A method and device for task offloading based on RIS-assisted mobile edge computing system

Abstract

The present application relates to a kind of based on RIS auxiliary mobile edge computing system task unloading method and device, method includes: S1, initialization RIS reflection coefficient matrix, the transmit power of user equipment, local computing frequency and initial objective function value;S2, introduce first auxiliary variable, according to channel information update the transmit power of user equipment, local computing frequency, obtain the target transmit power of user equipment, target local computing frequency;S3, introduce second auxiliary variable, update RIS reflection coefficient matrix, obtain target RIS reflection coefficient matrix;S4, according to first auxiliary variable, target RIS reflection coefficient matrix, the target transmit power of user equipment and target local computing frequency, calculate new objective function;S5, if the difference of objective function value produced by adjacent two iterations is less than preset threshold, it is judged as convergence, and the optimal solution is output.The present application optimizes RIS reflection coefficient, improves computing performance, realizes algorithm fast convergence, and the total bit number of user equipment computing task is significantly improved.

Description

Technical Field

[0001] This application relates to the field of communication technology, and in particular to a method and apparatus for task offloading in a RIS-assisted mobile edge computing system. Background Technology

[0002] With the rapid development of 5G / 6G communication technologies, computing-intensive applications such as autonomous driving, virtual reality, 3D games, and intelligent transportation are emerging. User devices, limited by computing resources and battery capacity, struggle to independently complete highly complex computing tasks. Mobile edge computing (MEC) deploys computing resources at network edge access points, reducing the burden on user devices and lowering task processing latency through task offloading. It is a core technology supporting low-latency and high-reliability applications.

[0003] In real-world wireless environments, non-line-of-sight links often form between user equipment and edge access points due to obstructions from buildings and trees, resulting in high path loss and low transmission rates, severely impacting task offloading efficiency. Reconfigurable Smart Surfaces (RIS), comprising numerous low-cost passive reflective elements, can actively optimize the wireless propagation channel by adjusting the reflection phase, improving link quality and expanding coverage, thus providing an effective solution to address the performance bottlenecks of MEC systems in non-line-of-sight scenarios.

[0004] Existing research on RIS-assisted MEC systems primarily focuses on independently optimizing communication functions and local computing resources, neglecting the coupling relationship between transmit power and local computing frequency under user total power constraints. Furthermore, many studies employ fixed-phase or simple optimization strategies, failing to fully leverage the performance-enhancing effect of RIS on system computation and hindering the achievement of optimal overall system performance. Therefore, designing efficient task offloading and resource joint optimization methods for multi-user, power-constrained, and non-line-of-sight transmission scenarios is of significant research value and application implications. Summary of the Invention

[0005] Therefore, it is necessary to provide a RIS-assisted task offloading method and apparatus for mobile edge computing systems to address the aforementioned technical problems.

[0006] In a first aspect, the present invention provides a task offloading method for a RIS-assisted mobile edge computing system, wherein K user devices offload tasks to an AP via a RIS, and the RIS includes N reflection units. The task offloading method includes the following steps: S1. Obtain channel information from the user equipment to the RIS. k Channel information g0 from RIS to AP, and noise power Initialize the transmit power of the user equipment. Local computing frequency of user equipment and RIS reflection coefficient matrix Calculate the initial objective function value ; S2, Introduce the first auxiliary variable Based on the channel information g from the user equipment to the RIS k RIS to AP channel information g0 updates user equipment transmit power Local computing frequency of user equipment To obtain the target transmit power of the user equipment Target local computing frequency of user equipment ;in, , , , , `diag(.)` is the diagonalization operation. For the first The reflection coefficient of each reflecting unit; S3. Introduce a second auxiliary variable. Update the RIS reflection coefficient matrix The target RIS reflection coefficient matrix is ​​obtained. ; S4. Based on the first auxiliary variable Target transmit power of user equipment Target local computing frequency of user equipment and the target RIS reflection coefficient matrix Calculate the objective function Obtain the maximum total number of bits for the user equipment's computing tasks; and denot the obtained objective function value as... ; S5. If the difference between the objective function values ​​generated by two adjacent iterations If the value is less than a preset threshold, determine if the solution has converged and output the optimal solution; otherwise, [the solution is considered converged]. Then return to step S2.

