decision returned by the optimal base station. It should be noted that the user-association-based computing offloading optimization system provided in the above embodiment and the user-association-based computing offloading optimization method provided in the above embodiment belong to the same concept, wherein the specific manner in which each system and unit performs the operation has been described in detail in the method embodiment and will not be repeated here. In actual application, the user-association-based computing offloading optimization system provided in the above embodiment can allocate the above functions to different systems as needed, that is, divide the internal structure of the device into different systems to complete all or part of the functions described above, and this is not limited here.

[0096] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions and operations of the systems, methods and computer program products according to various embodiments of the present invention. Each box in the flowchart or block diagram can represent a module, program segment, or part of the code, and the above-mentioned module, program segment, or part of the code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the boxes can also occur in an order different from that marked in the accompanying drawings. For example, two boxes shown in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flowchart, and the combination of boxes in the block diagram or flowchart, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or can be implemented using a combination of dedicated hardware and computer instructions.

[0097] The units involved in the embodiments of the present invention may be implemented in software or hardware, and the units described may also be provided in a processor. In some cases, the names of these units do not limit the units themselves.

[0098] The above embodiments are merely illustrative of the principles and effects of the present invention and are not intended to limit the present invention. Anyone skilled in the art may modify or alter the above embodiments without departing from the spirit and scope of the present invention. Therefore, any equivalent modifications or alterations made by one of ordinary skill in the art without departing from the spirit and technical principles disclosed herein are intended to be covered by the claims of the present invention.

Claims

1. A computation offloading optimization method based on user association, characterized in that: Applicable to a multi-layer heterogeneous network comprising a plurality of user terminals, a plurality of micro base stations and a macro base station; The user association-based calculation offloading optimization method includes: The user terminal receives a user task request and selects an optimal base station associated with the user terminal from at least one micro base station covering the user terminal according to the task request; The optimal base station converts the non-smooth problem of minimizing system delay and energy consumption into a nonlinear multivariate function minimum problem using a smooth approximation function; The optimal base station solves the nonlinear multivariate function minimum problem using a particle swarm algorithm to obtain an optimal task offloading decision and the optimal base station allocation resource size; The best base station returns the best task offloading decision and the best base station allocated resource size to the corresponding user terminal; The user terminal executes the task offloading decision returned by the best base station.

2. The user association-based calculation offloading optimization method according to claim 1, characterized in that: The user terminal receives a user task request and selects an optimal base station associated with the user terminal according to the task request, including: a task set, wherein the task set is represented as: Among them, d n Indicates the size of the input calculation data, c n Indicates the computing resource requirements required for unit task input size, Indicates the maximum tolerable delay required to complete the task.

3. The user association-based calculation offloading optimization method according to claim 1, characterized in that: The step of the user terminal receiving a user task request and selecting an optimal base station associated with the user terminal according to the task request includes: The user terminal receives a user task request; The user terminal selects, according to the task request, a number of micro base stations that overlap and cover the user terminal; The user terminal calculates the channel gain between the user terminal and each of the micro base stations covering the user terminal and the requested load of each of the micro base stations covering the user terminal; The user terminal calculates, based on the channel gain between the user terminal and each of the micro base stations covering the user terminal and the requested load of each of the micro base stations covering the user terminal, an association index between the user terminal and each of the micro base stations covering the user terminal; The user terminal selects the micro base station with the largest association index as the best base station associated with the user terminal.

4. The user association-based calculation offloading optimization method according to claim 3 is characterized in that: The user terminal calculates the association index between the user terminal and each of the micro base stations covering the user terminal according to the channel gain between the user terminal and each of the micro base stations covering the user terminal and the requested load of each of the micro base stations covering the user terminal, which is implemented by the following formula: Where: b n,k represents the association index between the user terminal and each micro base station under overlapping coverage, h n,k represents the channel gain between base station k and user n, load k represents the request load of base station k.

5. The user association-based calculation offloading optimization method according to claim 1, characterized in that: The optimal base station uses a smooth approximation function to convert the non-smooth problem of minimizing system delay and energy consumption into a nonlinear multivariate function minimum problem, which is achieved by the following formula: Among them, X represents the task offloading decision, F represents the size of the base station allocated resources, Indicates the local computing latency, represents the user terminal under the best base station, Indicates the delay that the user terminal spends on the micro base station, represents the calculation delay of the macro base station, E n represents energy consumption, μ represents the smoothing parameter, and λ represents the weight factor of delay and energy consumption.

6. The user association-based calculation offloading optimization method according to claim 1, characterized in that: The optimal base station solves the nonlinear multivariate function minimum problem using a particle swarm algorithm to obtain an optimal task offloading decision and the optimal base station allocation resource size, including: The base station initializes the nonlinear multivariate function minimum problem into a plurality of particles; The base station searches for the optimal individual particle and the optimal particle group through a fitness function during iteration of the particles, and updates the positions and velocities of the particles; The base station obtains the optimal task offloading decision and the optimal base station allocation resource size according to the individual particle optimum and the particle swarm optimum.

7. The user association-based calculation offloading optimization method according to claim 6, characterized in that: The base station searches for the optimal individual particle and the optimal particle swarm through the fitness function during the iteration of the particles, which is achieved by the following formula: Among them, Fit represents fitness, η represents penalty function, t n represents the delay, λ represents the weight factor of delay and energy consumption, Indicates the calculation delay of the macro base station. represents the delay that user n spends on base station m, E n Indicates energy consumption.

8. The user association-based calculation offloading optimization method according to claim 6, characterized in that: The updating of the positions and velocities of the particles is achieved by the following formula: V k =W·V k +C1·R1·(E best -M k )+C2·R2·(G best -M k ) Among them, MK represents the k-th decision method, VK represents the speed of the k-th particle, C1 and C2 are acceleration constants, W is the inertia factor, R1 and R2 are random numbers in [0,1], and G best Indicates the global optimum found by the entire particle search.

9. The user association-based calculation offloading optimization method according to claim 1, characterized in that: Also includes: The macro base station solves the linear programming problem through an optimization toolbox to obtain the optimal task offloading decision.

10. A computing offloading optimization system based on user association, characterized in that: It includes several user terminals, several micro base stations and one macro base station; The plurality of user terminals, the plurality of micro base stations and the macro base station are configured as follows: The user terminal receives a user task request and selects an optimal base station associated with the user terminal from at least one micro base station covering the user terminal according to the task request; The optimal base station converts the non-smooth problem of minimizing system delay and energy consumption into a nonlinear multivariate function minimum problem using a smooth approximation function; The optimal base station solves the nonlinear multivariate function minimum problem using a particle swarm algorithm to obtain an optimal task offloading decision and the optimal base station allocation resource size; The best base station returns the best task offloading decision and the best base station allocated resource size to the corresponding user terminal; The user terminal executes the task offloading decision returned by the best base station.

Images