[0007] Furthermore, in step S2, the first auxiliary variable No. The expression for each element is:

[0008] in, for The The element is the nth element. Transmit power of each user equipment; For noise power, It is the first Equivalent channel from each user equipment to the AP.

[0009] Furthermore, step S2 includes the following steps: S21. Introduce the first auxiliary variable. Based on the channel information g from the user equipment to the RIS k The channel information g0 from RIS to AP is used to calculate and update the transmit power of the user equipment. Local computing frequency of user equipment And based on the updated transmit power and local computing frequency The total operating power of the user equipment is calculated. :

[0010] , =1~K in, , T Total working hours B For system bandwidth; These are coefficients related to the user equipment hardware architecture; The number of CPU cycles required to compute 1 bit of data; Negative Lagrange multipliers;

[0011] in, For the first The circuit power consumed when a user device task is unloaded; S22, Judgment and Does it meet the requirements? and ,in, For the first Maximum local computing frequency per user device: If the conditions are met, then the calculation will be performed. As the first Target transmit power of individual user equipment , will calculate As the first Target local computing frequency for each user device ; otherwise, and Invalidate and execute S23; S23, Re-examine the first The transmit power of each user equipment and the local computing frequency are calculated, and the calculated results are used to calculate the transmit power of each user equipment and the local computing frequency. As the first Target transmit power of individual user equipment , will calculate As the first Target local computing frequency for each user device The formula is as follows:

[0012] and

[0013] in, ; definition , , , ;like Then let , Otherwise, , .

[0014] Furthermore, in step S3, the second auxiliary variable The expression is:

[0015] in, for The Each element.

[0016] Furthermore, step S3 includes the following steps: S30, Calculation , recorded as and define ; S31, Combination Optimize the RIS reflection coefficient matrix : S32, Calculate again , recorded as and define ; S33, Determine the output of step S32 With the present Is the relative norm difference between them less than or equal to the second preset threshold? If it is less than, then convergence occurs, and the optimized RIS reflection coefficient matrix is ​​taken as the target RIS reflection coefficient matrix. ; If not restrained, Return to step S31 to update .

[0017] Furthermore, step S31 includes the following steps: S31-1, Order =1; S31-2, Calculate and update , Represented as:

[0018] in, , The superscript * indicates complex conjugation. Representing the The equivalent channel weighting coefficients of each reflection unit, Representing the The equivalent interference coupling coefficient of each reflecting unit, yes The One element, ; S31-3, := +1, if >N, let =1; S31-4, if If convergence is achieved, output the result; otherwise, return to step S31-2.

[0019] Secondly, the present invention provides a RIS-assisted task offloading device for a mobile edge computing system, used to implement the above method, comprising: The data acquisition and initialization module is used to acquire channel information g from the user equipment to the RIS. k Channel information g0 from RIS to AP, and noise power Initialize the transmit power of the user equipment. Local computing frequency of user equipment and RIS reflection coefficient matrix Calculate the initial objective function value ; The optimization solution module is used to introduce the first auxiliary variable. Based on the channel information g from the user equipment to the RIS k RIS to AP channel information g0 updates user equipment transmit power Local computing frequency of user equipment To obtain the target transmit power of the user equipment Target local computing frequency of user equipment ;in, , , , , `diag(.)` is the diagonalization operation. For the first The reflection coefficient of each reflective unit; introducing a second auxiliary variable. Update the RIS reflection coefficient matrix The target RIS reflection coefficient matrix is ​​obtained. According to the first auxiliary variable Target transmit power of user equipment Target local computing frequency of user equipment and the target RIS reflection coefficient matrix Calculate the objective function The maximum total number of bits for the user equipment's computational tasks is obtained, and the calculated objective function value is denoted as... If the difference between the objective function values ​​generated by two consecutive iterations If the value is less than a preset threshold, determine if the solution has converged and output the optimal solution; otherwise, [the solution is considered converged]. Return to the iteration.

[0020] Thirdly, the present invention provides an electronic device comprising: a processor and a memory, the memory being used to store computer program code, the computer program code including computer instructions, wherein, when the processor executes the computer instructions, the electronic device performs the method described above.

[0021] Fourthly, the present invention provides a computer-readable storage medium storing computer-executable instructions that, when executed by a processor, implement the above-described method.

[0022] Compared with existing technologies, this invention considers the coupling between transmit power and local computing frequency under total power constraints, jointly optimizes communication and computing resources, intelligently optimizes the reflection coefficient of RIS, improves non-line-of-sight links, enhances offloading rate and computing performance, achieves fast algorithm convergence, and significantly increases the total number of bits for user equipment computing tasks. Attached Figure Description

[0023] Figure 1 This is a flowchart illustrating the task offloading method for a RIS-assisted mobile edge computing system as described in this invention. Figure 2 This is a schematic diagram illustrating the iterative process of the total number of system computing task bits in two scenarios (K=3 and K=5) for the RIS-assisted mobile edge computing system task offloading method described in this invention. Figure 3 This diagram illustrates the impact of the number of user devices on the number of system computing task bits under the RIS-assisted mobile edge computing system task offloading method described in this invention and three benchmark schemes. Detailed Implementation

[0024] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0025] The embodiments of the present invention will be further described below with reference to the accompanying drawings.

[0026] One embodiment of the present invention discloses a task offloading method for a RIS-assisted mobile edge computing system, applicable to RIS-assisted mobile edge computing systems. The system includes a single-antenna access point (AP) with a deployed MEC server and K single-antenna user equipments (User Equipments). The User Equipments establish a communication link with the AP via the RIS to offload tasks to the AP. The RIS includes N reflector elements. , Let N represent the sets of N reflection units and K user equipment, respectively. This represents the channel information from RIS to AP. Indicates the first Channel information from each user equipment to the RIS; the reflection coefficient matrix of the RIS is defined as follows. ,in, For the first The reflection coefficient of each reflective unit, and diag(.) is the diagonalization operation.

[0027] Assume the user equipment uses frequency division multiple access (FDMA) and its tasks can be partially offloaded, meaning some tasks can be computed locally and others remotely. For uplink data transmission of the user equipment, the AP receives the first... The signal of each user equipment is: (1) No. The signal-to-noise ratio of each user device is

[0028] in, (2) It is the first Transmit power of each user equipment It's noise. The power.

[0029] No. The computing tasks of each user device are divided into the following two parts.

[0030] (1) Local calculation: the first The number of local computing bits per user device is (3) in (period / s) and (period / bit) represents the number of... The local computing frequency of each user device and the number of CPU cycles required to compute 1 bit of data, where T is the total working time. The local computing power of the user device is... ,in These are coefficients related to the user equipment hardware architecture.

[0031] (2) Remote unloading: The system bandwidth is B, the first The number of task bits that each user equipment offloads to the AP during the entire working time T is (4) No. The power consumed by each user device during remote offloading is

[0032] in, It is the first The circuit power consumed when a user device task is offloaded. The total power of each user equipment should meet the following constraints.

[0033] in, It is the first Total operating power of each user device.

[0034] According to (3) and (4), the first The total number of bits for each user device computing task is

[0035] Proposed joint optimization of user equipment transmit power Local computing frequency of user equipment RIS reflection coefficient matrix To maximize the sum of computing task bits across all user devices (i.e., the total number of computing task bits in the system), an optimization model is constructed, and the expression for the optimization model is:

[0036] For P1, C1 indicates that the total power of each user equipment is limited; C2 indicates that the local computing frequency of each user is greater than or equal to 0, and the transmit power of each user is also greater than or equal to 0; C3 indicates that the local maximum computing frequency of the user equipment is limited; C4 indicates that the coefficient of each reflective unit of RIS is a unit modulus.

[0037] The transmit power of the user equipment .

[0038] The local computing frequency of the user equipment .

[0039] The RIS-assisted task offloading method for mobile edge computing systems, such as Figure 1 ,include: S1. Obtain channel information from the user equipment to the RIS. k Channel information g0 from RIS to AP and noise power Initialize the transmit power of the user equipment. Local computing frequency of user equipment RIS reflection coefficient matrix Calculate the initial objective function value ,in, , , The superscript T represents transpose.

[0040] Because the objective function of P1 Including K logarithmic terms, the nonlinear nature of this optimization problem makes it difficult to solve directly. Therefore, a first auxiliary variable is introduced. The objective function is processed and transformed into... Solve the problem.

[0041] make Then P1 is equivalent to:

[0042] The objective function of P2 is:

[0043] S2, Introduce the first auxiliary variable According to channel information g k And g0 calculates and updates the transmit power of the user equipment. Local computing frequency of user equipment To obtain the target transmit power of the user equipment Target local computing frequency of user equipment ,in, , .

[0044] Specifically, the expression for the first auxiliary variable is:

[0045] in, for The kth element, for The k The element, i.e., the th element Transmit power of each user equipment; For noise power, It is the first Equivalent channel from each user equipment to the AP.

[0046] Step S2 includes: S21. Introduce the first auxiliary variable. According to channel information g k g0 calculates and updates the transmit power of the user equipment. Local computing frequency of user equipment And based on the updated transmit power and local computing frequency The total operating power of the user equipment is calculated. .

[0047] Specifically, the following formula is used for the first... The transmit power of each user equipment and the local computing frequency are used for calculation: , =1~K in, , T Total working hours B For system bandwidth; These are coefficients related to the user equipment hardware architecture; The number of CPU cycles required to compute 1 bit of data; The negative Lagrange multiplier is not true.

[0048] Based on the updated transmit power and local computing frequency The following formula is used to calculate the first... Total operating power of each user equipment :

[0049] in, For the first The circuit power consumed when a user device task is unloaded.

[0050] S22, Judgment and Does it meet the requirements? and ,in, For the first Maximum local computing frequency per user device: If the conditions are met, then the calculation will be performed. As the first Target transmit power of individual user equipment , will calculate As the first Target local computing frequency for each user device ; otherwise, and Invalidate and execute S23.

[0051] S23, Re-examine the first The transmit power of each user equipment and the local computing frequency are calculated, and the calculated results are used to calculate the transmit power of each user equipment and the local computing frequency. As the first Target transmit power of individual user equipment , will calculate As the first Target local computing frequency for each user device The specific calculation formula is as follows:

[0052] and

[0053] in, .

[0054] definition , , , ;like Then let , Otherwise, , .

[0055] S3. Introduce a second auxiliary variable. Update the RIS reflection coefficient matrix .

[0056] Specifically, step S3 includes: S30, Calculation :

[0057] Will Recorded as and define .

[0058] S31, Combination Optimize the RIS reflection coefficient matrix .

[0059] In this embodiment, step S31 includes: S31-1, Order =1; S31-2, Calculate and update , Represented as:

[0060] in, , The superscript * indicates complex conjugation. Representing the The equivalent channel weighting coefficients of each reflection unit, Representing the The equivalent interference coupling coefficient of each reflecting unit, yes The One element, ; S31-3, := +1, if >N, let =1; S31-4, if If convergence is achieved, output the result; otherwise, return to step S31-2.

[0061] S32, combined with the output of step S1 Calculate again , recorded as and define .

[0062] S33, Determine the output of S32 With the present Is the relative norm difference between them less than or equal to the second preset threshold? If it is less than, then convergence occurs, and the optimized RIS reflection coefficient matrix is ​​taken as the target reflection coefficient matrix. ; If not restrained, Return to step S31 to update .

[0063] S4. Based on the first auxiliary variable Target transmit power of user equipment Target local computing frequency of user equipment and the target RIS reflection coefficient matrix Calculate the objective function Defined as:

[0064] Obtain the maximum total number of bits for the user equipment's computational tasks; and denot the obtained objective function value as... ;

[0065] S5. If the difference between the objective function values ​​generated by two adjacent iterations Less than another preset threshold (e.g.) If the objective function value is less than 0.001, determine if it has converged and output the optimal solution; otherwise, assign the calculated current objective function value to the variable storing the previous iteration value. Return to step S2 and calculate the target transmit power based on the power obtained in this iteration. Target RIS reflection coefficient matrix Calculate the first auxiliary variable in the new iteration. .

[0066] At this point, the sum of the number of bits for all user equipment computation tasks reaches its maximum.

[0067] Example: In this embodiment, the RIS includes N=40 reflection units, the system bandwidth B=10 MHz, and the K user equipments are... Local computing power of individual user devices =60 cycles / second, the first Total operating power of each user equipment =1 W, hardware architecture coefficient =5×10⁻⁹ W / (MHz)³, circuit power loss =0.05 W, the first Maximum local computing frequency per user device =500 MHz, noise power =0.01 W. Table 1 below provides the main system parameter settings.

[0068] Table 1

[0069] S1, Obtain the first Channel information g from user equipment to RIS k Channel information g0 from RIS to AP and noise power Initialize the iteration index j=0 and the transmit power of the user equipment. Local computing frequency of user equipment RIS reflection coefficient matrix ,in, `diag(.)` is the diagonalization operation. For the first One reflective unit The reflection coefficient; , The superscript T represents transpose, and K is the total number of users.

[0070] Initialization parameters: First calculate

[0071] Then, let

[0072]

[0073] Each element in can be written as ,but Can be randomly selected from Range of values, complete Initialization.

[0074] Finally, calculate the initial objective function value. .

[0075] S2, Introduce the first auxiliary variable Calculate and update the transmit power of user equipment Local computing frequency of user equipment .

[0076] The expression for the first auxiliary variable is:

[0077] in, for The kth element, for The k The element, i.e., the th element Transmit power of each user equipment For the first The equivalent channel from a user equipment to an AP is expressed by formula (2).

[0078] Calculate and update and :

[0079] , k=1~K in, K represents the total number of users; These are coefficients related to the user equipment hardware architecture. These are non-negative Lagrange multipliers, obtained from the following equation:

[0080] get Substitute and The expression is used to find the optimal resource allocation strategy, namely the transmit power and local computing frequency of each user equipment.

[0081] judge and Does it meet the requirements? and If the conditions are met, then and That is what is required; otherwise, and This is invalid. Calculate the following two solutions.

[0082] First solution: Calculate first

[0083] according to

[0084] but

[0085]

[0086] Second solution: , Compare the two solutions and select the one with the higher number of computational bits as the optimal solution. Definition , , , .like Then let , Otherwise, , .

[0087] and the above The update result is denoted as ,Will The update result is denoted as .

[0088] S3. Introduce a second auxiliary variable. Update the RIS reflection coefficient matrix Record the update result as .

[0089] S4, according to , , , Calculate the objective function And the calculated objective function value is denoted as The objective function is defined as:

[0090] Among them, the equivalent channel Calculate according to the above formula (2).

[0091] S5. If the difference between the values ​​of the objective function generated by two adjacent iterations If the value is less than a preset threshold (e.g., 0.001), convergence is determined, and the optimal solution to the problem is output; otherwise, Return to S2.

[0092] It can be seen that when the total number of user devices is set, the results of each step of the algorithm can be calculated accurately.

[0093] After measurement, the channel information and noise power of the user-RIS and RIS-AP are obtained. Then, the optimal transmit power of the user equipment is calculated according to the method provided by this invention. Local computing frequency of user equipment and RIS reflection coefficient matrix The optimized data is fed back to the corresponding user devices and RIS controller for parameter configuration. All user devices start executing tasks synchronously, with a working time of T (parameters known). Each user device simultaneously performs local computation and remote offloading of its respective task.

[0094] This invention solves the technical problems of low task offloading efficiency, difficulty in optimizing the coupling between user power and computing resources, and complexity in solving non-convex problems under non-line-of-sight channels, and achieves fast algorithm convergence and a significant increase in the total number of computing bits of the system.

[0095] The method provided by this invention considers the coupling between transmit power and local frequency under total power constraints, jointly optimizes communication and computing resources, and improves efficiency; it uses the block coordinate descent method to decompose non-convex problems, derives analytical solutions for power and frequency, and iteratively optimizes the RIS phase, resulting in low algorithm complexity and fast convergence speed; it intelligently optimizes the RIS reflection coefficient, improves non-line-of-sight links, and enhances offloading rate and computing performance; it is applicable to multi-user, channel-obstructed, and total power-constrained edge computing, has strong versatility, and can be deployed in 5G / 6G RIS-MEC systems.

[0096] This invention uses the following three benchmark schemes for performance comparison: 1) Local computation: Only local devices are used for computation, and tasks are not unloaded to remote servers.

[0097] 2) Remote Unloading: Only remote unloading is used for computation tasks, and local computation is not used.

[0098] 3) All reflection coefficients are 1: The partial offloading mode is adopted, but the algorithm only optimizes the local calculation frequency and transmission power, and the reflection coefficients are all set to 1.

[0099] like Figure 2 This paper demonstrates the iterative process of calculating the total number of bits in the system task using the method provided by this invention in two scenarios (K=3 and K=5). It can be seen that the proposed algorithm converges after a small number of iterations.

[0100] like Figure 3 The paper demonstrates the impact of the number of user devices on the number of system computing task bits. It can be seen that the method provided by this invention consistently achieves the highest total number of system computing task bits, and this number steadily increases with the number of users. This indicates that the method provided by this invention can effectively utilize multi-user diversity gain and RIS beamforming capabilities, significantly improving the overall system computing performance. Compared to three benchmark schemes—local computing, remote computing, etc.—the method provided by this invention maintains optimal performance under different user number scenarios, verifying its effectiveness and robustness in RIS-assisted edge computing systems. Compared to schemes such as pure local computing, fully remote offloading, and RIS reflection coefficient fixed at 1, under the same system parameter conditions, the method provided by this invention has a faster convergence speed and a significantly higher total number of system computing task bits.

[0101] In another embodiment, the present invention also provides a RIS-assisted task offloading device for a mobile edge computing system, used to implement the above-described task offloading method. The device includes a data acquisition and initialization module and an optimization and solution module. The data acquisition and initialization module is used to acquire channel information from the user equipment to the RIS. k Channel information g0 from RIS to AP, and noise power Initialize the transmit power of the user equipment. Local computing frequency of user equipment and RIS reflection coefficient matrix Calculate the initial objective function value ,in, , , `diag(.)` is the diagonalization operation. For the first The reflection coefficient of each reflecting unit; The optimization solution module is used to introduce the first auxiliary variable. Based on the channel information g from the user equipment to the RIS k RIS to AP channel information g0 updates user equipment transmit power Local computing frequency of user equipment To obtain the target transmit power of the user equipment Target local computing frequency of user equipment ;in, , Introducing a second auxiliary variable Update the RIS reflection coefficient matrix The target RIS reflection coefficient matrix is ​​obtained. According to the first auxiliary variable Target transmit power of user equipment Target local computing frequency of user equipment and the target RIS reflection coefficient matrix Calculate the objective function The maximum total number of bits for the user equipment's computational tasks is obtained, and the calculated objective function value is denoted as... If the difference between the objective function values ​​generated by two consecutive iterations If the value is less than a preset threshold, determine if the solution has converged and output the optimal solution; otherwise, [the solution is considered converged]. .

[0102] In another embodiment, the present invention also provides an electronic device, including a processor and a memory, the memory being used to store computer program code, the computer program code including computer instructions, wherein the electronic device performs the task offloading method described above when the processor executes the computer instructions.

[0103] In another embodiment, the present invention also provides a computer-readable storage medium storing at least one instruction, which is loaded and executed by a processor to implement the steps of the multi-level computing task offloading method described above. For example, the computer-readable storage medium may be a ROM, random access memory (RAM), or CD. If the functions of ROM, magnetic tape, floppy disk, and optical data storage devices are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention.

[0104] The aforementioned storage media include: USB flash drives, external hard drives, and read-only memory (ROM). Various media capable of storing program code, such as ROM (ROM-only memory), random access memory (RAM), magnetic disks, or optical disks, can be used. It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Furthermore, at least some steps in the flowcharts of the embodiments described above may include multiple steps or stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least a portion of the steps or stages of other steps.

[0105] The terminology used in the embodiments of this application is for the purpose of describing particular embodiments only and is not intended to limit the embodiments of this application. The singular forms “a,” “the,” and “the” used in the embodiments and claims of this application are also intended to include the plural forms, unless the context clearly indicates otherwise. It should also be understood that, unless otherwise stated, “a plurality” and “several” refer to two or more; “and / or” refers to and includes any or all possible combinations of one or more associated listed items; “first,” “second,” “third,” etc., are used only to distinguish and not to describe a particular order or sequence, nor should they be construed as indicating or implying relative importance. When the above description relates to drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. In the description of this application, those skilled in the art can understand the specific meaning of the above terms in this application according to the specific circumstances.

[0106] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0107] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A task offloading method for a RIS-assisted mobile edge computing system, wherein K user devices offload tasks to an AP via a RIS, and the RIS comprises N reflection units, characterized in that... The task unloading method includes the following steps: S1. Obtain channel information from the user equipment to the RIS. k Channel information g0 from RIS to AP, and noise power Initialize the transmit power of the user equipment. Local computing frequency of user equipment and RIS reflection coefficient matrix Calculate the initial objective function value ,in, , , `diag(.)` is the diagonalization operation. For the first The reflection coefficient of each reflecting unit; S2. Introduce the first auxiliary variable. Based on the channel information g from the user equipment to the RIS k RIS to AP channel information g0 updates user equipment transmit power Local computing frequency of user equipment To obtain the target transmit power of the user equipment Target local computing frequency of user equipment ,in, , ; S3. Introduce a second auxiliary variable. Update the RIS reflection coefficient matrix The target RIS reflection coefficient matrix is ​​obtained. ; S4. Based on the first auxiliary variable Target transmit power of user equipment Target local computing frequency of user equipment and the target RIS reflection coefficient matrix Calculate the objective function Obtain the maximum total number of bits for the user equipment's computing tasks; and denot the obtained objective function value as... ; S5. If the difference between the objective function values ​​generated by two adjacent iterations If the value is less than a preset threshold, determine if the solution has converged and output the optimal solution; otherwise, [the solution is considered converged]. Then return to step S2.

2. The task unloading method according to claim 1, characterized in that, The first auxiliary variable in step S2 No. The expression for each element is: in, for The The element is the nth element. Transmit power of each user equipment; For noise power, It is the first The equivalent channel from each user equipment to the AP.

3. The task unloading method according to claim 2, characterized in that, Step S2 includes the following steps: S21. Introduce the first auxiliary variable. Based on the channel information g from the user equipment to the RIS k The channel information g0 from RIS to AP is used to calculate and update the transmit power of the user equipment. Local computing frequency of user equipment And based on the updated transmit power and local computing frequency The total operating power of the user equipment is calculated. : ， =1~K in, , T Total working hours B For system bandwidth; These are coefficients related to the user equipment hardware architecture; The number of CPU cycles required to compute 1 bit of data; Negative Lagrange multipliers; in, For the first The circuit power consumed when a user device task is unloaded; S22, Judgment and Does it meet the requirements? and ,in, For the first Maximum local computing frequency per user device: If the conditions are met, then the calculation will be performed. As the first Target transmit power of individual user equipment , will calculate As the first Target local computing frequency for each user device ; otherwise, and Invalidate and execute S23; S23, Re-examine the first The transmit power of each user equipment and the local computing frequency are calculated, and the calculated results are used to calculate the transmit power of each user equipment and the local computing frequency. As the first Target transmit power of individual user equipment , will calculate As the first Target local computing frequency for each user device The formula is as follows: and in, ; definition , , , ;like Then let , Otherwise, , .

4. The task unloading method according to claim 3, characterized in that, In step S3, the second auxiliary variable The expression is: in, for The Each element.

5. The task unloading method according to claim 4, characterized in that, Step S3 includes the following steps: S30, Calculation , recorded as and define ; S31, Combination Optimize the RIS reflection coefficient matrix : S32, Calculate again , recorded as and define ; S33, Determine the output of step S32 With the present Is the relative norm difference between them less than or equal to the second preset threshold? If it is less than, then convergence occurs, and the optimized RIS reflection coefficient matrix is ​​taken as the target RIS reflection coefficient matrix. ; If not restrained, Return to step S31 to update .

6. The task unloading method according to claim 5, characterized in that, Step S31 includes the following steps: S31-1, Order =1; S31-2, Calculate and update , Represented as: in, , The superscript * indicates complex conjugation. Representing the The equivalent channel weighting coefficients of each reflection unit, Representing the The equivalent interference coupling coefficient of each reflecting unit, yes The One element, ; S31-3, := +1, if >N, let =1; S31-4, if If convergence is achieved, output the result; otherwise, return to step S31-2.

7. A RIS-assisted task offloading device for a mobile edge computing system, used to implement the task offloading method according to any one of claims 1-6, characterized in that, include: The data acquisition and initialization module is used to acquire channel information g from the user equipment to the RIS. k Channel information g0 from RIS to AP, and noise power Initialize the transmit power of the user equipment. Local computing frequency of user equipment and RIS reflection coefficient matrix Calculate the initial objective function value ,in, , , `diag(.)` is the diagonalization operation. For the first The reflection coefficient of each reflecting unit; The optimization solution module is used to introduce the first auxiliary variable. Based on the channel information g from the user equipment to the RIS k RIS to AP channel information g0 updates user equipment transmit power Local computing frequency of user equipment To obtain the target transmit power of the user equipment Target local computing frequency of user equipment ;in, , Introducing a second auxiliary variable Update the RIS reflection coefficient matrix The target RIS reflection coefficient matrix is ​​obtained. According to the first auxiliary variable Target transmit power of user equipment Target local computing frequency of user equipment and the target RIS reflection coefficient matrix Calculate the objective function The maximum total number of bits for the user equipment's computational tasks is obtained, and the calculated objective function value is denoted as... If the difference between the objective function values ​​generated by two consecutive iterations If the value is less than a preset threshold, determine if the solution has converged and output the optimal solution; otherwise, [the solution is considered converged]. .

8. An electronic device, characterized in that, include: A processor and a memory, the memory being used to store computer program code, the computer program code including computer instructions, wherein, when the processor executes the computer instructions, the electronic device performs the method as described in any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions that, when executed by a processor, implement the method as described in any one of claims 1-6.

Images